6-Map_Assemblage.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d3f10f61",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copyright 2023 Google LLC\n",
    "#\n",
    "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
    "# you may not use this file except in compliance with the License.\n",
    "# You may obtain a copy of the License at\n",
    "#\n",
    "#     https://www.apache.org/licenses/LICENSE-2.0\n",
    "#\n",
    "# Unless required by applicable law or agreed to in writing, software\n",
    "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
    "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
    "# See the License for the specific language governing permissions and\n",
    "# limitations under the License.\n",
    "#\n",
    "# Developed for Google LLC by RedCastle Resources Inc\n",
    "# https://www.redcastleresources.com/"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5d35d3b",
   "metadata": {
    "id": "title:generic,gcp"
   },
   "source": [
    "# <img width=50px  src = 'https://apps.fs.usda.gov/lcms-viewer/src/assets/images/lcms-icon.png'> Lab 6: LCMS Map Assemblage\n",
    "\n",
    "<table align=\"left\">\n",
    "  <td>\n",
    "    <a href=\"https://colab.research.google.com/github/redcastle-resources/lcms-training/blob/main/6-Map_Assemblage.ipynb\">\n",
    "      <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
    "    </a>\n",
    "  </td>\n",
    "  <td>\n",
    "    <a href=\"https://github.com/redcastle-resources/lcms-training/blob/main/6-Map_Assemblage.ipynb\">\n",
    "      <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
    "      View on GitHub\n",
    "    </a>\n",
    "  </td>\n",
    "  <td>\n",
    "    <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://github.com/redcastle-resources/lcms-training/blob/main/6-Map_Assemblage.ipynb\">\n",
    "      <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
    "      Open in Vertex AI Workbench\n",
    "    </a>\n",
    "  </td>\n",
    "</table>\n",
    "<br/><br/><br/>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a5965a7",
   "metadata": {
    "id": "overview:mlops"
   },
   "source": [
    "\n",
    "## 6.0: Overview and Introduction\n",
    "\n",
    "\n",
    "This notebook teaches how to take raw model outputs and assemble final map classes. You will use two different map assemblage techniques, highest probability assemblage and threshold-based assemblage, to assemble the different map products. \n",
    "\n",
    "\n",
    "### 6.0.1: Objective\n",
    "\n",
    "In this tutorial, you learn how to manipulate raw model output GEE image arrays to create map outputs of a single product.\n",
    "\n",
    "This tutorial uses the following Google Cloud services:\n",
    "\n",
    "- `Google Earth Engine`\n",
    "\n",
    "The steps performed include:\n",
    "\n",
    "- Looking at raw GEE image array model output assets\n",
    "- Manipulating the image arrays for a basic map assemblage\n",
    "- Performing a more complicated map assemblage that balances omission and commission for Change"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "151709b3",
   "metadata": {},
   "source": [
    "### 6.0.2: Before you begin\n",
    "\n",
    "#### If you are working in Workbench: Set your current URL under `workbench_url`\n",
    "This gives the Map Viewer a url in which to host the viewer we will be generating. \n",
    "* This will be in your URL/search bar at the top of the browser window you are currently in\n",
    "* It will look something like `https://1234567890122-dot-us-west3.notebooks.googleusercontent.com/` (See the image below)\n",
    "\n",
    "![workspace url](img/workspace-url.png)\n",
    "\n",
    "#### Set a folder to use for all exports under `export_path_root` \n",
    "* This folder should be an assets folder in an existing GEE project.\n",
    "* By default, this folder is the same as the pre-baked folder (where outputs have already been created). \n",
    "* If you would like to create your own outputs, specify a different path for `export_path_root`, but leave the `pre_baked_path_root` as it was. This way, the pre-baked outputs can be shown at the end, instead of waiting for all exports to finish.\n",
    "* It will be something like `projects/projectID/assets/newFolder`\n",
    "* This folder does not have to already exist. If it does not exist, it will be created\n",
    "\n",
    "**If you are working in Qwiklabs and wish to export:** Copy the project ID from the 'Start Lab' screen into the `projectID` field in `export_path_root`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b28063e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "workbench_url = 'https://1307eb830a12e633-dot-us-central1.notebooks.googleusercontent.com/lab/tree/lcms-training/6-Map_Assemblage.ipynb'\n",
    "pre_baked_path_root  = 'projects/rcr-gee/assets/lcms-training'\n",
    "export_path_root = pre_baked_path_root\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58bf8756",
   "metadata": {},
   "source": [
    "#### Installation\n",
    "First, install necessary Python packages. Uncomment the first line to upgrade geeViz if necessary.\n",
    "\n",
    "Note that for this module, we're also importing many data science packages such as pandas. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b967dd6f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initializing GEE\n",
      "Not signed up for Earth Engine or project is not registered. Visit https://developers.google.com/earth-engine/guides/access\n",
      "Cached project id file path: C:\\Users\\ihousman\\.config\\earthengine\\credentials.proj_id\n",
      "Cached project id: rcr-gee\n",
      "Successfully initialized\n",
      "geeViz package folder: c:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\geeViz\n",
      "SQLalchemy is not installed. No support for SQL output.\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "#Module imports\n",
    "#!python -m pip install geeViz --upgrade\n",
    "try:\n",
    "    import geeViz.getImagesLib as getImagesLib\n",
    "except:\n",
    "    !python -m pip install geeViz\n",
    "    import geeViz.getImagesLib as getImagesLib\n",
    "\n",
    "import geeViz.changeDetectionLib as changeDetectionLib\n",
    "import geeViz.assetManagerLib as aml\n",
    "import geeViz.taskManagerLib as tml\n",
    "import geeViz.gee2Pandas as g2p\n",
    "import inspect,operator\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os,json\n",
    "\n",
    "try:\n",
    "    from sklearn.model_selection import train_test_split\n",
    "    from sklearn.model_selection import GroupShuffleSplit\n",
    "    from sklearn.model_selection import GroupKFold\n",
    "    from sklearn.ensemble import RandomForestClassifier\n",
    "    from sklearn import metrics \n",
    "except:\n",
    "    !pip install -U scikit-learn\n",
    "    from sklearn.model_selection import train_test_split\n",
    "    from sklearn.model_selection import GroupShuffleSplit\n",
    "    from sklearn.model_selection import GroupKFold\n",
    "    from sklearn.ensemble import RandomForestClassifier\n",
    "    from sklearn import metrics \n",
    "\n",
    "try:\n",
    "    import matplotlib.pyplot as plt\n",
    "except:\n",
    "    !pip install matplotlib\n",
    "    import matplotlib.pyplot as plt\n",
    "    \n",
    "\n",
    "    \n",
    "ee = getImagesLib.ee\n",
    "Map = getImagesLib.Map\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01ae56cc",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Set up your work environment\n",
    "\n",
    "Create a folder in your export path where you will export the composites. In addition, create a blank image collection where your composites will live.\n",
    "\n",
    "Currently, when running within Colab or Workbench, geeView uses a different project to authenticate through, so you may need to make your asset public to view from within Colab.\n",
    "\n",
    "**Warning!!**\n",
    "\n",
    "* **Read-only access is provided for all authenticated GEE users for pre-baked outputs**\n",
    "* **If you are using the pre-baked output location (`export_path_root = pre_baked_path_root`), you will see errors for any operation that tries to write, delete, or change access permissions to any asset in the pre-baked output location.**\n",
    "* **This is expected and will not stop you from successfully running this notebook - ignore these error messages should they appear.**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c9914314",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found the following sub directories:  ['lcms-training', 'lcms-training_module-6_assembledLCMSOutputs']\n",
      "Will attempt to create them if they do not exist\n",
      "Asset projects/rcr-gee/assets/lcms-training already exists\n",
      "Asset projects/rcr-gee/assets/lcms-training/lcms-training_module-6_assembledLCMSOutputs already exists\n",
      "Updating permissions for:  projects/rcr-gee/assets/lcms-training/lcms-training_module-6_assembledLCMSOutputs\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "# Bring in all folders/collections that are needed\n",
    "# These must already exist as they are created in previous notebooks\n",
    "export_rawLCMSOutputs_collection = f'{export_path_root}/lcms-training_module-5_rawLCMSOutputs'\n",
    "export_assembledLCMSOutputs_collection = f'{export_path_root}/lcms-training_module-6_assembledLCMSOutputs'\n",
    "export_timeSync_folder = f'{export_path_root}/lcms-training_module-4_timeSync'\n",
    "\n",
    "aml.create_asset(export_assembledLCMSOutputs_collection, asset_type = ee.data.ASSET_TYPE_IMAGE_COLL)\n",
    "\n",
    "# This is the pre-made TimeSync data\n",
    "# Creating this dataset is not covered in this set of notebooks\n",
    "timeSync_featureCollection = 'projects/rcr-gee/assets/training-materials/18_PRVI_AllPlots_TimeSync_Annualized_Table_secLC_wStrata2'\n",
    "\n",
    "# Currently geeView within Colab uses a different project to authenticate through, so you may need to make your asset public to view from within Colab\n",
    "aml.updateACL(export_assembledLCMSOutputs_collection,writers = [],all_users_can_read = True,readers = [])\n",
    "\n",
    "# Set folder for local model data\n",
    "local_model_data_folder = '/tmp/lcms-training/local_modeling'\n",
    "model_options_csv_filename = './tables/LCMS_model_options_table.csv'\n",
    "local_training_csv = os.path.join(local_model_data_folder,'timeSync_training_table.csv')\n",
    "\n",
    "\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "20eeca23",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "# set up map\n",
    "Map.clearMap()\n",
    "\n",
    "# reset port if necessary\n",
    "Map.port = 1236\n",
    "Map.proxy_url = workbench_url\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2acad9d2",
   "metadata": {},
   "source": [
    "## 6.1: Load raw LCMS outputs \n",
    "* You'll see that the raw LCMS outputs are all probabilities\n",
    "\n",
    "### 6.1.1: Load raw probabilities "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2dac4aef",
   "metadata": {},
   "outputs": [],
   "source": [
    "Map.clearMap()\n",
    "# Bring in raw LCMS model outputs\n",
    "raw_lcms = ee.ImageCollection(export_rawLCMSOutputs_collection)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "72333f9c-35c2-403a-be5e-bb8527da1800",
   "metadata": {},
   "outputs": [],
   "source": [
    "Map.clearMap()\n",
    "# Bring in raw LCMS model outputs\n",
    "raw_lcms = ee.ImageCollection(export_rawLCMSOutputs_collection)\n",
    "\n",
    "# Bring in existing LCMS data for the class names, numbers, and colors\n",
    "lcms_viz_dict = ee.ImageCollection(\"USFS/GTAC/LCMS/v2022-8\").filter('study_area==\"PRUSVI\"').first().toDictionary().getInfo()\n",
    "# print(lcms_viz_dict)\n",
    "    \n",
    "# Get some info\n",
    "products = raw_lcms.aggregate_histogram('product').keys().getInfo()\n",
    "\n",
    "\n",
    "# Specify missing classes in PRUSVI\n",
    "# Talls Shrubs (2 and 6)  and snow/ice (13) are not present and need filled back in\n",
    "# Change and Land_Use have all values\n",
    "missing_names = {'Change':[],\n",
    "                  'Land_Cover':['Tall Shrubs & Trees Mix (SEAK Only)',\n",
    "                                'Tall Shrubs (SEAK Only)',\n",
    "                                'Snow or Ice'],\n",
    "                  'Land_Use':[]\n",
    "                 }"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6578718d-bf1d-460c-985b-f4f8b29cbbdf",
   "metadata": {},
   "source": [
    "### 6.1.2: Set up array processing functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ad540544-ec07-4537-be18-589c91090261",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "def arrayToImage(img,bandNames):\n",
    "    return img.arrayProject([0])\\\n",
    "            .arrayFlatten([bandNames])\\\n",
    "            .copyProperties(img,['system:time_start'])\n",
    "\n",
    "# Function to add in a fill value in a 1-d image array of a given index and value\n",
    "def fillArray(img,index,fillValue=0):\n",
    "    imgLeft = img.arraySlice(0,0,index)\n",
    "    imgRight = img.arraySlice(0,index,None)\n",
    "    imgLeft = imgLeft.arrayCat(ee.Image(fillValue).toArray(),0)\n",
    "    out = imgLeft.arrayCat(imgRight,0)\n",
    "    return ee.Image(out.copyProperties(img).copyProperties(img,['system:time_start']))\n",
    "\n",
    "# Function to handle filling multiple values\n",
    "def fillArrayMulti(img,indices,fillValue):\n",
    "    for index in indices:\n",
    "        img = fillArray(img,index,fillValue)\n",
    "    return img\n",
    "\n",
    "print(\"Done\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b675bae5",
   "metadata": {},
   "source": [
    "### 6.1.2: Highest probability assemblage\n",
    "* Get the class code of the class with the highest probability - the most confident class\n",
    "* You'll see that the raw LCMS outputs are all probabilities\n",
    "* This is the most basic assemblage method and most commonly used\n",
    "* We will start with this approach and then illustrate when this simple approach may not create the best map in some instances\n",
    "* This map will take some time to load since it is loading 38 * 6 individual layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "53b949e9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adding layer: Most Probable LCMS Change\n",
      "Adding layer: Raw LCMS Change\n",
      "Adding layer: Most Probable LCMS Land Cover\n",
      "Adding layer: Raw LCMS Land Cover\n",
      "Adding layer: Most Probable LCMS Land Use\n",
      "Adding layer: Raw LCMS Land Use\n",
      "Starting webmap\n",
      "Using default refresh token for geeView: C:\\Users\\ihousman/.config/earthengine/credentials\n",
      "Starting local web server at: http://localhost:1236/geeView/\n",
      "HTTP server command: \"c:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\python.exe\" -m http.server  1236\n",
      "Done\n",
      "cwd z:\\Projects\\06_LCMS_4_NFS\\Scripts\\lcms-training\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"525px\"\n",
       "            src=\"http://localhost:1236/geeView/?accessToken=ya29.a0AfB_byDGSyTjiOIEPkDjP8IdBeS2NHRgjVyaIHqDvWeVmgJ0-3yjbDz_DaePDRFgdRTcIN8DF93OwKBWnlnR2rzmxEeApgPruMGhRcrHkJ4AhKn5a7BvJ4nyub6QZS8tkMLa9Frt8VfCfysQ7M7bSIGNPQU80JLjWiungjL0gp7taCgYKAb8SARASFQHGX2MibPGaqmmpQ7RlBeNXJEe-Dg0179\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x281e88bb2d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "Map.clearMap()\n",
    "\n",
    "max_prob_lcms_classes = {}\n",
    "\n",
    "# Provide some useful bands to visualize the probability bands with\n",
    "prob_viz_bands = {'Change':'Fast Loss,Gain,Stable',\n",
    "                  'Land_Cover':'Grass/Forb/Herb,Trees,Water',\n",
    "                  'Land_Use':'Developed,Forest,Non-Forest Wetland'}\n",
    "\n",
    "# Get the class code of the most confident class\n",
    "# This is the most basic assemblage method and most commonly used\n",
    "# We will start with this approach and then illustrate when this simple approach may not \n",
    "# create the best map in some instances\n",
    "def getMostProbableClass(raw_lcms_product_yr,product):\n",
    "    # Pull the index of the most probable value\n",
    "    # Since 0 is not used for LCMS outputs, add 1\n",
    "    max_prob_class = raw_lcms_product_yr.arrayArgmax().arrayGet(0).add(1).byte().rename([product])\n",
    "    max_prob_class = ee.Image(max_prob_class)\n",
    "\n",
    "    null_code = lcms_viz_dict[f'{product}_class_values'][-1]\n",
    "    max_prob_class = max_prob_class.unmask(null_code)\n",
    "    return max_prob_class.copyProperties(raw_lcms_product_yr,['system:time_start'])\n",
    "\n",
    "# Iterate across each product and assemble the most probable class collection from the raw\n",
    "for product in products:\n",
    "    # Pull the class names\n",
    "    class_names = lcms_viz_dict[f'{product}_class_names'][:-1]\n",
    "   \n",
    "    product_title = product.replace('_',' ')\n",
    "    \n",
    "    # Filter to only include the product of interest\n",
    "    raw_lcms_product = raw_lcms.filter(ee.Filter.eq('product',product))\n",
    "    \n",
    "    # Fill in any missing values\n",
    "    missing_names_product = missing_names[product]\n",
    "    missing_indices_product = [lcms_viz_dict[f'{product}_class_names'].index(missing_name_product) for missing_name_product in missing_names_product]\n",
    "    \n",
    "    if len(missing_names_product) > 0:\n",
    "        raw_lcms_product = raw_lcms_product.map(lambda img: fillArrayMulti(img,missing_indices_product,0))\n",
    "    \n",
    "    # Find the max probability class\n",
    "    maxProb_lcms_product = raw_lcms_product.map(lambda raw_lcms_product_yr:getMostProbableClass(raw_lcms_product_yr,product).set(lcms_viz_dict))\n",
    "    max_prob_lcms_classes[product] = maxProb_lcms_product\n",
    "    \n",
    "    # Convert raw model probabilities to bands for exploration\n",
    "    raw_lcms_product = raw_lcms_product.map(lambda img:arrayToImage(img,class_names))\n",
    "    \n",
    "    # Add the assembed final class maps\n",
    "    Map.addTimeLapse(maxProb_lcms_product,{'autoViz':True},f'Most Probable LCMS {product_title}')\n",
    "    \n",
    "     # Add the raw model probabilities to query\n",
    "    Map.addTimeLapse(raw_lcms_product,{'addToLegend':False,'bands':prob_viz_bands[product]},f'Raw LCMS {product_title}')\n",
    "Map.turnOnInspector()\n",
    "Map.setQueryDateFormat('YYYY')\n",
    "Map.view()\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01eac71f",
   "metadata": {},
   "source": [
    "#### Inspect\n",
    "* Click on the radio buttons next to the layers to turn them on\n",
    "\n",
    "* Turn on \"Raw LCMS Land Cover\" \n",
    "\n",
    "* While the highest probability assemblage method works well, you can introduce various rules to tailor the final map product to different audiences\n",
    "* In the case of LCMS, there is a focus on forest applications\n",
    "* We know that our models are generally not very confident about classifying change largely due to the limited number of training samples that experience change\n",
    "* We also know our land use classes are not mutually exclusive (e.g. non-forest wetlands can potentially occur in any land use)\n",
    "* Therefore we have a lot of confusion in our land use outputs\n",
    "\n",
    "\n",
    "* Inspect time series of change probability\n",
    "\n",
    "* Take note that the Stable is the most probable type of change\n",
    "* We will need a different assemblage method for change"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e464a26",
   "metadata": {},
   "source": [
    "## 6.2 Compute thresholds for change classes and assemble\n",
    "\n",
    "* Change is generally a rare class\n",
    "* As we noted earlier, uur training data have far more stable plots than change plots\n",
    "* As a result, most of the time, our models are not very confident of change\n",
    "* We do not use the class the model was most confident of for change classification\n",
    "* Instead, we take any of the 3 change classes that are above a threshold that balances omission and commission error and then take the most confident classs (should more than one be above their respective thresholds)\n",
    "* The next blocks compute the thresholds that balance omission and commission\n",
    "* We do this as a part of our model fitting and model assemblage\n",
    "\n",
    "### 6.2.1: Set up Random Forest Model for thresholding\n",
    "\n",
    "#### Load training data\n",
    "* we have to repeat some steps from the last module, module 5\n",
    "* to load in and prepare the training data again "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "24903f45",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Filtered number of training samples: 19777\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "##### This is all from module 5\n",
    "# Necessary in order to download data for local use\n",
    "\n",
    "# Bring in raw TS data\n",
    "timeSyncData = ee.FeatureCollection(timeSync_featureCollection)\n",
    "timeSync_fields = timeSyncData.first().toDictionary().keys().getInfo()\n",
    "\n",
    "# Now lets bring in all training data and prep it for modeling\n",
    "assets = ee.data.listAssets({'parent': export_timeSync_folder})['assets']\n",
    "\n",
    "# You may need to change the permissions for viewing model outputs in geeViz\n",
    "# Uncomment this if needed\n",
    "#for asset in assets:aml.updateACL(asset['name'],writers = [],all_users_can_read = True,readers = [])\n",
    "\n",
    "# Read in each year of extracted TimeSync data\n",
    "training_data = ee.FeatureCollection([ee.FeatureCollection(asset['name']) for asset in assets]).flatten()\n",
    "\n",
    "# Bring in existing LCMS data for the class names, numbers, and colors\n",
    "lcms_viz_dict = ee.ImageCollection(\"USFS/GTAC/LCMS/v2022-8\").filter('study_area==\"PRUSVI\"').first().toDictionary().getInfo()\n",
    "                              \n",
    "# Get the field names for prediction\n",
    "# Find any field that was not in the original TimeSync data and assume that is a predictor variable\n",
    "all_fields = training_data.first().toDictionary().keys().getInfo()\n",
    "predictor_field_names = [field for field in all_fields if field not in timeSync_fields]\n",
    "\n",
    "# Take out the date predictor and the cross-walked codes\n",
    "predictor_field_names = [nm for nm in predictor_field_names if nm != 'year' and nm.find('_Code') == -1]\n",
    "\n",
    "# Filter out any non null values (any training plot with missing predictor data will cause the model to fail entirely)\n",
    "training_data = training_data.filter(ee.Filter.notNull(predictor_field_names))\n",
    "print('Filtered number of training samples:',training_data.size().getInfo())\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "af6a1b2b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/tmp/lcms-training/local_modeling\\timeSync_training_table.csv  already exists\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>BARREN</th>\n",
       "      <th>BARREN-GRA</th>\n",
       "      <th>BARREN-IMP</th>\n",
