differential equation regression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "cbc22bff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:\n",
      "The TensorFlow contrib module will not be included in TensorFlow 2.0.\n",
      "For more information, please see:\n",
      "  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md\n",
      "  * https://github.com/tensorflow/addons\n",
      "  * https://github.com/tensorflow/io (for I/O related ops)\n",
      "If you depend on functionality not listed there, please file an issue.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/gpflow/session_manager.py:28: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/gpflow/misc.py:24: The name tf.GraphKeys is deprecated. Please use tf.compat.v1.GraphKeys instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/gpflow/training/tensorflow_optimizer.py:151: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"-1\"\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import tensorflow as tf\n",
    "\n",
    "import numpy as np\n",
    "from npde_master.utils import gen_data, plot_model, eval_model, em_int\n",
    "from npde_master.npde_helper import build_model, fit_model, save_model, load_model\n",
    "\n",
    "\n",
    "import os\n",
    "from os.path import join as oj\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import sem\n",
    "from collections import defaultdict\n",
    "\n",
    "from copy import deepcopy\n",
    "\n",
    "from utils.utils import cwd, set_up_plotting\n",
    "\n",
    "\n",
    "import time\n",
    "import datetime\n",
    "plt = set_up_plotting()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "01dedb91",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_and_fit_model(t, Y, de_type):\n",
    "    sess = tf.InteractiveSession()\n",
    "    de_type = de_type.lower()\n",
    "    if de_type == 'ode':\n",
    "\n",
    "    # for ode\n",
    "        npde = build_model(sess, t, Y, model='ode', sf0=1.0, ell0=np.ones(2), W=6, ktype=\"id\")\n",
    "\n",
    "        prior_sn = npde.sn.eval()\n",
    "        # print(f'Prior noise variances are {prior_sn}.')\n",
    "\n",
    "        npde = fit_model(sess, npde, t, Y, num_iter=500, print_every=50, eta=0.02, plot_=False)\n",
    "\n",
    "        posterior_sn = npde.sn.eval()\n",
    "\n",
    "    elif de_type == 'vdp':\n",
    "    # for vdp\n",
    "        npde = build_model(sess, t, Y, model='sde', sf0=1.0, ell0=np.ones(2), W=6, ellg0=[1e5], ktype=\"id\")\n",
    "\n",
    "        prior_sn = npde.sn.eval()\n",
    "\n",
    "        npde = fit_model(sess, npde, t, Y, Nw=100, num_iter=500, print_every=50, eta=0.02, plot_=False)\n",
    "\n",
    "        posterior_sn = npde.sn.eval()\n",
    "\n",
    "    elif de_type == 'sde':\n",
    "    # for sde \n",
    "        npde = build_model(sess, t, Y, model='sde', sf0=1.0, ell0=np.ones(2), W=6, ellg0=[1.0], ktype=\"id\", fix_Z=True)\n",
    "\n",
    "        prior_sn = npde.sn.eval()\n",
    "\n",
    "        npde = fit_model(sess, npde, t, Y, Nw=50, num_iter=500, print_every=50, eta=0.01, plot_=False)\n",
    "\n",
    "        posterior_sn = npde.sn.eval()\n",
    "    \n",
    "    else:\n",
    "        raise NotImplementedError(f\"DE type: {de_type} not implemented \")\n",
    "\n",
    "    del sess\n",
    "    return npde, prior_sn, posterior_sn\n",
    "\n",
    "def get_times(overall_t, overall_Y, baseline_obs):\n",
    "    baseline_obs = np.unique(baseline_obs, axis=0)\n",
    "\n",
    "    baseline_times = []\n",
    "    baseline_obs_time_synced = []\n",
    "    for t_i, y_i in zip(overall_t, overall_Y):\n",
    "        for ob_i in baseline_obs:\n",
    "            if (y_i == ob_i).all():\n",
    "                baseline_times.append(t_i)\n",
    "                baseline_obs_time_synced.append(list(y_i))\n",
    "                \n",
    "    baseline_obs_time_synced = np.asarray(baseline_obs_time_synced)\n",
    "    baseline_times = np.asarray(baseline_times)\n",
    "    assert len(baseline_times) == len(baseline_obs_time_synced)\n",
    "    return baseline_times, baseline_obs_time_synced \n",