       "      <th>BARREN-SHR</th>\n",
       "      <th>BARREN-TRE</th>\n",
       "      <th>BARREN-TS</th>\n",
       "      <th>CHANGE_DUR</th>\n",
       "      <th>CP_Code</th>\n",
       "      <th>DOM_LU_Code</th>\n",
       "      <th>DOM_SEC_LC_Code</th>\n",
       "      <th>...</th>\n",
       "      <th>swir2_LT_dur</th>\n",
       "      <th>swir2_LT_fitted</th>\n",
       "      <th>swir2_LT_mag</th>\n",
       "      <th>swir2_LT_slope</th>\n",
       "      <th>wetness_LT_diff</th>\n",
       "      <th>wetness_LT_dur</th>\n",
       "      <th>wetness_LT_fitted</th>\n",
       "      <th>wetness_LT_mag</th>\n",
       "      <th>wetness_LT_slope</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>19777.0</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.0</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "      <td>19777.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.005360</td>\n",
       "      <td>0.098094</td>\n",
       "      <td>0.004955</td>\n",
       "      <td>0.002174</td>\n",
       "      <td>0.0</td>\n",
       "      <td>14.327350</td>\n",
       "      <td>1.459372</td>\n",
       "      <td>3.307933</td>\n",
       "      <td>4.352885</td>\n",
       "      <td>...</td>\n",
       "      <td>26.305810</td>\n",
       "      <td>0.079596</td>\n",
       "      <td>0.005514</td>\n",
       "      <td>0.000503</td>\n",
       "      <td>-0.008305</td>\n",
       "      <td>27.750417</td>\n",
       "      <td>-0.073550</td>\n",
       "      <td>-0.011920</td>\n",
       "      <td>-0.000660</td>\n",
       "      <td>2011.236494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.073016</td>\n",
       "      <td>0.297449</td>\n",
       "      <td>0.070221</td>\n",
       "      <td>0.046579</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.767562</td>\n",
       "      <td>1.035758</td>\n",
       "      <td>1.297727</td>\n",
       "      <td>4.435511</td>\n",
       "      <td>...</td>\n",
       "      <td>11.267701</td>\n",
       "      <td>0.039772</td>\n",
       "      <td>0.033840</td>\n",
       "      <td>0.005742</td>\n",
       "      <td>0.029825</td>\n",
       "      <td>11.333125</td>\n",
       "      <td>0.044698</td>\n",
       "      <td>0.043909</td>\n",
       "      <td>0.006246</td>\n",
       "      <td>6.040297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.006700</td>\n",
       "      <td>-0.210300</td>\n",
       "      <td>-0.070100</td>\n",
       "      <td>-0.372316</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.360283</td>\n",
       "      <td>-0.393000</td>\n",
       "      <td>-0.176800</td>\n",
       "      <td>2000.517822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>0.053306</td>\n",
       "      <td>-0.011300</td>\n",
       "      <td>-0.000481</td>\n",
       "      <td>-0.022232</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>-0.097389</td>\n",
       "      <td>-0.034600</td>\n",
       "      <td>-0.001229</td>\n",
       "      <td>2006.169922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.069844</td>\n",
       "      <td>0.006200</td>\n",
       "      <td>0.000230</td>\n",
       "      <td>-0.007874</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>-0.067405</td>\n",
       "      <td>-0.014300</td>\n",
       "      <td>-0.000466</td>\n",
       "      <td>2011.295898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>0.095467</td>\n",
       "      <td>0.021800</td>\n",
       "      <td>0.000900</td>\n",
       "      <td>0.005050</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>-0.045750</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>0.000423</td>\n",
       "      <td>2016.476685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>0.373600</td>\n",
       "      <td>0.300500</td>\n",
       "      <td>0.204700</td>\n",
       "      <td>0.212600</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>0.072426</td>\n",
       "      <td>0.212600</td>\n",
       "      <td>0.053150</td>\n",
       "      <td>2021.411011</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 162 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        BARREN    BARREN-GRA    BARREN-IMP    BARREN-SHR    BARREN-TRE  \\\n",
       "count  19777.0  19777.000000  19777.000000  19777.000000  19777.000000   \n",
       "mean       0.0      0.005360      0.098094      0.004955      0.002174   \n",
       "std        0.0      0.073016      0.297449      0.070221      0.046579   \n",
       "min        0.0      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.0      0.000000      0.000000      0.000000      0.000000   \n",
       "50%        0.0      0.000000      0.000000      0.000000      0.000000   \n",
       "75%        0.0      0.000000      0.000000      0.000000      0.000000   \n",
       "max        0.0      1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "       BARREN-TS    CHANGE_DUR       CP_Code   DOM_LU_Code  DOM_SEC_LC_Code  \\\n",
       "count    19777.0  19777.000000  19777.000000  19777.000000     19777.000000   \n",
       "mean         0.0     14.327350      1.459372      3.307933         4.352885   \n",
       "std          0.0      6.767562      1.035758      1.297727         4.435511   \n",
       "min          0.0      1.000000      1.000000      1.000000         1.000000   \n",
       "25%          0.0      9.000000      1.000000      3.000000         1.000000   \n",
       "50%          0.0     17.000000      1.000000      3.000000         1.000000   \n",
       "75%          0.0     21.000000      1.000000      3.000000         9.000000   \n",
       "max          0.0     21.000000      4.000000      6.000000        14.000000   \n",
       "\n",
       "       ...  swir2_LT_dur  swir2_LT_fitted  swir2_LT_mag  swir2_LT_slope  \\\n",
       "count  ...  19777.000000     19777.000000  19777.000000    19777.000000   \n",
       "mean   ...     26.305810         0.079596      0.005514        0.000503   \n",
       "std    ...     11.267701         0.039772      0.033840        0.005742   \n",
       "min    ...      1.000000         0.006700     -0.210300       -0.070100   \n",
       "25%    ...     17.000000         0.053306     -0.011300       -0.000481   \n",
       "50%    ...     28.000000         0.069844      0.006200        0.000230   \n",
       "75%    ...     38.000000         0.095467      0.021800        0.000900   \n",
       "max    ...     38.000000         0.373600      0.300500        0.204700   \n",
       "\n",
       "       wetness_LT_diff  wetness_LT_dur  wetness_LT_fitted  wetness_LT_mag  \\\n",
       "count     19777.000000    19777.000000       19777.000000    19777.000000   \n",
       "mean         -0.008305       27.750417          -0.073550       -0.011920   \n",
       "std           0.029825       11.333125           0.044698        0.043909   \n",
       "min          -0.372316        1.000000          -0.360283       -0.393000   \n",
       "25%          -0.022232       19.000000          -0.097389       -0.034600   \n",
       "50%          -0.007874       32.000000          -0.067405       -0.014300   \n",
       "75%           0.005050       38.000000          -0.045750        0.010000   \n",
       "max           0.212600       38.000000           0.072426        0.212600   \n",
       "\n",
       "       wetness_LT_slope          year  \n",
       "count      19777.000000  19777.000000  \n",
       "mean          -0.000660   2011.236494  \n",
       "std            0.006246      6.040297  \n",
       "min           -0.176800   2000.517822  \n",
       "25%           -0.001229   2006.169922  \n",
       "50%           -0.000466   2011.295898  \n",
       "75%            0.000423   2016.476685  \n",
       "max            0.053150   2021.411011  \n",
       "\n",
       "[8 rows x 162 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "# Now will download the training table to a local location\n",
    "local_model_data_folder = '/tmp/lcms-training/local_modeling'\n",
    "local_training_csv = os.path.join(local_model_data_folder,'timeSync_training_table.csv')\n",
    "\n",
    "if not os.path.exists(local_model_data_folder):os.makedirs(local_model_data_folder)\n",
    "\n",
    "# Download the training data from a featureCollection to a local CSV\n",
    "# This function will automatically break the featureCollection into 5000 feature featureCollections\n",
    "# and then put it back together if it is larger than the 5000 feature limit set by GEE\n",
    "g2p.featureCollection_to_csv(training_data,local_training_csv,overwrite = False) # overwrite = false as default\n",
    "\n",
    "# Once the table is store locally, read it in\n",
    "training_df = pd.read_csv(local_training_csv)\n",
    "\n",
    "display(training_df.describe())\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ac069c9",
   "metadata": {},
   "source": [
    "#### Set up RF model parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "00204cd2-8151-4176-b77c-eadabfa2e2b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "# Set up k-fold for Change thresholding \n",
    "\n",
    "\n",
    "# load model options df\n",
    "model_options = pd.read_csv(model_options_csv_filename)\n",
    "\n",
    "# First set up a the predictor and reference field names\n",
    "model_options = model_options[model_options['Model Name'] == 'Non-correlated Predictors Top 30']\n",
    "\n",
    "# Specify the field for each product\n",
    "reference_field_dict = {'Land_Cover':{'field':'DOM_SEC_LC'},\n",
    "                        'Land_Use':{'field':'DOM_LU'},\n",
    "                        'Change':{'field':'CP'}\n",
    "                       }\n",
    "\n",
    "# Only needed for Change currently\n",
    "product_name = 'Change'\n",
    "\n",
    "groups = training_df['PLOTID'].squeeze()\n",
    "\n",
    "# Specify k (generally 5 -10 works well)\n",
    "k = 10\n",
    "n_jobs = 4\n",
    "\n",
    "# We will try to align parameters between the sklearn and GEE instances of RF\n",
    "nTrees = 150\n",
    "seed = 999\n",
    "\n",
    "min_samples_split = 2 # The minimum number of samples required to split an internal node\n",
    "min_samples_leaf = 1 # The minimum number of samples required to be at a leaf node\n",
    "max_features = 'sqrt' # The number of features to consider when looking for the best split ('sqrt' is the default in sklearn and GEE)\n",
    "max_samples = 0.632 # The number of samples to draw to train each tree. 0.632 is the default from the R instance. 0.5 is the default in GEE.\n",
    "\n",
    "\n",
    "# Set up groups\n",
    "gkf = GroupKFold(n_splits=k)\n",
    "foldNum = 1\n",
    "\n",
    "# Set up a random forest model\n",
    "rf = RandomForestClassifier(n_estimators = nTrees, \n",
    "                            random_state=seed,\n",
    "                            oob_score=False,\n",
    "                            min_samples_split = min_samples_split,\n",
    "                            min_samples_leaf= min_samples_leaf,\n",
    "                            max_features = max_features,\n",
    "                            max_samples = max_samples,\n",
    "                            n_jobs = n_jobs\n",
    "                            )\n",
    "# Pull predictors from table from 5.1\n",
    "# Some parsing is needed to read it in properly\n",
    "predictor_variable_names = model_options[model_options['Product Name'] == product_name]['Var Imp'].values[0]\n",
    "predictor_variable_names = predictor_variable_names[1:-1]\n",
    "predictor_variable_names=predictor_variable_names.replace(\"'\",\"\").split(', ')\n",
    "\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70f8020a",
   "metadata": {},
   "source": [
    "### 6.2.2 Run RF model\n",
    "\n",
    "* This will take some time to run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "1ca1e73f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold 1 has 17797 training samples and 1980 test samples\n",
      "Fitting model for fold 1, Change\n",
      "Fold 2 has 17797 training samples and 1980 test samples\n",
      "Fitting model for fold 2, Change\n",
      "Fold 3 has 17805 training samples and 1972 test samples\n",
      "Fitting model for fold 3, Change\n",
      "Fold 4 has 17805 training samples and 1972 test samples\n",
      "Fitting model for fold 4, Change\n",
      "Fold 5 has 17797 training samples and 1980 test samples\n",
      "Fitting model for fold 5, Change\n",
      "Fold 6 has 17798 training samples and 1979 test samples\n",
      "Fitting model for fold 6, Change\n",
      "Fold 7 has 17798 training samples and 1979 test samples\n",
      "Fitting model for fold 7, Change\n",
      "Fold 8 has 17798 training samples and 1979 test samples\n",
      "Fitting model for fold 8, Change\n",
      "Fold 9 has 17799 training samples and 1978 test samples\n",
      "Fitting model for fold 9, Change\n",
      "Fold 10 has 17799 training samples and 1978 test samples\n",
      "Fitting model for fold 10, Change\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "y_refs = []\n",
    "y_preds = []\n",
    "# Iterate over k folds\n",
    "for train_index, test_index in gkf.split(training_df, training_df, groups):\n",
    "    \n",
    "    \n",
    "    print(f'Fold {foldNum} has {len(train_index)} training samples and {len(test_index)} test samples')\n",
    "\n",
    "    # Pull the tran and test plots\n",
    "    k_train,k_test = training_df.iloc[train_index], training_df.iloc[test_index]\n",
    "    \n",
    "\n",
    "    print(f'Fitting model for fold {foldNum}, {product_name}')\n",
    "    \n",
    "   \n",
    "    # Get X and Y points for each group  \n",
    "    kx_train = k_train[predictor_variable_names]\n",
    "    ky_train = k_train[reference_field_dict[product_name]['field']+'_Code']\n",
    "    \n",
    "    kx_test = k_test[predictor_variable_names]\n",
    "    ky_test = k_test[reference_field_dict[product_name]['field']+'_Code']\n",
    "    \n",
    "    # Fit model\n",
    "    rf.fit(kx_train,ky_train)\n",
    "    \n",
    "    # Get predicted classes or probabilities for each Test Point\n",
    "    ky_pred = rf.predict_proba(kx_test)\n",
    "    \n",
    "    y_refs.extend(ky_test)\n",
    "    y_preds.extend(ky_pred)\n",
    "    foldNum+=1\n",
    "\n",
    "y_preds = np.array(y_preds)\n",
    "y_refs = np.array(y_refs)\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24c65c94",
   "metadata": {},
   "source": [
    "### 6.2.3: Compute proportion of votes that balance omission and comission\n",
    "* Next we'll compute the proportion of votes that balances omission and commission\n",
    "\n",
    "\n",
    "* Since a pixel may have probabilities above the threshold of more than one class, we choose the class with the model confidence that is the highest probability that is also above its respective threshold. \n",
    "* e.g. if the fast loss probability is 0.5 and the slow loss probability is 0.35, the pixel will be assigned fast loss\n",
    "* We will not include stable since stable will be any pixel that is not already assigned to one of the 3 change classes\n",
    "\n",
    "* Before jumping into computing the actual thresholds, in order to illustrate how we compute the threshold, the next block of code will walk you through applying a single threshold to a single class\n",
    "* It will then show the results of each step\n",
    "* You can change the class and threshold that is applied to explore its impacts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "64370ca7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Raw model probabilities:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Stable</th>\n",
       "      <th>Slow Loss</th>\n",
       "      <th>Fast Loss</th>\n",
       "      <th>Gain</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.026667</td>\n",
       "      <td>0.006667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.686667</td>\n",
       "      <td>0.013333</td>\n",
       "      <td>0.093333</td>\n",
       "      <td>0.206667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.013333</td>\n",
       "      <td>0.020000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.106667</td>\n",
       "      <td>0.226667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.893333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.073333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19772</th>\n",
       "      <td>0.960000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.020000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19773</th>\n",
       "      <td>0.713333</td>\n",
       "      <td>0.013333</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>0.140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19774</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.273333</td>\n",
       "      <td>0.226667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19775</th>\n",
       "      <td>0.546667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.073333</td>\n",
       "      <td>0.380000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19776</th>\n",
       "      <td>0.886667</td>\n",
       "      <td>0.006667</td>\n",
       "      <td>0.046667</td>\n",
       "      <td>0.060000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19777 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Stable  Slow Loss  Fast Loss      Gain\n",
       "0      0.966667   0.000000   0.026667  0.006667\n",
       "1      0.686667   0.013333   0.093333  0.206667\n",
       "2      0.966667   0.000000   0.013333  0.020000\n",
       "3      0.666667   0.000000   0.106667  0.226667\n",
       "4      0.893333   0.000000   0.033333  0.073333\n",
       "...         ...        ...        ...       ...\n",
       "19772  0.960000   0.000000   0.020000  0.020000\n",
       "19773  0.713333   0.013333   0.133333  0.140000\n",
       "19774  0.500000   0.000000   0.273333  0.226667\n",
       "19775  0.546667   0.000000   0.073333  0.380000\n",
       "19776  0.886667   0.006667   0.046667  0.060000\n",
       "\n",
       "[19777 rows x 4 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Raw Fast Loss column:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Fast Loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.026667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.093333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.013333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.106667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.033333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19772</th>\n",
       "      <td>0.020000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19773</th>\n",
       "      <td>0.133333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19774</th>\n",
       "      <td>0.273333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19775</th>\n",
       "      <td>0.073333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19776</th>\n",
       "      <td>0.046667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19777 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Fast Loss\n",
       "0       0.026667\n",
       "1       0.093333\n",
       "2       0.013333\n",
       "3       0.106667\n",
       "4       0.033333\n",
       "...          ...\n",
       "19772   0.020000\n",
       "19773   0.133333\n",
       "19774   0.273333\n",
       "19775   0.073333\n",
       "19776   0.046667\n",
       "\n",
       "[19777 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Observed and predicted binary:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Observed Fast Loss</th>\n",
       "      <th>Predicted Fast Loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19772</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19773</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19774</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19775</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19776</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19777 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Observed Fast Loss  Predicted Fast Loss\n",
       "0                       0                    0\n",
       "1                       0                    0\n",
       "2                       0                    0\n",
       "3                       0                    1\n",
       "4                       0                    0\n",
       "...                   ...                  ...\n",
       "19772                   0                    0\n",
       "19773                   0                    1\n",
       "19774                   0                    1\n",
       "19775                   0                    0\n",
       "19776                   0                    0\n",
       "\n",
       "[19777 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Resulting confusion matrix:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Observed Not Fast Loss</th>\n",
       "      <th>Observed Fast Loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Predicted Not Fast Loss</th>\n",
       "      <td>16249</td>\n",
       "      <td>476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Predicted Fast Loss</th>\n",
       "      <td>2627</td>\n",
       "      <td>425</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         Observed Not Fast Loss  Observed Fast Loss\n",
       "Predicted Not Fast Loss                   16249                 476\n",
       "Predicted Fast Loss                        2627                 425"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Fast Loss &gt; 0.1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Omission Error Rate</th>\n",
       "      <td>0.528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Commission Error Rate</th>\n",
       "      <td>0.861</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Fast Loss > 0.1\n",
       "Omission Error Rate              0.528\n",
       "Commission Error Rate            0.861"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# First, we'll look at all the predicted probabilities\n",
    "change_lookup = {'Stable':1,'Slow Loss':2,'Fast Loss':3,'Gain':4}\n",
    "print('Raw model probabilities:')\n",
    "display(pd.DataFrame(y_preds,columns=change_lookup.keys()))\n",
    "\n",
    "# Specify which change class to work with\n",
    "# Feel free to change the code\n",
    "change_class = 'Fast Loss'\n",
    "\n",
    "change_code = change_lookup[change_class]\n",
    "\n",
    "# Convert the reference data to yes/no \n",
    "y_ref_binary = np.where(y_refs==change_code, 1, 0)\n",
    "\n",
    "# Pull the probabilities for that column\n",
    "column_i = change_code-1\n",
    "column = y_preds[:,column_i]\n",
    "print(f'Raw {change_class} column:')\n",
    "display(pd.DataFrame(column,columns=[change_class]))\n",
    "\n",
    "# Specify a threshold\n",
    "threshold = 0.1\n",
    "\n",
    "# Apply the threshold\n",
    "thresholded_pred = np.where(column>threshold, 1, 0)\n",
    "obs_pred = np.array([y_ref_binary,thresholded_pred]).T\n",
    "print('Observed and predicted binary:')\n",
    "display(pd.DataFrame(obs_pred,columns=[f'Observed {change_class}',f'Predicted {change_class}']))\n",
    "\n",
    "# Create a confusion matrix\n",
    "cm = metrics.confusion_matrix(y_ref_binary,thresholded_pred).T\n",
    "print('Resulting confusion matrix:')\n",