    "\n",
    "\n",
    "def get_rmse(overall_t, overall_Y, baseline_obs, target, de_type):     \n",
    "    baseline_times, baseline_obs_synced = get_times(overall_t, overall_Y, baseline_obs)\n",
    "    npde, prior_sn, posterior_sn = create_and_fit_model([baseline_times], [baseline_obs_synced], de_type)\n",
    "    \n",
    "    target_times, target_synced = get_times(overall_t, overall_Y, target)\n",
    "    \n",
    "    return eval_model(npde, [target_times], [target_synced], Nw=1)\n",
    "\n",
    "def get_regression_df(mse_results):\n",
    "    data_df = defaultdict(list)\n",
    "    for collab_type, mse_list in mse_results.items():\n",
    "        baseline = collab_type.replace('-avg-mses', '').replace('-mses', '').replace('_obs','')\n",
    "        if baseline not in data_df['Baselines']:\n",
    "            data_df['Baselines'].append(baseline)\n",
    "        if '-avg-mses' in collab_type:\n",
    "            avg = np.mean(mse_list)\n",
    "            se = sem(mse_list)\n",
    "            data_df['Avg MSE'].append(avg)\n",
    "            data_df['Stderr'].append(se)\n",
    "        else:\n",
    "            stds = np.std(mse_list, axis=1)\n",
    "\n",
    "            mean_std_mse = np.mean(stds)\n",
    "            se_std_mse = sem(stds)\n",
    "            data_df['Std MSE'].append(mean_std_mse)\n",
    "            data_df['Stderr Std'].append(se_std_mse)\n",
    "    \n",
    "    return pd.DataFrame(data=data_df)\n",
    "\n",
    "\n",
    "def get_mses(collab_type, n, overall_t, overall_Y, baseline_obs, Ts, de_type):\n",
    "    if 'indiv' in collab_type:\n",
    "        mses = [get_rmse(overall_t, overall_Y, baseline_obs[i], Ts[i], de_type) for i in range(n) ]\n",
    "\n",
    "    else:\n",
    "        mses = []\n",
    "\n",
    "        baseline_obs = np.unique(baseline_obs, axis=0)\n",
    "        baseline_times, baseline_obs_synced = get_times(overall_t, overall_Y, baseline_obs)\n",
    "        npde, prior_sn, posterior_sn = create_and_fit_model([baseline_times], [baseline_obs_synced], de_type)\n",
    "        \n",
    "        for i, target in enumerate(Ts):\n",
    "            target = np.unique(target, axis=0)\n",
    "            target_times, target_synced = get_times(overall_t, overall_Y, target)\n",
    "            mses.append(eval_model(npde, [target_times], [target_synced], Nw=1))                                   \n",
    "    return mses\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2a840b58",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d86db334",
   "metadata": {},
   "outputs": [],
   "source": [
    "def run_regression_analysis(de_type='ODE'):\n",
    "    results_dir = oj('results', 'DE')\n",
    "    \n",
    "    with cwd(results_dir):\n",
    "        result_datas = []\n",
    "        for file in os.listdir():\n",
    "            if file.endswith('.npz') and de_type.upper() in file:\n",
    "                result_datas.append(np.load(file, allow_pickle=True))\n",
    "\n",
    "        mse_results = defaultdict(list)\n",
    "        \n",
    "        for trial_i in range(n_trials):\n",
    "            print(f'**** trial index {trial_i+1} **** ')\n",
    "            print()\n",
    "            obs = result_datas[trial_i]['obs'].item()\n",
    "            times = result_datas[trial_i]['times']\n",
    "\n",
    "            overall_t = result_datas[trial_i]['t'][trajectory_index]\n",
    "            overall_Y = result_datas[trial_i]['Y'][trajectory_index]\n",
    "            \n",
    "            Ts = result_datas[trial_i]['Ts']\n",
    "            for collab_type, baseline_obs in obs.items():   \n",
    "#                 if 'rand' in collab_type:\n",
    "                if 'entropy' in collab_type or 'indiv' in collab_type: \n",
    "                    try:\n",
    "                        mses = get_mses(collab_type, n, overall_t, overall_Y, baseline_obs, Ts, de_type)\n",
    "                        mse = np.mean(mses)\n",
    "\n",
    "                        ode_mse_results[collab_type+'-avg-mses'].append(mse)        \n",
    "                        ode_mse_results[collab_type+'-mses'].append(mses)\n",
    "\n",
    "                        print(f'****** {collab_type} MSEs: {mses} Avg MSE: {mse} ******')\n",
    "                    except Exception as inst:\n",
    "                        print(type(inst))    # the exception instance\n",
    "                        print(inst.args)     # arguments stored in .args\n",
    "                        print(inst)          # __str__ allows args to be printed directly,\n",
    "                                             # but may be overridden in exception subclasses\n",
    "\n",
    "        regression_dir = 'regression_results'\n",
    "#         os.makedirs(regression_dir)\n",
    "        \n",
    "#         time_str = datetime.datetime.fromtimestamp(time.time()).strftime('%m-%d-%H:%M:%S')\n",
    "\n",
    "#         regression_df = get_regression_df(mse_results)\n",
    "#         regression_df.to_latex(oj(regression_dir, f'{de_type}-regression-{time_str}.tex'), index=False) \n",
    "#         regression_df.to_csv(oj(regression_dir, f'{de_type}-regression-{time_str}.csv'), index=False) \n",
    "    return"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "aa7b239b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "**** trial index 1 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21432883 0.21690132]\n",
      "signal variance:       1.3461104682735887\n",
      "lengthscales:          [0.9438477  1.08880317]\n",
      "diff signal variance:  0.8305085514857685\n",
      "diff lengthscales:     [0.89734469]\n",
      "RMSE over the dimensions is: 3.9371207460941218\n",
      "name:                  npsde\n",
      "noise variance:        [0.21432883 0.21690132]\n",
      "signal variance:       1.3461104682735887\n",
      "lengthscales:          [0.9438477  1.08880317]\n",
      "diff signal variance:  0.8305085514857685\n",
      "diff lengthscales:     [0.89734469]\n",
      "RMSE over the dimensions is: 3.727436095003661\n",
      "name:                  npsde\n",
      "noise variance:        [0.21432883 0.21690132]\n",
      "signal variance:       1.3461104682735887\n",
      "lengthscales:          [0.9438477  1.08880317]\n",
      "diff signal variance:  0.8305085514857685\n",
      "diff lengthscales:     [0.89734469]\n",
      "RMSE over the dimensions is: 6.882562789145073\n",
      "**** greedy_obs_1 MSEs: [3.9371207460941218, 3.727436095003661, 6.882562789145073] Avg MSE: 4.849039876747619 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21469824 0.21123287]\n",
      "signal variance:       1.1180115103091877\n",
      "lengthscales:          [0.94662216 1.14066593]\n",
      "diff signal variance:  0.8726654401142518\n",
      "diff lengthscales:     [0.96067933]\n",
      "RMSE over the dimensions is: 3.7546719285084773\n",
      "name:                  npsde\n",
      "noise variance:        [0.21469824 0.21123287]\n",
      "signal variance:       1.1180115103091877\n",
      "lengthscales:          [0.94662216 1.14066593]\n",
      "diff signal variance:  0.8726654401142518\n",
      "diff lengthscales:     [0.96067933]\n",
      "RMSE over the dimensions is: 3.8154058355840528\n",
      "name:                  npsde\n",
      "noise variance:        [0.21469824 0.21123287]\n",
      "signal variance:       1.1180115103091877\n",
      "lengthscales:          [0.94662216 1.14066593]\n",
      "diff signal variance:  0.8726654401142518\n",
      "diff lengthscales:     [0.96067933]\n",
      "RMSE over the dimensions is: 6.37283873477589\n",
      "**** greedy_obs_2 MSEs: [3.7546719285084773, 3.8154058355840528, 6.37283873477589] Avg MSE: 4.64763883295614 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.22520997 0.2205616 ]\n",
      "signal variance:       1.1058162214773528\n",
      "lengthscales:          [0.98773267 1.05723752]\n",
      "diff signal variance:  0.8999157676455367\n",
      "diff lengthscales:     [0.94068676]\n",
      "RMSE over the dimensions is: 4.229803327857271\n",
      "name:                  npsde\n",
      "noise variance:        [0.22520997 0.2205616 ]\n",
      "signal variance:       1.1058162214773528\n",