    "display(pd.DataFrame(cm,index=[f'Predicted Not {change_class}',f'Predicted {change_class}'],columns=[f'Observed Not {change_class}',f'Observed {change_class}']))\n",
    "\n",
    "# Compute the error rates\n",
    "\n",
    "# Find the observed and predicted positives\n",
    "OP = np.sum(cm[:,1])\n",
    "PP = np.sum(cm[1])\n",
    "# Find the true positives\n",
    "TP = cm[1][1]\n",
    "\n",
    "# Compute precision and recall\n",
    "precision = TP/PP\n",
    "recall = TP/OP\n",
    "\n",
    "commission = 1-precision\n",
    "omission = 1-recall\n",
    "\n",
    "display(pd.DataFrame([omission,commission],index = ['Omission Error Rate','Commission Error Rate'],columns=[f'{change_class} > {threshold}']).round(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "4f39302e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Stable': 0.655, 'Slow Loss': 0.34500000000000003, 'Fast Loss': 0.17, 'Gain': 0.27}\n"
     ]
    }
   ],
   "source": [
    "# Next we'll compute the proportion of votes that balances omission and commission\n",
    "\n",
    "# Specify individual class names and codes for change\n",
    "change_class_codes = {'Stable':1,'Slow Loss':2,'Fast Loss':3,'Gain':4}\n",
    "\n",
    "# Make a function that will take a list of binary reference and probablity (0-1 float) \n",
    "# predicted values and find the optimal threshold to balance omission and commission\n",
    "def getBalancedThreshold(y_ref_binary,y_pred_probs,title='Omission-Commission Error',thresholds = list(np.arange(0,0.7,0.005)),simple_plot = True):\n",
    "    \n",
    "    omission_list = []\n",
    "    commission_list = []\n",
    "    \n",
    "    precision_list = []\n",
    "    recall_list = []\n",
    "\n",
    "    balanced_error_list = []\n",
    "    overall_error_list = []\n",
    "\n",
    "    # Iterate across each threshold and compute the omission and commision error rates\n",
    "    for threshold in thresholds:\n",
    "        \n",
    "        # Apply threshold to the probability\n",
    "        y_pred_binary = np.where(y_pred_probs>threshold, 1, 0)\n",
    "        \n",
    "        # Convert to 1:1 confusion matrix\n",
    "        cm = metrics.confusion_matrix(y_ref_binary,y_pred_binary).T\n",
    "        \n",
    "        # Find observed and predicted positives and negative sums (rows and columns)\n",
    "        OP = np.sum(cm[:,1])\n",
    "        ON = np.sum(cm[:,0])\n",
    "        PP = np.sum(cm[1])\n",
    "        PN = np.sum(cm[0])\n",
    "        \n",
    "        # Find overall sum\n",
    "        N = np.sum(cm)\n",
    "\n",
    "        # Give descriptive names to each of the four parts of the matrix\n",
    "        TP = cm[1][1]\n",
    "        TN = cm[0][0]\n",
    "        FP = cm[1][0]\n",
    "        FN = cm[0][1]\n",
    "\n",
    "        # Compute recall (1-omission) aka sensitivity\n",
    "        if OP == 0:\n",
    "            recall_sensitivity = 0\n",
    "        else:\n",
    "            recall_sensitivity = TP/OP\n",
    "        recall_list.append(recall_sensitivity)\n",
    "\n",
    "        # Compute precision (1-commission)\n",
    "        if PP == 0:\n",
    "            precision = 0\n",
    "        else:\n",
    "            precision = TP/PP\n",
    "        precision_list.append(precision)\n",
    "        \n",
    "        # Compute accuracy metrics\n",
    "        balanced_accuracy = (recall_sensitivity+ (1-FP/N))/2\n",
    "        balanced_error_list.append(1-balanced_accuracy)\n",
    "\n",
    "        overall_accuracy = (TP+TN)/N\n",
    "        overall_error_list.append(1-overall_accuracy)\n",
    "\n",
    "        # Populate commission and omission\n",
    "        commission_list.append(1-precision)\n",
    "        omission_list.append(1-recall_sensitivity)\n",
    "\n",
    "   \n",
    "    #Find the threshold that corresponds to the smallest difference between omission and commission\n",
    "    diff = np.abs(np.array(omission_list)-np.array(commission_list))\n",
    "    diff = np.stack((thresholds,diff,omission_list,commission_list),1)\n",
    "    diff = diff[diff[:, 1].argsort()]\n",
    "    crossover = diff[0]\n",
    "    optimal_thresh = crossover[0]\n",
    "    \n",
    "    # Plot the curves\n",
    "    plt.plot(thresholds, commission_list,color = '#FF834C',label='Commission')\n",
    "    plt.plot(thresholds,omission_list,color = '#00897b',label='Omission')\n",
    "\n",
    "    if not simple_plot:\n",
    "        plt.plot(thresholds, precision_list,color = '#FF004C',label='Precision')\n",
    "        plt.plot(thresholds,recall_list,color = '#00007b',label='Recall')\n",
    "\n",
    "        plt.plot(thresholds,balanced_error_list,color = '#BBBB7b',label='Balanced Error')\n",
    "        plt.plot(thresholds,overall_error_list,color = '#BB83BB',label='Overall Error')\n",
    "\n",
    "    plt.axvline(optimal_thresh,linestyle=':', color = '#372e2c', label = f'Optimal Probability: {optimal_thresh}')\n",
    "    plt.axhline(crossover[2],linestyle='-.', color = '#372e2c', label = f'Balanced Error Rate: {crossover[2].round(3)}')\n",
    "    plt.xlabel(f\"Probability\")\n",
    "    plt.ylabel(\"Error Rate\")\n",
    "   \n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "    \n",
    "    return optimal_thresh\n",
    "\n",
    "# Iterate across each class and find the optimal threshold\n",
    "out_threshold_dict = {}\n",
    "for change_class,change_code in list(change_class_codes.items()):\n",
    "    # print(change_class,change_code)\n",
    "    y_pred_prob_class =y_preds[:,change_code-1]\n",
    "    \n",
    "    y_ref_binary = np.where(y_refs==change_code, 1, 0)\n",
    "    # print(y_ref_binary)\n",
    "    rf_holdout_thresh = getBalancedThreshold(y_ref_binary,y_pred_prob_class,f'Commission - Omission Plot - {change_class}')\n",
    "    out_threshold_dict[change_class] = rf_holdout_thresh\n",
    "\n",
    "print(out_threshold_dict)\n",
    "\n",
    "# Save table for use \n",
    "model_change_thresholds_filename = os.path.join(local_model_data_folder,'LCMS_change_thresholds.json')\n",
    "o = open(model_change_thresholds_filename,'w')\n",
    "o.write(json.dumps(out_threshold_dict))\n",
    "o.close()\n",
    "\n",
    "\n",
    "\n",
    "# print('Done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68a427f2",
   "metadata": {},
   "source": [
    "* Notice the slow loss threshold makes little sense with error rate near 1. Recall we have very few slow loss samples. This indicates our models are unlikely to be able to predict slow loss at all.\n",
    "* Note that both fast loss and gain have thresholds far below 0.5 and error rates around 0.75 or so. This indicates our models are rarely confident of classifying these classes, and the omission and commission error rates are high relative to our reference dataset. \n",
    "* Simply put, TimeSync is much more sensitive to detecting change than any model is."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4e9b49d",
   "metadata": {},
   "source": [
    "### 6.2.4: Final Change Layer Assemblage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "996ec2b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adding layer: Assembled LCMS Change Time-Lapse\n",
      "Adding layer: LCMS Max Prob Default Change 2017\n",
      "Adding layer: LCMS Balanced Omission/Commission Change 2017\n",
      "Setting default query date format to: YYYY\n",
      "Starting webmap\n",
      "Using default refresh token for geeView: C:\\Users\\ihousman/.config/earthengine/credentials\n",
      "Local web server at: http://localhost:1236/geeView/ already serving.\n",
      "cwd z:\\Projects\\06_LCMS_4_NFS\\Scripts\\lcms-training\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"525px\"\n",
       "            src=\"http://localhost:1236/geeView/?accessToken=ya29.a0AfB_byB8NqrjDWoTBste45ZOcxwur9k26wC-0Rfnluq_JzPi_ZFZsaClu2w1OhGgOPjiCVdWTj4v3yd-0e4AS4YXHVdYYpDpP1I6gzUuQbyeLCYnr3Io-XO8vii1zCHymQNICPO6fLXqwqqQkOpbx3euOLpCRYOMAawr_mVoUoqoaCgYKAYsSARASFQHGX2MiyH8iaGeALM_t8NY7_DzC2g0179\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x281e88e0f90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Map.clearMap()\n",
    "\n",
    "studyArea = ee.FeatureCollection('projects/lcms-292214/assets/R8/PR_USVI/Ancillary/prusvi_boundary')\n",
    "\n",
    "# Specify which Change classes to include in assemblage\n",
    "# By default, LCMS only uses Slow Loss, Fast Loss, and Gain and any remaining pixel becomes stable\n",
    "o = open(model_change_thresholds_filename,'r')\n",
    "change_thresholds = json.load(o)\n",
    "o.close()\n",
    "\n",
    "# Specify which change clases to include\n",
    "#change_classes = ['Slow Loss','Fast Loss','Gain']\n",
    "change_classes = ['Fast Loss','Gain']\n",
    "\n",
    "# Filter to only include change classes we want to include \n",
    "# Multiply probability by 100 so it's the same scale as raw array output\n",
    "change_thresholds = {k:v*100  for k,v in change_thresholds.items() if k in change_classes}\n",
    "\n",
    "# Specify the Change product\n",
    "product = 'Change'\n",
    "\n",
    "# Pull the class names\n",
    "class_names = lcms_viz_dict[f'{product}_class_names']\n",
    "class_codes = lcms_viz_dict[f'{product}_class_values']\n",
    "class_dict = dict(zip(class_names,class_codes))\n",
    "\n",
    "# Pull the codes in the correct order\n",
    "class_order_codes = [class_dict[cl] for cl in list(change_thresholds.keys())]\n",
    "\n",
    "# Filter to only include the product of interest\n",
    "raw_lcms_product = raw_lcms.filter(ee.Filter.eq('product',product))\n",
    "    \n",
    "# Function to assemble change for a given year\n",
    "# Uses a specified set of bands and thresholds in order to find the highest probability class above its threshold\n",
    "# It then fills in the stable value as any remaining non null value\n",
    "# Nulls are then recoded to the non processing code\n",
    "def assembleChange(img,product):\n",
    "    # Convert array to multi-band image\n",
    "    img = ee.Image(arrayToImage(img,lcms_viz_dict[f'{product}_class_names'][:-1]))\n",
    "    \n",
    "    # Select only the bands we'd like to assemble\n",
    "    img = img.select(list(change_thresholds.keys()))\n",
    "    \n",
    "    # Find the most confident class\n",
    "    maxConf = img.reduce(ee.Reducer.max())\n",
    "    \n",
    "    # Find pixels that are the most confident and above the threshold\n",
    "    maxConfMask = img.eq(maxConf).And(img.gte(list(change_thresholds.values()))).selfMask()\n",
    "    \n",
    "    # Get class code for any pixel that is above its threshold for the most confident class\n",
    "    maxClass = ee.Image(class_order_codes).updateMask(maxConfMask).reduce(ee.Reducer.firstNonNull())\n",
    "    \n",
    "    # Fill stable back in\n",
    "    maxClass = maxClass.unmask(1)\n",
    "    \n",
    "    # Burn in non processing where the input was null\n",
    "    maxClass = maxClass.where(img.mask().reduce(ee.Reducer.min()).Not(),class_codes[-1])\n",
    "    \n",
    "    return maxClass.rename([product]).copyProperties(img).copyProperties(img,['system:time_start'])\n",
    "\n",
    "# Iterate across the change probabilities and assemble final change class\n",
    "assembledChange = raw_lcms_product.map(lambda img:assembleChange(img,product).set(lcms_viz_dict))\n",
    "\n",
    "# Explore  time lapse on the map\n",
    "Map.addTimeLapse(assembledChange,{'autoViz':True},'Assembled LCMS Change Time-Lapse')\n",
    "\n",
    "# Compare the default Change output (max model confidence) and the output that balances\n",
    "# omission and commission of the change categories\n",
    "year = 2017\n",
    "Map.addLayer(max_prob_lcms_classes['Change'].filter(ee.Filter.calendarRange(year,year,'year')),{'autoViz':True},f'LCMS Max Prob Default Change {year}')\n",
    "Map.addLayer(assembledChange.filter(ee.Filter.calendarRange(year,year,'year')),{'autoViz':True},f'LCMS Balanced Omission/Commission Change {year}')\n",
    "\n",
    "Map.turnOnInspector()\n",
    "Map.setQueryDateFormat('YYYY')\n",
    "Map.centerObject(ee.Geometry.Point(-66.095, 18.400), 13)\n",
    "Map.view()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5d96d9d",
   "metadata": {},
   "source": [
    "* Notice the Balanced Omission/Commission Change layer detects much more change than the Max Prob Default Change"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35c784a7-b621-44fd-96dd-01d374b4ee4a",
   "metadata": {},
   "source": [
    "#### Inspect raw probability of change layers\n",
    "\n",
    "* Compared against assembled change for a single year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "58724529-1c48-48a0-b65a-72ecfa0cbba0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adding layer: Raw Gain Prob 2017\n",
      "Adding layer: Raw Slow Loss Prob 2017\n",
      "Adding layer: Raw Fast Loss Prob 2017\n",
      "Adding layer: Assembled Change 2017\n",
      "Starting webmap\n",
      "Using default refresh token for geeView: C:\\Users\\ihousman/.config/earthengine/credentials\n",
      "Local web server at: http://localhost:1236/geeView/ already serving.\n",
      "cwd z:\\Projects\\06_LCMS_4_NFS\\Scripts\\lcms-training\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"525px\"\n",
       "            src=\"http://localhost:1236/geeView/?accessToken=ya29.a0AfB_byBLZhIV0oM1WN98bdOhS2A2Ya3tfatuPta99iLXEJeUjST2JFrr5p7tiaqw-E1HZm8sZEYgPElWxYTTNwEdD9UJimOZUN9ZaEKasgXKeHY4sFW1c1GOuDofUBtHfQqSN8zh6s0v5vedCG1M02pmmnTdu3tqh8ab0WYOmJMOaCgYKAfUSARASFQHGX2Mi70utWe6DDvaajFbfJznmBg0179\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x281e8acb2d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#### Inspect raw probability of change layers\n",
    "Map.clearMap()\n",
    "\n",
    "# select single year\n",
    "filterYear = 2017\n",
    "\n",
    "changeDetectionLib.lossMagPalette.reverse()\n",
    "\n",
    "# Divide by 100 to show prob 0-1 instead of 0-1\n",
    "imgs = raw_lcms_product.map(lambda img:arrayToImage(img, lcms_viz_dict['Change_class_names'][:-1]))\n",
    "imgs = imgs.select(list(['Slow Loss', 'Fast Loss', 'Gain']))\n",
    "imgs = imgs.filter(ee.Filter.calendarRange(filterYear, filterYear, 'year')).first().divide(100)\n",
    "\n",
    "\n",
    "# Select only the bands we'd like to visualize\n",
    "raw_slow_loss_prob = imgs.select([0])\n",
    "raw_fast_loss_prob = imgs.select([1])\n",
    "raw_gain_prob = imgs.select([2])\n",
    "    \n",
    "# Add raw loss probabilities to the map \n",
    "Map.addLayer(raw_gain_prob, {'min': 0.1, 'max': 0.4, 'palette': changeDetectionLib.gainMagPalette}, f'Raw Gain Prob {filterYear}') \n",
    "Map.addLayer(raw_slow_loss_prob, {'min': 0.1, 'max': 0.3, 'palette': changeDetectionLib.lossMagPalette}, f'Raw Slow Loss Prob {filterYear}')\n",
    "Map.addLayer(raw_fast_loss_prob, {'min': 0.1, 'max': 0.3, 'palette': changeDetectionLib.lossMagPalette}, f'Raw Fast Loss Prob {filterYear}')\n",
    "\n",
    "\n",
    "# Add assembled change to map for comparison\n",
    "Map.addLayer(assembledChange.filter(ee.Filter.calendarRange(filterYear, filterYear, 'year')).first(), {'autoViz': True}, f'Assembled Change {filterYear}')\n",
    "\n",
    "Map.turnOnInspector()\n",
    "Map.view()\n",
    "\n",
    "             "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48ed452a",
   "metadata": {},
   "source": [
    "#### Single Class Year Summaries\n",
    "\n",
    "* Sometimes instead of looking at individual years, we want to see the a composite of the most recent year a loss occurred \n",
    "* This example will illustrate how to show the most recent year of Fast Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "d8fd1661",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adding layer: Most Recent Year of Fast Loss\n",
      "Starting webmap\n",
      "Using default refresh token for geeView: C:\\Users\\ihousman/.config/earthengine/credentials\n",
      "Local web server at: http://localhost:1236/geeView/ already serving.\n",
      "cwd z:\\Projects\\06_LCMS_4_NFS\\Scripts\\lcms-training\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"525px\"\n",
       "            src=\"http://localhost:1236/geeView/?accessToken=ya29.a0AfB_byC_K_-qYeJxi7YOhoEicYmtF58bCR1KVJ7wc_nTlREeRfJAvrMDqV07BinZ6fWyWUars-lTuRPigH3ss6UrXBL3hBILH6D-6Q04evlyPXucamosrBRCrTmbP4lFNlfJ1WBG4KteTIjo9IV5bSj5gGbHw3aSL6d_aT3GXKAhaCgYKAekSARASFQHGX2MiNROdZ2Sg0XohYUjdQcg7kg0179\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x281e9cf7cd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Change', 'year']\n"
     ]
    }
   ],
   "source": [
    "Map.clearMap()\n",
    "# Viewing change outputs can be easier as a composite that shows the most recent year of a single change type\n",
    "# This example will show the most recent year of fast loss\n",
    "\n",
    "# First, add a year constant band\n",
    "assembledChange = assembledChange.map(getImagesLib.addYearBand)\n",
    "\n",
    "# Then mask where change equals fast loss (3)\n",
    "fastLoss = assembledChange.map(lambda img: img.updateMask(img.select([0]).eq(3)))\n",
    "\n",
    "# Find the most recent year\n",
    "mostRecentFastLoss = fastLoss.max().select(['year'])\n",
    "Map.addLayer(mostRecentFastLoss,{'min':1985,'max':2022,'palette':changeDetectionLib.lossYearPalette},'Most Recent Year of Fast Loss')\n",
    "\n",
    "Map.turnOnInspector()\n",
    "Map.view()\n",
    "print(assembledChange.first().bandNames().getInfo())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4dc2af0",
   "metadata": {},
   "source": [
    "## 6.4: Export assembled maps\n",
    "\n",
    "* Now lets export assembled assets\n",
    "\n",
    "* OPTIONAL!!!!\n",
    "* Optionally, we can export using the tile grid approach\n",
    "* For PRUSVI, LCMS does not need to export using this approach, but this is how you would set it up\n",
    "* First, we'll set up the study area and a tile to export across"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "23b5002b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done\n"
     ]
    }
   ],
   "source": [
    "Map.clearMap()\n",
    "# Now lets export assembled assets\n",
    "\n",
    "# OPTIONAL!!!!\n",
    "# Optionally, we can export using the tile grid approach\n",
    "# For PRUSVI, LCMS does not need to export using this approach, but this is how you would set it up\n",
    "\n",
    "# Specify whether to export across tile grid\n",
    "# Generally only needed if exports fail due to memory errors\n",
    "use_tile_grid = False\n",
    "\n",
    "# First, we'll set up the study area  to export across\n",
    "studyArea = ee.FeatureCollection('projects/lcms-292214/assets/R8/PR_USVI/Ancillary/prusvi_boundary')\n",
    "\n",
    "# Set the size (in meters) of the tiles\n",
    "# We can likely use a large tile for this step\n",
    "# If exports fail, reducing the tileSize is likely to help\n",
    "tileSize = 240000\n",
    "\n",
    "\n",
    "# Set the projection\n",
    "crs = getImagesLib.common_projections['NLCD_CONUS']['crs']\n",
    "transform  = getImagesLib.common_projections['NLCD_CONUS']['transform']\n",
    "scale = None\n",
    "projection = ee.Projection(crs,transform)\n",
    "\n",
    "\n",
    "# Get the grid\n",
    "grid = studyArea.geometry().coveringGrid(projection.atScale(tileSize))\n",
    "#Map.addLayer(grid,{},'Tile Grid {}m'.format(tileSize))\n",
    "\n",
    "# Get IDs of tiles\n",
    "ids = grid.aggregate_histogram('system:index').keys().getInfo()\n",
    "\n",
    "#Map.centerObject(grid)\n",
    "#Map.view()\n",
    "print(\"Done\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c61f3ba",
   "metadata": {},
   "source": [
    "* Iterate across each year and assemble full stack output\n",
    "* [Example of fully assembed bands](https://developers.google.com/earth-engine/datasets/catalog/USFS_GTAC_LCMS_v2022-8#bands)\n",
    "* We will not be producing the qa-bits bands"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b4832e2d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Exporting: LCMS_PRUSVI_v2022-Training_1985\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1986\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1987\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1988\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1989\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1990\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1991\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1992\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1993\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1994\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1995\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1996\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1997\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1998\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_1999\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2000\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2001\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2002\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2003\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2004\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2005\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2006\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2007\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2008\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2009\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2010\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2011\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2012\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2013\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2014\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2015\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2016\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2017\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2018\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2019\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2020\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2021\n",
      "Exporting: LCMS_PRUSVI_v2022-Training_2022\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "# Iterate across each year, assemble full stack output \n",
    "Map.clearMap()\n",
    "\n",
    "# Specify the products\n",
    "products = ['Change','Land_Cover','Land_Use']\n",
    "\n",
    "# Provide a study area name and version\n",
    "study_area_name = 'PRUSVI'\n",
    "version = '2022-Training'\n",
    "\n",
    "# Set up years to apply models across\n",
    "apply_years = list(range(1985,2022+1))\n",
    "\n",
    "\n",
    "# This will be populated on the first iteration\n",
    "pyramidingPolicy = None\n",
    "\n",
    "\n",
    "for apply_year in apply_years:\n",
    "    assembled_list = []\n",
    "    probability_list = []\n",
    "    for product in products:\n",
    "        # Pull the class names\n",
    "        class_names = lcms_viz_dict[f'{product}_class_names'][:-1]\n",
    "        \n",
    "        # Filter to only include the product of interest\n",
    "        raw_lcms_product = raw_lcms.filter(ee.Filter.eq('product',product))\n",
    "        \n",
    "        # Get rid of any characters that aren't allowed\n",
    "        class_names = [nm.replace(' ','-').replace('/','-').replace('&','and').replace('-(SEAK-Only)','') for nm in class_names]\n",
    "        \n",
    "        probability_names = [f'{product}_Raw_Probability_{class_name}' for class_name in class_names]\n",
    "        \n",
    "        product_title = product.replace('_',' ')\n",
    "\n",
    "        # Filter to only include the product of interest\n",
    "        raw_lcms_product_yr = raw_lcms\\\n",
    "                            .filter(ee.Filter.calendarRange(apply_year,apply_year,'year'))\\\n",
    "                            .filter(ee.Filter.eq('product',product)).first()\n",
    "        \n",
    "        # Fill in any missing values (Land Cover only for PRUSVI)\n",
    "        missing_names_product = missing_names[product]\n",
    "        missing_indices_product = [lcms_viz_dict[f'{product}_class_names'].index(missing_name_product) for missing_name_product in missing_names_product]\n",
    "        if len(missing_names_product) > 0:\n",
    "            raw_lcms_product_yr = fillArrayMulti(raw_lcms_product_yr,missing_indices_product,0)\n",
    "\n",
    "            \n",
    "        if product in ['Land_Cover','Land_Use']:\n",
    "            # Find the max probability class\n",
    "            maxProb_lcms_product_yr = getMostProbableClass(raw_lcms_product_yr,product)\n",
    "            assembled_list.append(maxProb_lcms_product_yr)\n",
    "        \n",
    "        elif product == 'Change':\n",
    "            change_lcms_product_yr = assembleChange(raw_lcms_product_yr,product)\n",
    "            assembled_list.append(change_lcms_product_yr)\n",
    "            \n",
    "        # Convert raw model probabilities to bands for exploration\n",
    "        prob_lcms_product_yr = arrayToImage(raw_lcms_product_yr,probability_names)\n",
    "        probability_list.append(prob_lcms_product_yr)\n",
    "\n",
    "    # Stack final image with assembed and raw probabilities   \n",
    "    final_stack = ee.Image(ee.Image.cat([ee.Image.cat(assembled_list),ee.Image.cat(probability_list)]).byte()\\\n",
    "                    .copyProperties(raw_lcms_product_yr,['study_area','year']))\n",
    "    \n",
    "    # Set up a pyramiding object for each band\n",
    "    # Final output bands are thematic and need a mode resampling method\n",
    "    # Probability bands are continuous, so mean is appropriate\n",
    "    if pyramidingPolicy == None:\n",
    "        bns = final_stack.bandNames().getInfo()\n",
    "        pyramidingPolicy = {}\n",
    "        for bn in bns:\n",
    "            if bn.find('_Probability_') == -1:\n",
    "                pyramidingPolicy[bn] = 'mode'\n",
    "            else:\n",
    "                pyramidingPolicy[bn] = 'mean'\n",
    "\n",
    "    if use_tile_grid:\n",
    "        for id in ids:\n",
    "            id_title = id.replace(',','-')\n",
    "\n",
    "            # Get the tile and buffer it so there are no missing pixels at tile edges\n",
    "            tile = grid.filter(ee.Filter.eq('system:index',id)).geometry().intersection(studyArea,240,projection).buffer(900)\n",
    "            final_stack = final_stack.set({'TileSize':tileSize,\n",
    "                     'TileID':id_title})\n",
    "\n",
    "\n",
    "            exportName = f'LCMS_{study_area_name}_v{version}_Tile-{tileSize}m_ID{id_title}_{apply_year}'\n",
    "            exportPath = f'{export_assembledLCMSOutputs_collection}/{exportName}'\n",
    "            \n",
    "\n",
    "            getImagesLib.exportToAssetWrapper(final_stack,exportName,exportPath,pyramidingPolicy,tile,scale,crs,transform,overwrite=False)\n",
    "\n",
    "    else:\n",
    "        exportName = f'LCMS_{study_area_name}_v{version}_{apply_year}'\n",
    "        exportPath = f'{export_assembledLCMSOutputs_collection}/{exportName}'\n",
    "\n",
    "       \n",
    "        getImagesLib.exportToAssetWrapper(final_stack,exportName,exportPath,pyramidingPolicy,studyArea,scale,crs,transform,overwrite=True)\n",
    "\n",
    "print('Done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "81c1449a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "35 tasks ready 2023-12-07 16:30:04\n",
      "3 tasks running 2023-12-07 16:30:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:02:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:02:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:02:45']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:09\n",
      "3 tasks running 2023-12-07 16:30:09\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:02:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:02:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:02:51']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:15\n",
      "3 tasks running 2023-12-07 16:30:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:02:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:02:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:02:56']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:20\n",
      "3 tasks running 2023-12-07 16:30:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:02:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:02']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:26\n",
      "3 tasks running 2023-12-07 16:30:26\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:02:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:08']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:32\n",
      "3 tasks running 2023-12-07 16:30:32\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:02:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:14']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:38\n",
      "3 tasks running 2023-12-07 16:30:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:20']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:44\n",
      "3 tasks running 2023-12-07 16:30:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:25']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:50\n",
      "3 tasks running 2023-12-07 16:30:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:31']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:30:56\n",
      "3 tasks running 2023-12-07 16:30:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:37']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:02\n",
      "3 tasks running 2023-12-07 16:31:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:43']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:08\n",
      "3 tasks running 2023-12-07 16:31:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:49']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:13\n",
      "3 tasks running 2023-12-07 16:31:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:03:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:03:55']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:19\n",
      "3 tasks running 2023-12-07 16:31:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:01']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:25\n",
      "3 tasks running 2023-12-07 16:31:25\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:06']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:31\n",
      "3 tasks running 2023-12-07 16:31:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:12']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:37\n",
      "3 tasks running 2023-12-07 16:31:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:03:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:18']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:42\n",
      "3 tasks running 2023-12-07 16:31:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:24']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:48\n",
      "3 tasks running 2023-12-07 16:31:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:30']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:31:54\n",
      "3 tasks running 2023-12-07 16:31:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:36']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:00\n",
      "3 tasks running 2023-12-07 16:32:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:42']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:06\n",
      "3 tasks running 2023-12-07 16:32:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:47']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:12\n",
      "3 tasks running 2023-12-07 16:32:12\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:53']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:18\n",
      "3 tasks running 2023-12-07 16:32:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:04:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:04:59']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:24\n",
      "3 tasks running 2023-12-07 16:32:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:05']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:30\n",
      "3 tasks running 2023-12-07 16:32:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:11']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:35\n",
      "3 tasks running 2023-12-07 16:32:35\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:04:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:17']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:41\n",
      "3 tasks running 2023-12-07 16:32:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:23']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:47\n",
      "3 tasks running 2023-12-07 16:32:47\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:28']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:53\n",
      "3 tasks running 2023-12-07 16:32:53\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:34']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:32:59\n",
      "3 tasks running 2023-12-07 16:32:59\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:40']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:04\n",
      "3 tasks running 2023-12-07 16:33:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:46']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:10\n",
      "3 tasks running 2023-12-07 16:33:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:51']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:16\n",
      "3 tasks running 2023-12-07 16:33:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:05:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:05:57']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:22\n",
      "3 tasks running 2023-12-07 16:33:22\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:03']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:28\n",
      "3 tasks running 2023-12-07 16:33:28\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:09']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:33\n",
      "3 tasks running 2023-12-07 16:33:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:05:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:15']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:39\n",
      "3 tasks running 2023-12-07 16:33:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:20']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:44\n",
      "3 tasks running 2023-12-07 16:33:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:26']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:50\n",
      "3 tasks running 2023-12-07 16:33:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:32']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:33:56\n",
      "3 tasks running 2023-12-07 16:33:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:37']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:02\n",
      "3 tasks running 2023-12-07 16:34:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:43']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:08\n",
      "3 tasks running 2023-12-07 16:34:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:49']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:14\n",
      "3 tasks running 2023-12-07 16:34:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:06:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:06:55']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:20\n",
      "3 tasks running 2023-12-07 16:34:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:01']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:26\n",
      "3 tasks running 2023-12-07 16:34:26\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:07']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:31\n",
      "3 tasks running 2023-12-07 16:34:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:13']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:37\n",
      "3 tasks running 2023-12-07 16:34:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:06:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:19']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:43\n",
      "3 tasks running 2023-12-07 16:34:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:24']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:48\n",
      "3 tasks running 2023-12-07 16:34:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:30']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:34:54\n",
      "3 tasks running 2023-12-07 16:34:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:35']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:00\n",
      "3 tasks running 2023-12-07 16:35:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:41']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:06\n",
      "3 tasks running 2023-12-07 16:35:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:47']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:12\n",
      "3 tasks running 2023-12-07 16:35:12\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:53']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:18\n",
      "3 tasks running 2023-12-07 16:35:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:07:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:07:59']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:24\n",
      "3 tasks running 2023-12-07 16:35:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:05']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:30\n",
      "3 tasks running 2023-12-07 16:35:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:11']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:35\n",
      "3 tasks running 2023-12-07 16:35:35\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:07:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:17']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:41\n",
      "3 tasks running 2023-12-07 16:35:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:22']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:46\n",
      "3 tasks running 2023-12-07 16:35:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:28']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:52\n",
      "3 tasks running 2023-12-07 16:35:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:34']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:35:58\n",
      "3 tasks running 2023-12-07 16:35:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:39']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:03\n",
      "3 tasks running 2023-12-07 16:36:03\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:45']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:09\n",
      "3 tasks running 2023-12-07 16:36:09\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:51']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:15\n",
      "3 tasks running 2023-12-07 16:36:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:08:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:08:57']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:21\n",
      "3 tasks running 2023-12-07 16:36:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:02']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:27\n",
      "3 tasks running 2023-12-07 16:36:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:08']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:33\n",
      "3 tasks running 2023-12-07 16:36:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:08:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:14']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:39\n",
      "3 tasks running 2023-12-07 16:36:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:20']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:44\n",
      "3 tasks running 2023-12-07 16:36:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:26']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:50\n",
      "3 tasks running 2023-12-07 16:36:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:31']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:36:56\n",
      "3 tasks running 2023-12-07 16:36:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:37']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:02\n",
      "3 tasks running 2023-12-07 16:37:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:43']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:08\n",
      "3 tasks running 2023-12-07 16:37:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:49']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:14\n",
      "3 tasks running 2023-12-07 16:37:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:09:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:09:55']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:20\n",
      "3 tasks running 2023-12-07 16:37:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:01']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:25\n",
      "3 tasks running 2023-12-07 16:37:25\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:07']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:31\n",
      "3 tasks running 2023-12-07 16:37:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:12']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:37\n",
      "3 tasks running 2023-12-07 16:37:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:09:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:18']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:42\n",
      "3 tasks running 2023-12-07 16:37:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:24']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:48\n",
      "3 tasks running 2023-12-07 16:37:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:30']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:37:54\n",
      "3 tasks running 2023-12-07 16:37:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1986', '0:10:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:35']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:38:00\n",
      "2 tasks running 2023-12-07 16:38:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:41']\n",
      "\n",
      "\n",
      "35 tasks ready 2023-12-07 16:38:05\n",
      "2 tasks running 2023-12-07 16:38:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:47']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:11\n",
      "3 tasks running 2023-12-07 16:38:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:52']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:17\n",
      "3 tasks running 2023-12-07 16:38:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:10:58']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:22\n",
      "3 tasks running 2023-12-07 16:38:22\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:04']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:28\n",
      "3 tasks running 2023-12-07 16:38:28\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:10']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:34\n",
      "3 tasks running 2023-12-07 16:38:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:10:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:16']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:40\n",
      "3 tasks running 2023-12-07 16:38:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:11:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:22']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:46\n",
      "3 tasks running 2023-12-07 16:38:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:11:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:28']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:52\n",
      "3 tasks running 2023-12-07 16:38:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:11:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:33']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:38:58\n",
      "3 tasks running 2023-12-07 16:38:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:11:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:40']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:39:04\n",
      "3 tasks running 2023-12-07 16:39:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:00:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:11:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:46']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:39:10\n",
      "3 tasks running 2023-12-07 16:39:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1987', '0:11:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:51']\n",
      "\n",
      "\n",
      "34 tasks ready 2023-12-07 16:39:15\n",
      "2 tasks running 2023-12-07 16:39:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:11:57']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:21\n",
      "3 tasks running 2023-12-07 16:39:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:03']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:27\n",
      "3 tasks running 2023-12-07 16:39:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:08']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:33\n",
      "3 tasks running 2023-12-07 16:39:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:14']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:39\n",
      "3 tasks running 2023-12-07 16:39:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:20']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:45\n",
      "3 tasks running 2023-12-07 16:39:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:26']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:51\n",
      "3 tasks running 2023-12-07 16:39:51\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:32']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:39:56\n",
      "3 tasks running 2023-12-07 16:39:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:38']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:02\n",
      "3 tasks running 2023-12-07 16:40:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:01:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:43']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:08\n",
      "3 tasks running 2023-12-07 16:40:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:49']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:14\n",
      "3 tasks running 2023-12-07 16:40:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:12:55']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:20\n",
      "3 tasks running 2023-12-07 16:40:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:00:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:01']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:26\n",
      "3 tasks running 2023-12-07 16:40:26\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:07']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:32\n",