      "lengthscales:          [0.98773267 1.05723752]\n",
      "diff signal variance:  0.8999157676455367\n",
      "diff lengthscales:     [0.94068676]\n",
      "RMSE over the dimensions is: 3.8138798488252688\n",
      "name:                  npsde\n",
      "noise variance:        [0.22520997 0.2205616 ]\n",
      "signal variance:       1.1058162214773528\n",
      "lengthscales:          [0.98773267 1.05723752]\n",
      "diff signal variance:  0.8999157676455367\n",
      "diff lengthscales:     [0.94068676]\n",
      "RMSE over the dimensions is: 6.856487850679862\n",
      "**** greedy_obs_3 MSEs: [4.229803327857271, 3.8138798488252688, 6.856487850679862] Avg MSE: 4.966723675787467 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21218455 0.2109741 ]\n",
      "signal variance:       1.1869524792911261\n",
      "lengthscales:          [0.94565297 1.16041894]\n",
      "diff signal variance:  0.8293309101713909\n",
      "diff lengthscales:     [0.92584904]\n",
      "RMSE over the dimensions is: 3.70625904114334\n",
      "name:                  npsde\n",
      "noise variance:        [0.21218455 0.2109741 ]\n",
      "signal variance:       1.1869524792911261\n",
      "lengthscales:          [0.94565297 1.16041894]\n",
      "diff signal variance:  0.8293309101713909\n",
      "diff lengthscales:     [0.92584904]\n",
      "RMSE over the dimensions is: 3.9385932377896404\n",
      "name:                  npsde\n",
      "noise variance:        [0.21218455 0.2109741 ]\n",
      "signal variance:       1.1869524792911261\n",
      "lengthscales:          [0.94565297 1.16041894]\n",
      "diff signal variance:  0.8293309101713909\n",
      "diff lengthscales:     [0.92584904]\n",
      "RMSE over the dimensions is: 6.703070658898033\n",
      "**** greedy_obs_4 MSEs: [3.70625904114334, 3.9385932377896404, 6.703070658898033] Avg MSE: 4.782640979277004 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21569868 0.21176796]\n",
      "signal variance:       1.1189336114814912\n",
      "lengthscales:          [1.00776459 1.08946515]\n",
      "diff signal variance:  0.869934574023678\n",
      "diff lengthscales:     [0.94291251]\n",
      "RMSE over the dimensions is: 3.959341067590531\n",
      "name:                  npsde\n",
      "noise variance:        [0.21569868 0.21176796]\n",
      "signal variance:       1.1189336114814912\n",
      "lengthscales:          [1.00776459 1.08946515]\n",
      "diff signal variance:  0.869934574023678\n",
      "diff lengthscales:     [0.94291251]\n",
      "RMSE over the dimensions is: 3.730352750774405\n",
      "name:                  npsde\n",
      "noise variance:        [0.21569868 0.21176796]\n",
      "signal variance:       1.1189336114814912\n",
      "lengthscales:          [1.00776459 1.08946515]\n",
      "diff signal variance:  0.869934574023678\n",
      "diff lengthscales:     [0.94291251]\n",
      "RMSE over the dimensions is: 6.650311391855747\n",
      "**** greedy_sum_obs MSEs: [3.959341067590531, 3.730352750774405, 6.650311391855747] Avg MSE: 4.780001736740228 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21315271 0.21159585]\n",
      "signal variance:       1.1307068062594254\n",
      "lengthscales:          [0.95188769 1.1136779 ]\n",
      "diff signal variance:  0.8572269692742251\n",
      "diff lengthscales:     [0.99583286]\n",
      "RMSE over the dimensions is: 3.8747221184857095\n",
      "name:                  npsde\n",
      "noise variance:        [0.21315271 0.21159585]\n",
      "signal variance:       1.1307068062594254\n",
      "lengthscales:          [0.95188769 1.1136779 ]\n",
      "diff signal variance:  0.8572269692742251\n",
      "diff lengthscales:     [0.99583286]\n",
      "RMSE over the dimensions is: 3.6570728131159043\n",
      "name:                  npsde\n",
      "noise variance:        [0.21315271 0.21159585]\n",
      "signal variance:       1.1307068062594254\n",
      "lengthscales:          [0.95188769 1.1136779 ]\n",
      "diff signal variance:  0.8572269692742251\n",
      "diff lengthscales:     [0.99583286]\n",