      "3 tasks running 2023-12-07 16:40:32\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:13']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:38\n",
      "3 tasks running 2023-12-07 16:40:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:19']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:43\n",
      "3 tasks running 2023-12-07 16:40:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:25']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:49\n",
      "3 tasks running 2023-12-07 16:40:49\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:31']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:40:55\n",
      "3 tasks running 2023-12-07 16:40:55\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:37']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:01\n",
      "3 tasks running 2023-12-07 16:41:01\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:42']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:06\n",
      "3 tasks running 2023-12-07 16:41:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:02:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:48']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:12\n",
      "3 tasks running 2023-12-07 16:41:12\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:53']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:18\n",
      "3 tasks running 2023-12-07 16:41:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:01:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:13:59']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:24\n",
      "3 tasks running 2023-12-07 16:41:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:14:05']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:30\n",
      "3 tasks running 2023-12-07 16:41:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1985', '0:14:12']\n",
      "\n",
      "\n",
      "33 tasks ready 2023-12-07 16:41:36\n",
      "2 tasks running 2023-12-07 16:41:36\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:28']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:41:42\n",
      "3 tasks running 2023-12-07 16:41:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:34']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:41:48\n",
      "3 tasks running 2023-12-07 16:41:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:40']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:41:54\n",
      "3 tasks running 2023-12-07 16:41:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:45']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:41:59\n",
      "3 tasks running 2023-12-07 16:41:59\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:51']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:05\n",
      "3 tasks running 2023-12-07 16:42:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:03:57']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:11\n",
      "3 tasks running 2023-12-07 16:42:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:03']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:17\n",
      "3 tasks running 2023-12-07 16:42:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:02:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:09']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:23\n",
      "3 tasks running 2023-12-07 16:42:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:14']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:28\n",
      "3 tasks running 2023-12-07 16:42:28\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:20']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:34\n",
      "3 tasks running 2023-12-07 16:42:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:26']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:39\n",
      "3 tasks running 2023-12-07 16:42:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:00:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:31']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:45\n",
      "3 tasks running 2023-12-07 16:42:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:37']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:51\n",
      "3 tasks running 2023-12-07 16:42:51\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:42']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:42:56\n",
      "3 tasks running 2023-12-07 16:42:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:48']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:02\n",
      "3 tasks running 2023-12-07 16:43:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:53']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:08\n",
      "3 tasks running 2023-12-07 16:43:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:04:59']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:13\n",
      "3 tasks running 2023-12-07 16:43:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:05']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:19\n",
      "3 tasks running 2023-12-07 16:43:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:03:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:11']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:25\n",
      "3 tasks running 2023-12-07 16:43:25\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:17']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:31\n",
      "3 tasks running 2023-12-07 16:43:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:22']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:37\n",
      "3 tasks running 2023-12-07 16:43:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:01:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:28']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:43\n",
      "3 tasks running 2023-12-07 16:43:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:34']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:48\n",
      "3 tasks running 2023-12-07 16:43:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:40']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:43:54\n",
      "3 tasks running 2023-12-07 16:43:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:46']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:00\n",
      "3 tasks running 2023-12-07 16:44:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:51']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:05\n",
      "3 tasks running 2023-12-07 16:44:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:05:57']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:11\n",
      "3 tasks running 2023-12-07 16:44:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:03']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:17\n",
      "3 tasks running 2023-12-07 16:44:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:04:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:09']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:23\n",
      "3 tasks running 2023-12-07 16:44:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:14']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:28\n",
      "3 tasks running 2023-12-07 16:44:28\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:20']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:34\n",
      "3 tasks running 2023-12-07 16:44:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:02:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:26']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:40\n",
      "3 tasks running 2023-12-07 16:44:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:32']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:46\n",
      "3 tasks running 2023-12-07 16:44:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:38']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:51\n",
      "3 tasks running 2023-12-07 16:44:51\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:43']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:44:57\n",
      "3 tasks running 2023-12-07 16:44:57\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:49']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:03\n",
      "3 tasks running 2023-12-07 16:45:03\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:06:55']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:09\n",
      "3 tasks running 2023-12-07 16:45:09\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:01']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:15\n",
      "3 tasks running 2023-12-07 16:45:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:05:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:07']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:21\n",
      "3 tasks running 2023-12-07 16:45:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:13']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:27\n",
      "3 tasks running 2023-12-07 16:45:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:18']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:32\n",
      "3 tasks running 2023-12-07 16:45:32\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:24']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:38\n",
      "3 tasks running 2023-12-07 16:45:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:03:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:30']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:44\n",
      "3 tasks running 2023-12-07 16:45:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:36']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:50\n",
      "3 tasks running 2023-12-07 16:45:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:42']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:45:56\n",
      "3 tasks running 2023-12-07 16:45:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:48']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:02\n",
      "3 tasks running 2023-12-07 16:46:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:54']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:08\n",
      "3 tasks running 2023-12-07 16:46:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:07:59']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:13\n",
      "3 tasks running 2023-12-07 16:46:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:05']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:19\n",
      "3 tasks running 2023-12-07 16:46:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:06:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:11']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:25\n",
      "3 tasks running 2023-12-07 16:46:25\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:16']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:31\n",
      "3 tasks running 2023-12-07 16:46:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:22']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:36\n",
      "3 tasks running 2023-12-07 16:46:36\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:04:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:28']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:42\n",
      "3 tasks running 2023-12-07 16:46:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:34']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:48\n",
      "3 tasks running 2023-12-07 16:46:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:40']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:46:54\n",
      "3 tasks running 2023-12-07 16:46:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:46']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:00\n",
      "3 tasks running 2023-12-07 16:47:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:52']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:05\n",
      "3 tasks running 2023-12-07 16:47:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:08:57']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:11\n",
      "3 tasks running 2023-12-07 16:47:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:03']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:17\n",
      "3 tasks running 2023-12-07 16:47:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:07:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:09']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:23\n",
      "3 tasks running 2023-12-07 16:47:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:15']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:29\n",
      "3 tasks running 2023-12-07 16:47:29\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:21']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:35\n",
      "3 tasks running 2023-12-07 16:47:35\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:05:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:26']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:41\n",
      "3 tasks running 2023-12-07 16:47:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:32']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:46\n",
      "3 tasks running 2023-12-07 16:47:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:38']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:52\n",
      "3 tasks running 2023-12-07 16:47:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:44']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:47:58\n",
      "3 tasks running 2023-12-07 16:47:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:50']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:04\n",
      "3 tasks running 2023-12-07 16:48:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:09:56']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:10\n",
      "3 tasks running 2023-12-07 16:48:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:01']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:16\n",
      "3 tasks running 2023-12-07 16:48:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:08:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:07']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:21\n",
      "3 tasks running 2023-12-07 16:48:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:13']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:27\n",
      "3 tasks running 2023-12-07 16:48:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:19']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:32\n",
      "3 tasks running 2023-12-07 16:48:32\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:24']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:38\n",
      "3 tasks running 2023-12-07 16:48:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:06:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:30']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:44\n",
      "3 tasks running 2023-12-07 16:48:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:36']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:50\n",
      "3 tasks running 2023-12-07 16:48:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:42']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:48:56\n",
      "3 tasks running 2023-12-07 16:48:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:48']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:02\n",
      "3 tasks running 2023-12-07 16:49:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:10:54']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:08\n",
      "3 tasks running 2023-12-07 16:49:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:11:00']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:14\n",
      "3 tasks running 2023-12-07 16:49:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:11:06']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:19\n",
      "3 tasks running 2023-12-07 16:49:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:09:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:11:11']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:25\n",
      "3 tasks running 2023-12-07 16:49:25\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1988', '0:11:17']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:31\n",
      "2 tasks running 2023-12-07 16:49:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:10']\n",
      "\n",
      "\n",
      "32 tasks ready 2023-12-07 16:49:37\n",
      "2 tasks running 2023-12-07 16:49:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:07:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:16']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:49:43\n",
      "3 tasks running 2023-12-07 16:49:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:21']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:49:48\n",
      "3 tasks running 2023-12-07 16:49:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:27']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:49:54\n",
      "3 tasks running 2023-12-07 16:49:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:33']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:00\n",
      "3 tasks running 2023-12-07 16:50:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:38']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:06\n",
      "3 tasks running 2023-12-07 16:50:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:44']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:11\n",
      "3 tasks running 2023-12-07 16:50:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:50']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:17\n",
      "3 tasks running 2023-12-07 16:50:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:10:56']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:23\n",
      "3 tasks running 2023-12-07 16:50:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:02']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:29\n",
      "3 tasks running 2023-12-07 16:50:29\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:08']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:34\n",
      "3 tasks running 2023-12-07 16:50:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:08:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:13']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:40\n",
      "3 tasks running 2023-12-07 16:50:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:00:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:19']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:46\n",
      "3 tasks running 2023-12-07 16:50:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:24']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:52\n",
      "3 tasks running 2023-12-07 16:50:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:30']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:50:58\n",
      "3 tasks running 2023-12-07 16:50:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:36']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:04\n",
      "3 tasks running 2023-12-07 16:51:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:42']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:09\n",
      "3 tasks running 2023-12-07 16:51:09\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:48']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:15\n",
      "3 tasks running 2023-12-07 16:51:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:53']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:21\n",
      "3 tasks running 2023-12-07 16:51:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:11:59']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:27\n",
      "3 tasks running 2023-12-07 16:51:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:05']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:33\n",
      "3 tasks running 2023-12-07 16:51:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:11']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:39\n",
      "3 tasks running 2023-12-07 16:51:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:01:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:09:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:17']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:45\n",
      "3 tasks running 2023-12-07 16:51:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:10:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:23']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:50\n",
      "3 tasks running 2023-12-07 16:51:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:10:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:29']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:51:56\n",
      "3 tasks running 2023-12-07 16:51:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:10:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:34']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:52:01\n",
      "3 tasks running 2023-12-07 16:52:01\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:10:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:40']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:52:07\n",
      "3 tasks running 2023-12-07 16:52:07\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:10:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:46']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:52:13\n",
      "3 tasks running 2023-12-07 16:52:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1990', '0:10:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:52']\n",
      "\n",
      "\n",
      "31 tasks ready 2023-12-07 16:52:19\n",
      "2 tasks running 2023-12-07 16:52:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:12:58']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:24\n",
      "3 tasks running 2023-12-07 16:52:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:03']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:30\n",
      "3 tasks running 2023-12-07 16:52:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:09']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:36\n",
      "3 tasks running 2023-12-07 16:52:36\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:02:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:14']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:41\n",
      "3 tasks running 2023-12-07 16:52:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:20']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:47\n",
      "3 tasks running 2023-12-07 16:52:47\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:26']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:53\n",
      "3 tasks running 2023-12-07 16:52:53\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:31']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:52:58\n",
      "3 tasks running 2023-12-07 16:52:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:37']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:04\n",
      "3 tasks running 2023-12-07 16:53:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:42']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:10\n",
      "3 tasks running 2023-12-07 16:53:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:48']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:15\n",
      "3 tasks running 2023-12-07 16:53:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:00:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:13:54']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:21\n",
      "3 tasks running 2023-12-07 16:53:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:00']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:27\n",
      "3 tasks running 2023-12-07 16:53:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:05']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:32\n",