      "RMSE over the dimensions is: 6.700667986273594\n",
      "**** dynamic_beta_obs MSEs: [3.8747221184857095, 3.6570728131159043, 6.700667986273594] Avg MSE: 4.744154305958403 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21231601 0.211573  ]\n",
      "signal variance:       1.133527987752562\n",
      "lengthscales:          [0.99317111 1.04232777]\n",
      "diff signal variance:  0.867431316899961\n",
      "diff lengthscales:     [0.96733784]\n",
      "RMSE over the dimensions is: 3.9094441638596207\n",
      "name:                  npsde\n",
      "noise variance:        [0.21231601 0.211573  ]\n",
      "signal variance:       1.133527987752562\n",
      "lengthscales:          [0.99317111 1.04232777]\n",
      "diff signal variance:  0.867431316899961\n",
      "diff lengthscales:     [0.96733784]\n",
      "RMSE over the dimensions is: 3.957151005677245\n",
      "name:                  npsde\n",
      "noise variance:        [0.21231601 0.211573  ]\n",
      "signal variance:       1.133527987752562\n",
      "lengthscales:          [0.99317111 1.04232777]\n",
      "diff signal variance:  0.867431316899961\n",
      "diff lengthscales:     [0.96733784]\n",
      "RMSE over the dimensions is: 6.9135995980375125\n",
      "**** joint_obs MSEs: [3.9094441638596207, 3.957151005677245, 6.9135995980375125] Avg MSE: 4.926731589191459 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21085772 0.20742548]\n",
      "signal variance:       1.1443748894311505\n",
      "lengthscales:          [1.01281477 1.08083825]\n",
      "diff signal variance:  0.8403134669510173\n",
      "diff lengthscales:     [0.87126878]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 4.013052048219262\n",
      "name:                  npsde\n",
      "noise variance:        [0.21085772 0.20742548]\n",
      "signal variance:       1.1443748894311505\n",
      "lengthscales:          [1.01281477 1.08083825]\n",
      "diff signal variance:  0.8403134669510173\n",
      "diff lengthscales:     [0.87126878]\n",
      "RMSE over the dimensions is: 3.7726927314296503\n",
      "name:                  npsde\n",
      "noise variance:        [0.21085772 0.20742548]\n",
      "signal variance:       1.1443748894311505\n",
      "lengthscales:          [1.01281477 1.08083825]\n",
      "diff signal variance:  0.8403134669510173\n",
      "diff lengthscales:     [0.87126878]\n",
      "RMSE over the dimensions is: 6.98800321826565\n",
      "**** rand_obs MSEs: [4.013052048219262, 3.7726927314296503, 6.98800321826565] Avg MSE: 4.924582665971521 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.19923394 0.19674557]\n",
      "signal variance:       1.2004068980487435\n",
      "lengthscales:          [0.82731412 1.18389405]\n",
      "diff signal variance:  0.9175167702488756\n",
      "diff lengthscales:     [0.98389786]\n",
      "RMSE over the dimensions is: 3.5530944321355857\n",
      "name:                  npsde\n",
      "noise variance:        [0.19923394 0.19674557]\n",
      "signal variance:       1.2004068980487435\n",
      "lengthscales:          [0.82731412 1.18389405]\n",
      "diff signal variance:  0.9175167702488756\n",
      "diff lengthscales:     [0.98389786]\n",
      "RMSE over the dimensions is: 3.6952728694126726\n",
      "name:                  npsde\n",
      "noise variance:        [0.19923394 0.19674557]\n",
      "signal variance:       1.2004068980487435\n",
      "lengthscales:          [0.82731412 1.18389405]\n",
      "diff signal variance:  0.9175167702488756\n",
      "diff lengthscales:     [0.98389786]\n",
      "RMSE over the dimensions is: 3.8258033143334305\n",
      "**** entropy_obs MSEs: [3.5530944321355857, 3.6952728694126726, 3.8258033143334305] Avg MSE: 3.6913902052938963 ****\n",
      "**** trial index 2 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.428434   0.36865686]\n",
      "signal variance:       0.991143585050166\n",
      "lengthscales:          [1.00894181 0.97146869]\n",
      "diff signal variance:  0.9258303187299289\n",
      "diff lengthscales:     [0.99857917]\n",
      "RMSE over the dimensions is: 0.7000241681757922\n",