      "3 tasks running 2023-12-07 16:53:32\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:11']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:38\n",
      "3 tasks running 2023-12-07 16:53:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:03:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:17']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:44\n",
      "3 tasks running 2023-12-07 16:53:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:23']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:49\n",
      "3 tasks running 2023-12-07 16:53:49\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:28']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:53:55\n",
      "3 tasks running 2023-12-07 16:53:55\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:34']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:01\n",
      "3 tasks running 2023-12-07 16:54:01\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:40']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:07\n",
      "3 tasks running 2023-12-07 16:54:07\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:45']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:13\n",
      "3 tasks running 2023-12-07 16:54:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:51']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:18\n",
      "3 tasks running 2023-12-07 16:54:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:01:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:14:57']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:24\n",
      "3 tasks running 2023-12-07 16:54:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:03']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:30\n",
      "3 tasks running 2023-12-07 16:54:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:08']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:35\n",
      "3 tasks running 2023-12-07 16:54:35\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:04:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:14']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:41\n",
      "3 tasks running 2023-12-07 16:54:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:20']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:46\n",
      "3 tasks running 2023-12-07 16:54:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:25']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:52\n",
      "3 tasks running 2023-12-07 16:54:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:31']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:54:58\n",
      "3 tasks running 2023-12-07 16:54:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:37']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:04\n",
      "3 tasks running 2023-12-07 16:55:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:43']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:10\n",
      "3 tasks running 2023-12-07 16:55:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:49']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:16\n",
      "3 tasks running 2023-12-07 16:55:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:02:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:15:55']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:21\n",
      "3 tasks running 2023-12-07 16:55:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:00']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:27\n",
      "3 tasks running 2023-12-07 16:55:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:06']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:33\n",
      "3 tasks running 2023-12-07 16:55:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:12']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:39\n",
      "3 tasks running 2023-12-07 16:55:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:05:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:17']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:45\n",
      "3 tasks running 2023-12-07 16:55:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:23']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:50\n",
      "3 tasks running 2023-12-07 16:55:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:29']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:55:56\n",
      "3 tasks running 2023-12-07 16:55:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:35']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:02\n",
      "3 tasks running 2023-12-07 16:56:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:40']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:07\n",
      "3 tasks running 2023-12-07 16:56:07\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:46']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:13\n",
      "3 tasks running 2023-12-07 16:56:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:52']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:19\n",
      "3 tasks running 2023-12-07 16:56:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:03:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:16:57']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:24\n",
      "3 tasks running 2023-12-07 16:56:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:03']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:30\n",
      "3 tasks running 2023-12-07 16:56:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:09']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:36\n",
      "3 tasks running 2023-12-07 16:56:36\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:06:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:15']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:42\n",
      "3 tasks running 2023-12-07 16:56:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:20']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:48\n",
      "3 tasks running 2023-12-07 16:56:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:27']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:53\n",
      "3 tasks running 2023-12-07 16:56:53\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:32']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:56:59\n",
      "3 tasks running 2023-12-07 16:56:59\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:38']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:57:05\n",
      "3 tasks running 2023-12-07 16:57:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:43']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:57:10\n",
      "3 tasks running 2023-12-07 16:57:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:49']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:57:16\n",
      "3 tasks running 2023-12-07 16:57:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:04:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1989', '0:17:55']\n",
      "\n",
      "\n",
      "30 tasks ready 2023-12-07 16:57:22\n",
      "2 tasks running 2023-12-07 16:57:22\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:41']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:57:28\n",
      "3 tasks running 2023-12-07 16:57:28\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:47']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:57:33\n",
      "3 tasks running 2023-12-07 16:57:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:53']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:57:39\n",
      "3 tasks running 2023-12-07 16:57:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:07:59']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:57:45\n",
      "3 tasks running 2023-12-07 16:57:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:04']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:57:51\n",
      "3 tasks running 2023-12-07 16:57:51\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:10']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:57:56\n",
      "3 tasks running 2023-12-07 16:57:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:16']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:02\n",
      "3 tasks running 2023-12-07 16:58:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:22']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:08\n",
      "3 tasks running 2023-12-07 16:58:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:28']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:14\n",
      "3 tasks running 2023-12-07 16:58:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:34']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:20\n",
      "3 tasks running 2023-12-07 16:58:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:00:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:05:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:39']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:26\n",
      "3 tasks running 2023-12-07 16:58:26\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:45']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:31\n",
      "3 tasks running 2023-12-07 16:58:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:51']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:37\n",
      "3 tasks running 2023-12-07 16:58:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:08:57']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:43\n",
      "3 tasks running 2023-12-07 16:58:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:02']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:49\n",
      "3 tasks running 2023-12-07 16:58:49\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:08']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:58:54\n",
      "3 tasks running 2023-12-07 16:58:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:14']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:00\n",
      "3 tasks running 2023-12-07 16:59:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:19']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:06\n",
      "3 tasks running 2023-12-07 16:59:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:25']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:11\n",
      "3 tasks running 2023-12-07 16:59:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:31']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:17\n",
      "3 tasks running 2023-12-07 16:59:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:06:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:36']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:23\n",
      "3 tasks running 2023-12-07 16:59:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:01:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:42']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:29\n",
      "3 tasks running 2023-12-07 16:59:29\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:48']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:34\n",
      "3 tasks running 2023-12-07 16:59:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:54']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:40\n",
      "3 tasks running 2023-12-07 16:59:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:09:59']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:45\n",
      "3 tasks running 2023-12-07 16:59:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:05']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:51\n",
      "3 tasks running 2023-12-07 16:59:51\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:10']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 16:59:57\n",
      "3 tasks running 2023-12-07 16:59:57\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:16']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:02\n",
      "3 tasks running 2023-12-07 17:00:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:22']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:08\n",
      "3 tasks running 2023-12-07 17:00:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:27']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:14\n",
      "3 tasks running 2023-12-07 17:00:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:33']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:20\n",
      "3 tasks running 2023-12-07 17:00:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:02:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:07:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:39']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:25\n",
      "3 tasks running 2023-12-07 17:00:25\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:45']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:31\n",
      "3 tasks running 2023-12-07 17:00:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:50']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:37\n",
      "3 tasks running 2023-12-07 17:00:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:10:56']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:43\n",
      "3 tasks running 2023-12-07 17:00:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:02']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:49\n",
      "3 tasks running 2023-12-07 17:00:49\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:08']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:00:54\n",
      "3 tasks running 2023-12-07 17:00:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:14']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:00\n",
      "3 tasks running 2023-12-07 17:01:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:20']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:06\n",
      "3 tasks running 2023-12-07 17:01:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:25']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:12\n",
      "3 tasks running 2023-12-07 17:01:12\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:31']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:18\n",
      "3 tasks running 2023-12-07 17:01:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:03:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:08:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:37']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:24\n",
      "3 tasks running 2023-12-07 17:01:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:43']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:29\n",
      "3 tasks running 2023-12-07 17:01:29\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:49']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:35\n",
      "3 tasks running 2023-12-07 17:01:35\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:11:55']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:41\n",
      "3 tasks running 2023-12-07 17:01:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:00']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:47\n",
      "3 tasks running 2023-12-07 17:01:47\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:06']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:53\n",
      "3 tasks running 2023-12-07 17:01:53\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:12']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:01:58\n",
      "3 tasks running 2023-12-07 17:01:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:18']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:04\n",
      "3 tasks running 2023-12-07 17:02:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:24']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:10\n",
      "3 tasks running 2023-12-07 17:02:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:29']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:16\n",
      "3 tasks running 2023-12-07 17:02:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:09:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:35']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:22\n",
      "3 tasks running 2023-12-07 17:02:22\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:04:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:41']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:27\n",
      "3 tasks running 2023-12-07 17:02:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:47']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:33\n",
      "3 tasks running 2023-12-07 17:02:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:52']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:39\n",
      "3 tasks running 2023-12-07 17:02:39\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:12:58']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:45\n",
      "3 tasks running 2023-12-07 17:02:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:04']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:50\n",
      "3 tasks running 2023-12-07 17:02:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:10']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:02:56\n",
      "3 tasks running 2023-12-07 17:02:56\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:16']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:02\n",
      "3 tasks running 2023-12-07 17:03:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:22']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:08\n",
      "3 tasks running 2023-12-07 17:03:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:27']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:13\n",
      "3 tasks running 2023-12-07 17:03:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:33']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:19\n",
      "3 tasks running 2023-12-07 17:03:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:05:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:10:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:38']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:24\n",
      "3 tasks running 2023-12-07 17:03:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:44']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:30\n",
      "3 tasks running 2023-12-07 17:03:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:50']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:36\n",
      "3 tasks running 2023-12-07 17:03:36\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:13:55']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:42\n",
      "3 tasks running 2023-12-07 17:03:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:01']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:48\n",
      "3 tasks running 2023-12-07 17:03:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:07']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:54\n",
      "3 tasks running 2023-12-07 17:03:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:13']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:03:59\n",
      "3 tasks running 2023-12-07 17:03:59\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:19']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:05\n",
      "3 tasks running 2023-12-07 17:04:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:24']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:11\n",
      "3 tasks running 2023-12-07 17:04:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:30']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:16\n",
      "3 tasks running 2023-12-07 17:04:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:11:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:36']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:22\n",
      "3 tasks running 2023-12-07 17:04:22\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:06:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:42']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:28\n",
      "3 tasks running 2023-12-07 17:04:28\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:47']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:34\n",
      "3 tasks running 2023-12-07 17:04:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:53']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:40\n",
      "3 tasks running 2023-12-07 17:04:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:16']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:14:59']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:45\n",
      "3 tasks running 2023-12-07 17:04:45\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:15:05']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:51\n",
      "3 tasks running 2023-12-07 17:04:51\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:15:10']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:04:57\n",
      "3 tasks running 2023-12-07 17:04:57\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:15:16']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:05:02\n",
      "3 tasks running 2023-12-07 17:05:02\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:15:22']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:05:08\n",
      "3 tasks running 2023-12-07 17:05:08\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:15:27']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:05:14\n",
      "3 tasks running 2023-12-07 17:05:14\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1991', '0:15:33']\n",
      "\n",
      "\n",
      "29 tasks ready 2023-12-07 17:05:20\n",
      "2 tasks running 2023-12-07 17:05:20\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:07:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:12:59']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:05:26\n",
      "3 tasks running 2023-12-07 17:05:26\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:05']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:05:31\n",
      "3 tasks running 2023-12-07 17:05:31\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:10']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:05:37\n",
      "3 tasks running 2023-12-07 17:05:37\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:16']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:05:43\n",
      "3 tasks running 2023-12-07 17:05:43\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:22']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:05:48\n",
      "3 tasks running 2023-12-07 17:05:48\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:28']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:05:54\n",
      "3 tasks running 2023-12-07 17:05:54\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:33']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:06:00\n",
      "3 tasks running 2023-12-07 17:06:00\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:39']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:06:06\n",
      "3 tasks running 2023-12-07 17:06:06\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:45']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:06:12\n",
      "3 tasks running 2023-12-07 17:06:12\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:51']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:06:18\n",
      "3 tasks running 2023-12-07 17:06:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:00:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:13:57']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:06:23\n",
      "3 tasks running 2023-12-07 17:06:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:08:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1992', '0:14:02']\n",
      "\n",
      "\n",
      "28 tasks ready 2023-12-07 17:06:29\n",
      "2 tasks running 2023-12-07 17:06:29\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:05']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:06:34\n",
      "3 tasks running 2023-12-07 17:06:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:10']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:06:40\n",
      "3 tasks running 2023-12-07 17:06:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:16']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:06:46\n",
      "3 tasks running 2023-12-07 17:06:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:22']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:06:52\n",
      "3 tasks running 2023-12-07 17:06:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:28']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:06:58\n",
      "3 tasks running 2023-12-07 17:06:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:34']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:04\n",
      "3 tasks running 2023-12-07 17:07:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:40']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:10\n",
      "3 tasks running 2023-12-07 17:07:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:46']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:16\n",
      "3 tasks running 2023-12-07 17:07:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:53']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:52']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:21\n",
      "3 tasks running 2023-12-07 17:07:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:01:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:09:57']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:27\n",
      "3 tasks running 2023-12-07 17:07:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:00:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:03']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:33\n",
      "3 tasks running 2023-12-07 17:07:33\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:10']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:09']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:38\n",
      "3 tasks running 2023-12-07 17:07:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:14']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:44\n",
      "3 tasks running 2023-12-07 17:07:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:20']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:50\n",
      "3 tasks running 2023-12-07 17:07:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:26']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:07:55\n",
      "3 tasks running 2023-12-07 17:07:55\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:33']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:31']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:01\n",
      "3 tasks running 2023-12-07 17:08:01\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:37']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:07\n",
      "3 tasks running 2023-12-07 17:08:07\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:43']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:13\n",
      "3 tasks running 2023-12-07 17:08:13\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:40']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:49']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:19\n",
      "3 tasks running 2023-12-07 17:08:19\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:46']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:02:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:10:55']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:24\n",
      "3 tasks running 2023-12-07 17:08:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:00']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:30\n",
      "3 tasks running 2023-12-07 17:08:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:01:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:06']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:36\n",
      "3 tasks running 2023-12-07 17:08:36\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:03']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:12']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:42\n",
      "3 tasks running 2023-12-07 17:08:42\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:18']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:47\n",
      "3 tasks running 2023-12-07 17:08:47\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:24']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:53\n",
      "3 tasks running 2023-12-07 17:08:53\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:29']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:08:59\n",
      "3 tasks running 2023-12-07 17:08:59\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:35']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:09:05\n",
      "3 tasks running 2023-12-07 17:09:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:41']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:09:11\n",
      "3 tasks running 2023-12-07 17:09:11\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:47']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:09:17\n",
      "3 tasks running 2023-12-07 17:09:17\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:44']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:03:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:53']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:09:23\n",
      "3 tasks running 2023-12-07 17:09:23\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:50']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1993', '0:11:59']\n",
      "\n",
      "\n",
      "27 tasks ready 2023-12-07 17:09:29\n",
      "2 tasks running 2023-12-07 17:09:29\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:02:56']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:06']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:09:34\n",
      "3 tasks running 2023-12-07 17:09:34\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:00']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:12']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:09:40\n",
      "3 tasks running 2023-12-07 17:09:40\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:06']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:18']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:09:46\n",
      "3 tasks running 2023-12-07 17:09:46\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:12']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:24']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:09:52\n",
      "3 tasks running 2023-12-07 17:09:52\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:19']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:29']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:09:58\n",
      "3 tasks running 2023-12-07 17:09:58\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:23']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:25']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:35']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:04\n",
      "3 tasks running 2023-12-07 17:10:04\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:29']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:31']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:41']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:09\n",
      "3 tasks running 2023-12-07 17:10:09\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:35']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:47']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:15\n",
      "3 tasks running 2023-12-07 17:10:15\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:41']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:53']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:21\n",
      "3 tasks running 2023-12-07 17:10:21\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:47']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:48']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:04:58']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:27\n",
      "3 tasks running 2023-12-07 17:10:27\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:52']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:54']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:04']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:32\n",
      "3 tasks running 2023-12-07 17:10:32\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:00:58']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:03:59']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:10']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:38\n",
      "3 tasks running 2023-12-07 17:10:38\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:04']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:05']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:15']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:44\n",
      "3 tasks running 2023-12-07 17:10:44\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:09']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:11']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:21']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:50\n",
      "3 tasks running 2023-12-07 17:10:50\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:15']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:17']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:27']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:10:55\n",
      "3 tasks running 2023-12-07 17:10:55\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:21']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:22']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:33']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:01\n",
      "3 tasks running 2023-12-07 17:11:01\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:27']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:28']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:38']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:07\n",
      "3 tasks running 2023-12-07 17:11:07\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:34']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:44']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:12\n",
      "3 tasks running 2023-12-07 17:11:12\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:38']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:39']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:50']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:18\n",
      "3 tasks running 2023-12-07 17:11:18\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:45']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:05:55']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:24\n",
      "3 tasks running 2023-12-07 17:11:24\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:49']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:51']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:01']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:30\n",
      "3 tasks running 2023-12-07 17:11:30\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:01:55']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:04:57']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:07']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:35\n",
      "3 tasks running 2023-12-07 17:11:35\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:01']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:02']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:13']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:41\n",
      "3 tasks running 2023-12-07 17:11:41\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:07']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:08']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:19']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:47\n",
      "3 tasks running 2023-12-07 17:11:47\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:13']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:14']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:24']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:53\n",
      "3 tasks running 2023-12-07 17:11:53\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:18']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:20']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:30']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:11:59\n",
      "3 tasks running 2023-12-07 17:11:59\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:24']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:26']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:36']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:12:05\n",
      "3 tasks running 2023-12-07 17:12:05\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:30']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:32']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:42']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:12:10\n",
      "3 tasks running 2023-12-07 17:12:10\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:36']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:37']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:48']\n",
      "\n",
      "\n",
      "26 tasks ready 2023-12-07 17:12:16\n",
      "3 tasks running 2023-12-07 17:12:16\n",
      "Running names:\n",
      "['LCMS_PRUSVI_v2022-Training_1996', '0:02:42']\n",
      "['LCMS_PRUSVI_v2022-Training_1995', '0:05:43']\n",
      "['LCMS_PRUSVI_v2022-Training_1994', '0:06:54']\n",
      "\n",
      "\n"
     ]
    },
    {
     "ename": "SSLError",
     "evalue": "HTTPSConnectionPool(host='earthengine.googleapis.com', port=443): Max retries exceeded with url: /v1/projects/rcr-gee/operations?pageSize=500&alt=json (Caused by SSLError(SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1002)')))",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mSSLEOFError\u001b[0m                               Traceback (most recent call last)",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\connectionpool.py:714\u001b[0m, in \u001b[0;36mHTTPConnectionPool.urlopen\u001b[1;34m(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, **response_kw)\u001b[0m\n\u001b[0;32m    713\u001b[0m \u001b[39m# Make the request on the httplib connection object.\u001b[39;00m\n\u001b[1;32m--> 714\u001b[0m httplib_response \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_make_request(\n\u001b[0;32m    715\u001b[0m     conn,\n\u001b[0;32m    716\u001b[0m     method,\n\u001b[0;32m    717\u001b[0m     url,\n\u001b[0;32m    718\u001b[0m     timeout\u001b[39m=\u001b[39;49mtimeout_obj,\n\u001b[0;32m    719\u001b[0m     body\u001b[39m=\u001b[39;49mbody,\n\u001b[0;32m    720\u001b[0m     headers\u001b[39m=\u001b[39;49mheaders,\n\u001b[0;32m    721\u001b[0m     chunked\u001b[39m=\u001b[39;49mchunked,\n\u001b[0;32m    722\u001b[0m )\n\u001b[0;32m    724\u001b[0m \u001b[39m# If we're going to release the connection in ``finally:``, then\u001b[39;00m\n\u001b[0;32m    725\u001b[0m \u001b[39m# the response doesn't need to know about the connection. Otherwise\u001b[39;00m\n\u001b[0;32m    726\u001b[0m \u001b[39m# it will also try to release it and we'll have a double-release\u001b[39;00m\n\u001b[0;32m    727\u001b[0m \u001b[39m# mess.\u001b[39;00m\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\connectionpool.py:403\u001b[0m, in \u001b[0;36mHTTPConnectionPool._make_request\u001b[1;34m(self, conn, method, url, timeout, chunked, **httplib_request_kw)\u001b[0m\n\u001b[0;32m    402\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[1;32m--> 403\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_validate_conn(conn)\n\u001b[0;32m    404\u001b[0m \u001b[39mexcept\u001b[39;00m (SocketTimeout, BaseSSLError) \u001b[39mas\u001b[39;00m e:\n\u001b[0;32m    405\u001b[0m     \u001b[39m# Py2 raises this as a BaseSSLError, Py3 raises it as socket timeout.\u001b[39;00m\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\connectionpool.py:1053\u001b[0m, in \u001b[0;36mHTTPSConnectionPool._validate_conn\u001b[1;34m(self, conn)\u001b[0m\n\u001b[0;32m   1052\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mgetattr\u001b[39m(conn, \u001b[39m\"\u001b[39m\u001b[39msock\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39mNone\u001b[39;00m):  \u001b[39m# AppEngine might not have  `.sock`\u001b[39;00m\n\u001b[1;32m-> 1053\u001b[0m     conn\u001b[39m.\u001b[39;49mconnect()\n\u001b[0;32m   1055\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m conn\u001b[39m.\u001b[39mis_verified:\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\connection.py:419\u001b[0m, in \u001b[0;36mHTTPSConnection.connect\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    417\u001b[0m     context\u001b[39m.\u001b[39mload_default_certs()\n\u001b[1;32m--> 419\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39msock \u001b[39m=\u001b[39m ssl_wrap_socket(\n\u001b[0;32m    420\u001b[0m     sock\u001b[39m=\u001b[39;49mconn,\n\u001b[0;32m    421\u001b[0m     keyfile\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mkey_file,\n\u001b[0;32m    422\u001b[0m     certfile\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mcert_file,\n\u001b[0;32m    423\u001b[0m     key_password\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mkey_password,\n\u001b[0;32m    424\u001b[0m     ca_certs\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mca_certs,\n\u001b[0;32m    425\u001b[0m     ca_cert_dir\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mca_cert_dir,\n\u001b[0;32m    426\u001b[0m     ca_cert_data\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mca_cert_data,\n\u001b[0;32m    427\u001b[0m     server_hostname\u001b[39m=\u001b[39;49mserver_hostname,\n\u001b[0;32m    428\u001b[0m     ssl_context\u001b[39m=\u001b[39;49mcontext,\n\u001b[0;32m    429\u001b[0m     tls_in_tls\u001b[39m=\u001b[39;49mtls_in_tls,\n\u001b[0;32m    430\u001b[0m )\n\u001b[0;32m    432\u001b[0m \u001b[39m# If we're using all defaults and the connection\u001b[39;00m\n\u001b[0;32m    433\u001b[0m \u001b[39m# is TLSv1 or TLSv1.1 we throw a DeprecationWarning\u001b[39;00m\n\u001b[0;32m    434\u001b[0m \u001b[39m# for the host.\u001b[39;00m\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\util\\ssl_.py:449\u001b[0m, in \u001b[0;36mssl_wrap_socket\u001b[1;34m(sock, keyfile, certfile, cert_reqs, ca_certs, server_hostname, ssl_version, ciphers, ssl_context, ca_cert_dir, key_password, ca_cert_data, tls_in_tls)\u001b[0m\n\u001b[0;32m    448\u001b[0m \u001b[39mif\u001b[39;00m send_sni:\n\u001b[1;32m--> 449\u001b[0m     ssl_sock \u001b[39m=\u001b[39m _ssl_wrap_socket_impl(\n\u001b[0;32m    450\u001b[0m         sock, context, tls_in_tls, server_hostname\u001b[39m=\u001b[39;49mserver_hostname\n\u001b[0;32m    451\u001b[0m     )\n\u001b[0;32m    452\u001b[0m \u001b[39melse\u001b[39;00m:\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\util\\ssl_.py:493\u001b[0m, in \u001b[0;36m_ssl_wrap_socket_impl\u001b[1;34m(sock, ssl_context, tls_in_tls, server_hostname)\u001b[0m\n\u001b[0;32m    492\u001b[0m \u001b[39mif\u001b[39;00m server_hostname:\n\u001b[1;32m--> 493\u001b[0m     \u001b[39mreturn\u001b[39;00m ssl_context\u001b[39m.\u001b[39;49mwrap_socket(sock, server_hostname\u001b[39m=\u001b[39;49mserver_hostname)\n\u001b[0;32m    494\u001b[0m \u001b[39melse\u001b[39;00m:\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\ssl.py:517\u001b[0m, in \u001b[0;36mSSLContext.wrap_socket\u001b[1;34m(self, sock, server_side, do_handshake_on_connect, suppress_ragged_eofs, server_hostname, session)\u001b[0m\n\u001b[0;32m    511\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mwrap_socket\u001b[39m(\u001b[39mself\u001b[39m, sock, server_side\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m,\n\u001b[0;32m    512\u001b[0m                 do_handshake_on_connect\u001b[39m=\u001b[39m\u001b[39mTrue\u001b[39;00m,\n\u001b[0;32m    513\u001b[0m                 suppress_ragged_eofs\u001b[39m=\u001b[39m\u001b[39mTrue\u001b[39;00m,\n\u001b[0;32m    514\u001b[0m                 server_hostname\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, session\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m):\n\u001b[0;32m    515\u001b[0m     \u001b[39m# SSLSocket class handles server_hostname encoding before it calls\u001b[39;00m\n\u001b[0;32m    516\u001b[0m     \u001b[39m# ctx._wrap_socket()\u001b[39;00m\n\u001b[1;32m--> 517\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49msslsocket_class\u001b[39m.\u001b[39;49m_create(\n\u001b[0;32m    518\u001b[0m         sock\u001b[39m=\u001b[39;49msock,\n\u001b[0;32m    519\u001b[0m         server_side\u001b[39m=\u001b[39;49mserver_side,\n\u001b[0;32m    520\u001b[0m         do_handshake_on_connect\u001b[39m=\u001b[39;49mdo_handshake_on_connect,\n\u001b[0;32m    521\u001b[0m         suppress_ragged_eofs\u001b[39m=\u001b[39;49msuppress_ragged_eofs,\n\u001b[0;32m    522\u001b[0m         server_hostname\u001b[39m=\u001b[39;49mserver_hostname,\n\u001b[0;32m    523\u001b[0m         context\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m,\n\u001b[0;32m    524\u001b[0m         session\u001b[39m=\u001b[39;49msession\n\u001b[0;32m    525\u001b[0m     )\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\ssl.py:1075\u001b[0m, in \u001b[0;36mSSLSocket._create\u001b[1;34m(cls, sock, server_side, do_handshake_on_connect, suppress_ragged_eofs, server_hostname, context, session)\u001b[0m\n\u001b[0;32m   1074\u001b[0m             \u001b[39mraise\u001b[39;00m \u001b[39mValueError\u001b[39;00m(\u001b[39m\"\u001b[39m\u001b[39mdo_handshake_on_connect should not be specified for non-blocking sockets\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[1;32m-> 1075\u001b[0m         \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdo_handshake()\n\u001b[0;32m   1076\u001b[0m \u001b[39mexcept\u001b[39;00m (\u001b[39mOSError\u001b[39;00m, \u001b[39mValueError\u001b[39;00m):\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\ssl.py:1346\u001b[0m, in \u001b[0;36mSSLSocket.do_handshake\u001b[1;34m(self, block)\u001b[0m\n\u001b[0;32m   1345\u001b[0m         \u001b[39mself\u001b[39m\u001b[39m.\u001b[39msettimeout(\u001b[39mNone\u001b[39;00m)\n\u001b[1;32m-> 1346\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_sslobj\u001b[39m.\u001b[39mdo_handshake()\n\u001b[0;32m   1347\u001b[0m \u001b[39mfinally\u001b[39;00m:\n",
      "\u001b[1;31mSSLEOFError\u001b[0m: EOF occurred in violation of protocol (_ssl.c:1002)",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mMaxRetryError\u001b[0m                             Traceback (most recent call last)",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\requests\\adapters.py:486\u001b[0m, in \u001b[0;36mHTTPAdapter.send\u001b[1;34m(self, request, stream, timeout, verify, cert, proxies)\u001b[0m\n\u001b[0;32m    485\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[1;32m--> 486\u001b[0m     resp \u001b[39m=\u001b[39m conn\u001b[39m.\u001b[39;49murlopen(\n\u001b[0;32m    487\u001b[0m         method\u001b[39m=\u001b[39;49mrequest\u001b[39m.\u001b[39;49mmethod,\n\u001b[0;32m    488\u001b[0m         url\u001b[39m=\u001b[39;49murl,\n\u001b[0;32m    489\u001b[0m         body\u001b[39m=\u001b[39;49mrequest\u001b[39m.\u001b[39;49mbody,\n\u001b[0;32m    490\u001b[0m         headers\u001b[39m=\u001b[39;49mrequest\u001b[39m.\u001b[39;49mheaders,\n\u001b[0;32m    491\u001b[0m         redirect\u001b[39m=\u001b[39;49m\u001b[39mFalse\u001b[39;49;00m,\n\u001b[0;32m    492\u001b[0m         assert_same_host\u001b[39m=\u001b[39;49m\u001b[39mFalse\u001b[39;49;00m,\n\u001b[0;32m    493\u001b[0m         preload_content\u001b[39m=\u001b[39;49m\u001b[39mFalse\u001b[39;49;00m,\n\u001b[0;32m    494\u001b[0m         decode_content\u001b[39m=\u001b[39;49m\u001b[39mFalse\u001b[39;49;00m,\n\u001b[0;32m    495\u001b[0m         retries\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mmax_retries,\n\u001b[0;32m    496\u001b[0m         timeout\u001b[39m=\u001b[39;49mtimeout,\n\u001b[0;32m    497\u001b[0m         chunked\u001b[39m=\u001b[39;49mchunked,\n\u001b[0;32m    498\u001b[0m     )\n\u001b[0;32m    500\u001b[0m \u001b[39mexcept\u001b[39;00m (ProtocolError, \u001b[39mOSError\u001b[39;00m) \u001b[39mas\u001b[39;00m err:\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\connectionpool.py:798\u001b[0m, in \u001b[0;36mHTTPConnectionPool.urlopen\u001b[1;34m(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, **response_kw)\u001b[0m\n\u001b[0;32m    796\u001b[0m     e \u001b[39m=\u001b[39m ProtocolError(\u001b[39m\"\u001b[39m\u001b[39mConnection aborted.\u001b[39m\u001b[39m\"\u001b[39m, e)\n\u001b[1;32m--> 798\u001b[0m retries \u001b[39m=\u001b[39m retries\u001b[39m.\u001b[39;49mincrement(\n\u001b[0;32m    799\u001b[0m     method, url, error\u001b[39m=\u001b[39;49me, _pool\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m, _stacktrace\u001b[39m=\u001b[39;49msys\u001b[39m.\u001b[39;49mexc_info()[\u001b[39m2\u001b[39;49m]\n\u001b[0;32m    800\u001b[0m )\n\u001b[0;32m    801\u001b[0m retries\u001b[39m.\u001b[39msleep()\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\urllib3\\util\\retry.py:592\u001b[0m, in \u001b[0;36mRetry.increment\u001b[1;34m(self, method, url, response, error, _pool, _stacktrace)\u001b[0m\n\u001b[0;32m    591\u001b[0m \u001b[39mif\u001b[39;00m new_retry\u001b[39m.\u001b[39mis_exhausted():\n\u001b[1;32m--> 592\u001b[0m     \u001b[39mraise\u001b[39;00m MaxRetryError(_pool, url, error \u001b[39mor\u001b[39;00m ResponseError(cause))\n\u001b[0;32m    594\u001b[0m log\u001b[39m.\u001b[39mdebug(\u001b[39m\"\u001b[39m\u001b[39mIncremented Retry for (url=\u001b[39m\u001b[39m'\u001b[39m\u001b[39m%s\u001b[39;00m\u001b[39m'\u001b[39m\u001b[39m): \u001b[39m\u001b[39m%r\u001b[39;00m\u001b[39m\"\u001b[39m, url, new_retry)\n",
      "\u001b[1;31mMaxRetryError\u001b[0m: HTTPSConnectionPool(host='earthengine.googleapis.com', port=443): Max retries exceeded with url: /v1/projects/rcr-gee/operations?pageSize=500&alt=json (Caused by SSLError(SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1002)')))",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mSSLError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32mz:\\Projects\\06_LCMS_4_NFS\\Scripts\\lcms-training\\6-Map_Assemblage.ipynb Cell 41\u001b[0m line \u001b[0;36m3\n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m \u001b[39m# Can track tasks here or at https://code.earthengine.google.com/tasks\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m \u001b[39m# If you'd like to track the tasks, use this:\u001b[39;00m\n\u001b[1;32m----> <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m tml\u001b[39m.\u001b[39;49mtrackTasks2()\n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=4'>5</a>\u001b[0m \u001b[39m# If you want to cancel all running tasks, you can use this function\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=5'>6</a>\u001b[0m \u001b[39m# tml.batchCancel()\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m \n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=7'>8</a>\u001b[0m \u001b[39m# If you want to empty the collection of all images\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=8'>9</a>\u001b[0m \u001b[39m# aml.batchDelete(export_rawLCMSOutputs_collection, type = 'imageCollection')\u001b[39;00m\n\u001b[0;32m     <a href='vscode-notebook-cell:/z%3A/Projects/06_LCMS_4_NFS/Scripts/lcms-training/6-Map_Assemblage.ipynb#X55sZmlsZQ%3D%3D?line=10'>11</a>\u001b[0m \u001b[39mprint\u001b[39m(\u001b[39m'\u001b[39m\u001b[39mdone\u001b[39m\u001b[39m'\u001b[39m)\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\geeViz\\taskManagerLib.py:54\u001b[0m, in \u001b[0;36mtrackTasks2\u001b[1;34m(credential_name, id_list, task_count)\u001b[0m\n\u001b[0;32m     52\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mtrackTasks2\u001b[39m(credential_name \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m,id_list \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m,task_count \u001b[39m=\u001b[39m \u001b[39m1\u001b[39m):\n\u001b[0;32m     53\u001b[0m   \u001b[39mwhile\u001b[39;00m task_count \u001b[39m>\u001b[39m \u001b[39m0\u001b[39m:\n\u001b[1;32m---> 54\u001b[0m     tasks \u001b[39m=\u001b[39m ee\u001b[39m.\u001b[39;49mdata\u001b[39m.\u001b[39;49mgetTaskList()\n\u001b[0;32m     55\u001b[0m     \u001b[39mif\u001b[39;00m id_list \u001b[39m!=\u001b[39m \u001b[39mNone\u001b[39;00m:\n\u001b[0;32m     56\u001b[0m       tasks \u001b[39m=\u001b[39m [i \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m tasks \u001b[39mif\u001b[39;00m i[\u001b[39m'\u001b[39m\u001b[39mdescription\u001b[39m\u001b[39m'\u001b[39m] \u001b[39min\u001b[39;00m id_list]\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\ee\\deprecation.py:30\u001b[0m, in \u001b[0;36mDeprecated.<locals>.Decorator.<locals>.Wrapper\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     23\u001b[0m \u001b[39m@functools\u001b[39m\u001b[39m.\u001b[39mwraps(func)\n\u001b[0;32m     24\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mWrapper\u001b[39m(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs):\n\u001b[0;32m     25\u001b[0m   warnings\u001b[39m.\u001b[39mwarn_explicit(\n\u001b[0;32m     26\u001b[0m       \u001b[39m'\u001b[39m\u001b[39m%s\u001b[39;00m\u001b[39m() is deprecated: \u001b[39m\u001b[39m%s\u001b[39;00m\u001b[39m'\u001b[39m \u001b[39m%\u001b[39m (func\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m, message),\n\u001b[0;32m     27\u001b[0m       category\u001b[39m=\u001b[39m\u001b[39mDeprecationWarning\u001b[39;00m,\n\u001b[0;32m     28\u001b[0m       filename\u001b[39m=\u001b[39mfunc\u001b[39m.\u001b[39m\u001b[39m__code__\u001b[39m\u001b[39m.\u001b[39mco_filename,\n\u001b[0;32m     29\u001b[0m       lineno\u001b[39m=\u001b[39mfunc\u001b[39m.\u001b[39m\u001b[39m__code__\u001b[39m\u001b[39m.\u001b[39mco_firstlineno \u001b[39m+\u001b[39m \u001b[39m1\u001b[39m)\n\u001b[1;32m---> 30\u001b[0m   \u001b[39mreturn\u001b[39;00m func(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\ee\\data.py:1541\u001b[0m, in \u001b[0;36mgetTaskList\u001b[1;34m()\u001b[0m\n\u001b[0;32m   1531\u001b[0m \u001b[39m@deprecation\u001b[39m\u001b[39m.\u001b[39mDeprecated(\u001b[39m'\u001b[39m\u001b[39mUse listOperations\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[0;32m   1532\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mgetTaskList\u001b[39m() \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m List[Any]:\n\u001b[0;32m   1533\u001b[0m \u001b[39m  \u001b[39m\u001b[39m\"\"\"Retrieves a list of the user's tasks.\u001b[39;00m\n\u001b[0;32m   1534\u001b[0m \n\u001b[0;32m   1535\u001b[0m \u001b[39m  Returns:\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m   1538\u001b[0m \u001b[39m    canceled or failed tasks.\u001b[39;00m\n\u001b[0;32m   1539\u001b[0m \u001b[39m  \"\"\"\u001b[39;00m\n\u001b[0;32m   1540\u001b[0m   \u001b[39mreturn\u001b[39;00m [_cloud_api_utils\u001b[39m.\u001b[39mconvert_operation_to_task(o)\n\u001b[1;32m-> 1541\u001b[0m           \u001b[39mfor\u001b[39;00m o \u001b[39min\u001b[39;00m listOperations()]\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\ee\\data.py:1564\u001b[0m, in \u001b[0;36mlistOperations\u001b[1;34m(project)\u001b[0m\n\u001b[0;32m   1558\u001b[0m request \u001b[39m=\u001b[39m (\n\u001b[0;32m   1559\u001b[0m     _get_cloud_projects()\n\u001b[0;32m   1560\u001b[0m     \u001b[39m.\u001b[39moperations()\n\u001b[0;32m   1561\u001b[0m     \u001b[39m.\u001b[39mlist(pageSize\u001b[39m=\u001b[39m_TASKLIST_PAGE_SIZE, name\u001b[39m=\u001b[39mproject)\n\u001b[0;32m   1562\u001b[0m )\n\u001b[0;32m   1563\u001b[0m \u001b[39mwhile\u001b[39;00m request \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m-> 1564\u001b[0m   response \u001b[39m=\u001b[39m _execute_cloud_call(request)\n\u001b[0;32m   1565\u001b[0m   operations \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m response\u001b[39m.\u001b[39mget(\u001b[39m'\u001b[39m\u001b[39moperations\u001b[39m\u001b[39m'\u001b[39m, [])\n\u001b[0;32m   1566\u001b[0m   request \u001b[39m=\u001b[39m _get_cloud_projects()\u001b[39m.\u001b[39moperations()\u001b[39m.\u001b[39mlist_next(request, response)\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\ee\\data.py:354\u001b[0m, in \u001b[0;36m_execute_cloud_call\u001b[1;34m(call, num_retries)\u001b[0m\n\u001b[0;32m    340\u001b[0m \u001b[39m\u001b[39m\u001b[39m\"\"\"Executes a Cloud API call and translates errors to EEExceptions.\u001b[39;00m\n\u001b[0;32m    341\u001b[0m \n\u001b[0;32m    342\u001b[0m \u001b[39mArgs:\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    351\u001b[0m \u001b[39m  EEException if the call fails.\u001b[39;00m\n\u001b[0;32m    352\u001b[0m \u001b[39m\"\"\"\u001b[39;00m\n\u001b[0;32m    353\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[1;32m--> 354\u001b[0m   \u001b[39mreturn\u001b[39;00m call\u001b[39m.\u001b[39;49mexecute(num_retries\u001b[39m=\u001b[39;49mnum_retries)\n\u001b[0;32m    355\u001b[0m \u001b[39mexcept\u001b[39;00m googleapiclient\u001b[39m.\u001b[39merrors\u001b[39m.\u001b[39mHttpError \u001b[39mas\u001b[39;00m e:\n\u001b[0;32m    356\u001b[0m   \u001b[39mraise\u001b[39;00m _translate_cloud_exception(e)\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\googleapiclient\\_helpers.py:130\u001b[0m, in \u001b[0;36mpositional.<locals>.positional_decorator.<locals>.positional_wrapper\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m    128\u001b[0m     \u001b[39melif\u001b[39;00m positional_parameters_enforcement \u001b[39m==\u001b[39m POSITIONAL_WARNING:\n\u001b[0;32m    129\u001b[0m         logger\u001b[39m.\u001b[39mwarning(message)\n\u001b[1;32m--> 130\u001b[0m \u001b[39mreturn\u001b[39;00m wrapped(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\googleapiclient\\http.py:923\u001b[0m, in \u001b[0;36mHttpRequest.execute\u001b[1;34m(self, http, num_retries)\u001b[0m\n\u001b[0;32m    920\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mheaders[\u001b[39m\"\u001b[39m\u001b[39mcontent-length\u001b[39m\u001b[39m\"\u001b[39m] \u001b[39m=\u001b[39m \u001b[39mstr\u001b[39m(\u001b[39mlen\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mbody))\n\u001b[0;32m    922\u001b[0m \u001b[39m# Handle retries for server-side errors.\u001b[39;00m\n\u001b[1;32m--> 923\u001b[0m resp, content \u001b[39m=\u001b[39m _retry_request(\n\u001b[0;32m    924\u001b[0m     http,\n\u001b[0;32m    925\u001b[0m     num_retries,\n\u001b[0;32m    926\u001b[0m     \u001b[39m\"\u001b[39;49m\u001b[39mrequest\u001b[39;49m\u001b[39m\"\u001b[39;49m,\n\u001b[0;32m    927\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_sleep,\n\u001b[0;32m    928\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_rand,\n\u001b[0;32m    929\u001b[0m     \u001b[39mstr\u001b[39;49m(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49muri),\n\u001b[0;32m    930\u001b[0m     method\u001b[39m=\u001b[39;49m\u001b[39mstr\u001b[39;49m(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mmethod),\n\u001b[0;32m    931\u001b[0m     body\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mbody,\n\u001b[0;32m    932\u001b[0m     headers\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mheaders,\n\u001b[0;32m    933\u001b[0m )\n\u001b[0;32m    935\u001b[0m \u001b[39mfor\u001b[39;00m callback \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mresponse_callbacks:\n\u001b[0;32m    936\u001b[0m     callback(resp)\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\googleapiclient\\http.py:191\u001b[0m, in \u001b[0;36m_retry_request\u001b[1;34m(http, num_retries, req_type, sleep, rand, uri, method, *args, **kwargs)\u001b[0m\n\u001b[0;32m    189\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m    190\u001b[0m     exception \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n\u001b[1;32m--> 191\u001b[0m     resp, content \u001b[39m=\u001b[39m http\u001b[39m.\u001b[39;49mrequest(uri, method, \u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[0;32m    192\u001b[0m \u001b[39m# Retry on SSL errors and socket timeout errors.\u001b[39;00m\n\u001b[0;32m    193\u001b[0m \u001b[39mexcept\u001b[39;00m _ssl_SSLError \u001b[39mas\u001b[39;00m ssl_error:\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\google_auth_httplib2.py:218\u001b[0m, in \u001b[0;36mAuthorizedHttp.request\u001b[1;34m(self, uri, method, body, headers, redirections, connection_type, **kwargs)\u001b[0m\n\u001b[0;32m    215\u001b[0m     body_stream_position \u001b[39m=\u001b[39m body\u001b[39m.\u001b[39mtell()\n\u001b[0;32m    217\u001b[0m \u001b[39m# Make the request.\u001b[39;00m\n\u001b[1;32m--> 218\u001b[0m response, content \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mhttp\u001b[39m.\u001b[39;49mrequest(\n\u001b[0;32m    219\u001b[0m     uri,\n\u001b[0;32m    220\u001b[0m     method,\n\u001b[0;32m    221\u001b[0m     body\u001b[39m=\u001b[39;49mbody,\n\u001b[0;32m    222\u001b[0m     headers\u001b[39m=\u001b[39;49mrequest_headers,\n\u001b[0;32m    223\u001b[0m     redirections\u001b[39m=\u001b[39;49mredirections,\n\u001b[0;32m    224\u001b[0m     connection_type\u001b[39m=\u001b[39;49mconnection_type,\n\u001b[0;32m    225\u001b[0m     \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs\n\u001b[0;32m    226\u001b[0m )\n\u001b[0;32m    228\u001b[0m \u001b[39m# If the response indicated that the credentials needed to be\u001b[39;00m\n\u001b[0;32m    229\u001b[0m \u001b[39m# refreshed, then refresh the credentials and re-attempt the\u001b[39;00m\n\u001b[0;32m    230\u001b[0m \u001b[39m# request.\u001b[39;00m\n\u001b[0;32m    231\u001b[0m \u001b[39m# A stored token may expire between the time it is retrieved and\u001b[39;00m\n\u001b[0;32m    232\u001b[0m \u001b[39m# the time the request is made, so we may need to try twice.\u001b[39;00m\n\u001b[0;32m    233\u001b[0m \u001b[39mif\u001b[39;00m (\n\u001b[0;32m    234\u001b[0m     response\u001b[39m.\u001b[39mstatus \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_refresh_status_codes\n\u001b[0;32m    235\u001b[0m     \u001b[39mand\u001b[39;00m _credential_refresh_attempt \u001b[39m<\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_max_refresh_attempts\n\u001b[0;32m    236\u001b[0m ):\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\ee\\_cloud_api_utils.py:63\u001b[0m, in \u001b[0;36m_Http.request\u001b[1;34m(***failed resolving arguments***)\u001b[0m\n\u001b[0;32m     61\u001b[0m \u001b[39mwith\u001b[39;00m requests\u001b[39m.\u001b[39mSession() \u001b[39mas\u001b[39;00m session:\n\u001b[0;32m     62\u001b[0m   session\u001b[39m.\u001b[39mmax_redirects \u001b[39m=\u001b[39m redirections\n\u001b[1;32m---> 63\u001b[0m   response \u001b[39m=\u001b[39m session\u001b[39m.\u001b[39;49mrequest(\n\u001b[0;32m     64\u001b[0m       method, uri, data\u001b[39m=\u001b[39;49mbody, headers\u001b[39m=\u001b[39;49mheaders, timeout\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_timeout)\n\u001b[0;32m     65\u001b[0m   headers \u001b[39m=\u001b[39m \u001b[39mdict\u001b[39m(response\u001b[39m.\u001b[39mheaders)\n\u001b[0;32m     66\u001b[0m   headers[\u001b[39m'\u001b[39m\u001b[39mstatus\u001b[39m\u001b[39m'\u001b[39m] \u001b[39m=\u001b[39m response\u001b[39m.\u001b[39mstatus_code\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\requests\\sessions.py:589\u001b[0m, in \u001b[0;36mSession.request\u001b[1;34m(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)\u001b[0m\n\u001b[0;32m    584\u001b[0m send_kwargs \u001b[39m=\u001b[39m {\n\u001b[0;32m    585\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39mtimeout\u001b[39m\u001b[39m\"\u001b[39m: timeout,\n\u001b[0;32m    586\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39mallow_redirects\u001b[39m\u001b[39m\"\u001b[39m: allow_redirects,\n\u001b[0;32m    587\u001b[0m }\n\u001b[0;32m    588\u001b[0m send_kwargs\u001b[39m.\u001b[39mupdate(settings)\n\u001b[1;32m--> 589\u001b[0m resp \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49msend(prep, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49msend_kwargs)\n\u001b[0;32m    591\u001b[0m \u001b[39mreturn\u001b[39;00m resp\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\requests\\sessions.py:703\u001b[0m, in \u001b[0;36mSession.send\u001b[1;34m(self, request, **kwargs)\u001b[0m\n\u001b[0;32m    700\u001b[0m start \u001b[39m=\u001b[39m preferred_clock()\n\u001b[0;32m    702\u001b[0m \u001b[39m# Send the request\u001b[39;00m\n\u001b[1;32m--> 703\u001b[0m r \u001b[39m=\u001b[39m adapter\u001b[39m.\u001b[39;49msend(request, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[0;32m    705\u001b[0m \u001b[39m# Total elapsed time of the request (approximately)\u001b[39;00m\n\u001b[0;32m    706\u001b[0m elapsed \u001b[39m=\u001b[39m preferred_clock() \u001b[39m-\u001b[39m start\n",
      "File \u001b[1;32mc:\\Users\\ihousman\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\requests\\adapters.py:517\u001b[0m, in \u001b[0;36mHTTPAdapter.send\u001b[1;34m(self, request, stream, timeout, verify, cert, proxies)\u001b[0m\n\u001b[0;32m    513\u001b[0m         \u001b[39mraise\u001b[39;00m ProxyError(e, request\u001b[39m=\u001b[39mrequest)\n\u001b[0;32m    515\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(e\u001b[39m.\u001b[39mreason, _SSLError):\n\u001b[0;32m    516\u001b[0m         \u001b[39m# This branch is for urllib3 v1.22 and later.\u001b[39;00m\n\u001b[1;32m--> 517\u001b[0m         \u001b[39mraise\u001b[39;00m SSLError(e, request\u001b[39m=\u001b[39mrequest)\n\u001b[0;32m    519\u001b[0m     \u001b[39mraise\u001b[39;00m \u001b[39mConnectionError\u001b[39;00m(e, request\u001b[39m=\u001b[39mrequest)\n\u001b[0;32m    521\u001b[0m \u001b[39mexcept\u001b[39;00m ClosedPoolError \u001b[39mas\u001b[39;00m e:\n",
      "\u001b[1;31mSSLError\u001b[0m: HTTPSConnectionPool(host='earthengine.googleapis.com', port=443): Max retries exceeded with url: /v1/projects/rcr-gee/operations?pageSize=500&alt=json (Caused by SSLError(SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1002)')))"
     ]
    }
   ],
   "source": [
    "# Can track tasks here or at https://code.earthengine.google.com/tasks\n",
    "# If you'd like to track the tasks, use this:\n",
    "tml.trackTasks2()\n",
    "\n",
    "# If you want to cancel all running tasks, you can use this function\n",
    "# tml.batchCancel()\n",
    "\n",
    "# If you want to empty the collection of all images\n",
    "# aml.batchDelete(export_rawLCMSOutputs_collection, type = 'imageCollection')\n",
    "\n",
    "print('done')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88c2f60d",
   "metadata": {},
   "source": [
    "### Inspect final outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "254dda56",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adding layer: Full Stack Output\n",
      "Adding layer: LCMS Change\n",
      "Adding layer: LCMS Land_Cover\n",
      "Adding layer: LCMS Land_Use\n",
      "Starting webmap\n",
      "Using default refresh token for geeView: C:\\Users\\ihousman/.config/earthengine/credentials\n",
      "Local web server at: http://localhost:1236/geeView/ already serving.\n",
      "cwd z:\\Projects\\06_LCMS_4_NFS\\Scripts\\lcms-training\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"525px\"\n",
       "            src=\"http://localhost:1236/geeView/?accessToken=ya29.a0AfB_byBuPUr-_z7Ouu44r-H-OHRZY8sLB_AX-vJvZnDH2WlZuQGmE0BnPPHQej6eu9KApmjhgPi5Pa8mAhZpPLapqDcf-GW-C_ZaC6ixeE_9EeAFAmFGO5a2ziXo3MESwJ3du6ZDU4SILfdb4J2gHtkEt1jNh6wwXHYs_1Py1ytnaCgYKAaoSARASFQHGX2MiCsbu82xRjr2Z54Nf5xHq3g0179\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x281eb6a11d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# View final outputs\n",
    "Map.clearMap()\n",
    "lcms = ee.ImageCollection(export_assembledLCMSOutputs_collection)\n",
    "\n",
    "Map.addLayer(lcms,{'opacity':0},'Full Stack Output')\n",
    "lcms = lcms.map(lambda img:img.set(lcms_viz_dict))\n",
    "for product in products:\n",
    "    Map.addTimeLapse(lcms.select([product]),{'autoViz':True},f'LCMS {product}')\n",
    "\n",
    "Map.turnOnInspector()\n",
    "Map.view()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c9467fb",
   "metadata": {},
   "source": [
    "## Lab 6 challenge:\n",
    "\n",
    "**If you are accessing this lab through Qwiklabs, this challenge will be assessed for completion in Lab 6.**\n",
    "\n",
    "Create a layer that displays the most recent year of Gain. Extract the year value for the given location. Export the resulting value as a .csv. \n",
    "\n",
    "\n",
    "1. Create a layer that displays the most recent year of gain (class 4). Optionally view the layer in geeViz\n",
    "\n",
    "    * Use the pre-based assembed Change class:\n",
    "\n",
    "    ```python\n",
    "          lcms = ee.ImageCollection(export_assembledLCMSOutputs_collection).select(['Change'])\n",
    "\n",
    "      ```\n",
    "<br>\n",
    "\n",
    "2. Extract the year value for the following location. \n",
    "\n",
    "* Use a point with these coordinates: `([-66.127, 18.362])`\n",
    "        \n",
    "* Use the following function: `g2p.extractPointValuesToDataFrame`\n",
    "    <br>\n",
    "    Example:\n",
    "    ```python\n",
    "            extracted_values = g2p.extractPointValuesToDataFrame(\n",
    "            mostRecentGainYear,\n",
    "            ee.Geometry.Point([-66.127, 18.362]),\n",
    "            scale=30,\n",
    "            crs = \"EPSG:5070\", \n",
    "            transform = None,\n",
    "            reducer = ee.Reducer.first(),\n",
    "            includeNonSystemProperties = False,\n",
    "            includeSystemProperties=True\n",
    "            )\n",
    "    ```\n",
    "    <br> \n",
    "3. Save extracted values to a csv file.\n",
    "\n",
    "   * Save csv to this path: `\"/tmp/challenge/module_6_challenge_answer.csv\"`\n",
    "     * **Note: The path to the csv must exactly match the path above.** \n",
    "    <br>\n",
    "    \n",
    "    * Create the `\"/tmp/challenge\"` folder if it does not already exist.\n",
    "      \n",
    "        Example:\n",
    "    ```python\n",
    "        out_csv = \"/tmp/challenge/module_6_challenge_answer.csv\"\n",
    "        if not os.path.exists(os.path.dirname(out_csv)):os.makedirs(os.path.dirname(out_csv))\n",
    "    ```\n",
    "<br>\n",
    "\n",
    "4.  Check that the output csv exists.\n",
    "    \n",
    "    * Example: \n",
    "    ```python\n",
    "        print(os.path.exists(out_csv))\n",
    "    ```\n",
    "<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "id": "ce9d212a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# insert challenge code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68ca858b",
   "metadata": {},
   "source": [
    "## Done with Module 6"
   ]
  }
 ],
 "metadata": {
  "environment": {
   "kernel": "python3",
   "name": "tf2-gpu.2-11.m111",
   "type": "gcloud",
   "uri": "gcr.io/deeplearning-platform-release/tf2-gpu.2-11:m111"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}