      "name:                  npsde\n",
      "noise variance:        [0.428434   0.36865686]\n",
      "signal variance:       0.991143585050166\n",
      "lengthscales:          [1.00894181 0.97146869]\n",
      "diff signal variance:  0.9258303187299289\n",
      "diff lengthscales:     [0.99857917]\n",
      "RMSE over the dimensions is: 3.0120963273736723\n",
      "name:                  npsde\n",
      "noise variance:        [0.428434   0.36865686]\n",
      "signal variance:       0.991143585050166\n",
      "lengthscales:          [1.00894181 0.97146869]\n",
      "diff signal variance:  0.9258303187299289\n",
      "diff lengthscales:     [0.99857917]\n",
      "RMSE over the dimensions is: 4.929783565111187\n",
      "**** greedy_obs_1 MSEs: [0.7000241681757922, 3.0120963273736723, 4.929783565111187] Avg MSE: 2.8806346868868835 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21016298 0.23619773]\n",
      "signal variance:       1.1167607013070122\n",
      "lengthscales:          [1.05872961 0.93845748]\n",
      "diff signal variance:  0.8532947578290807\n",
      "diff lengthscales:     [0.92734998]\n",
      "RMSE over the dimensions is: 15.182058119952636\n",
      "name:                  npsde\n",
      "noise variance:        [0.21016298 0.23619773]\n",
      "signal variance:       1.1167607013070122\n",
      "lengthscales:          [1.05872961 0.93845748]\n",
      "diff signal variance:  0.8532947578290807\n",
      "diff lengthscales:     [0.92734998]\n",
      "RMSE over the dimensions is: 4.442046215165831\n",
      "name:                  npsde\n",
      "noise variance:        [0.21016298 0.23619773]\n",
      "signal variance:       1.1167607013070122\n",
      "lengthscales:          [1.05872961 0.93845748]\n",
      "diff signal variance:  0.8532947578290807\n",
      "diff lengthscales:     [0.92734998]\n",
      "RMSE over the dimensions is: 4.996128640762264\n",
      "**** greedy_obs_2 MSEs: [15.182058119952636, 4.442046215165831, 4.996128640762264] Avg MSE: 8.206744325293577 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.22056635 0.21661259]\n",
      "signal variance:       1.1578649326860744\n",
      "lengthscales:          [0.98546188 1.00033068]\n",
      "diff signal variance:  0.8758864702261084\n",
      "diff lengthscales:     [0.89229958]\n",
      "RMSE over the dimensions is: 0.942525732575727\n",
      "name:                  npsde\n",
      "noise variance:        [0.22056635 0.21661259]\n",
      "signal variance:       1.1578649326860744\n",
      "lengthscales:          [0.98546188 1.00033068]\n",
      "diff signal variance:  0.8758864702261084\n",
      "diff lengthscales:     [0.89229958]\n",
      "RMSE over the dimensions is: 3.7088600964850276\n",
      "name:                  npsde\n",
      "noise variance:        [0.22056635 0.21661259]\n",
      "signal variance:       1.1578649326860744\n",
      "lengthscales:          [0.98546188 1.00033068]\n",
      "diff signal variance:  0.8758864702261084\n",
      "diff lengthscales:     [0.89229958]\n",
      "RMSE over the dimensions is: 4.934132223540958\n",
      "**** greedy_obs_3 MSEs: [0.942525732575727, 3.7088600964850276, 4.934132223540958] Avg MSE: 3.195172684200571 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.24991016 0.21192379]\n",
      "signal variance:       1.137123902103656\n",
      "lengthscales:          [0.93782477 0.99363009]\n",
      "diff signal variance:  0.8949817144533216\n",
      "diff lengthscales:     [0.88706898]\n",
      "RMSE over the dimensions is: 0.8640490660132585\n",
      "name:                  npsde\n",
      "noise variance:        [0.24991016 0.21192379]\n",
      "signal variance:       1.137123902103656\n",
      "lengthscales:          [0.93782477 0.99363009]\n",
      "diff signal variance:  0.8949817144533216\n",
      "diff lengthscales:     [0.88706898]\n",
      "RMSE over the dimensions is: 4.383611650907755\n",
      "name:                  npsde\n",
      "noise variance:        [0.24991016 0.21192379]\n",
      "signal variance:       1.137123902103656\n",
      "lengthscales:          [0.93782477 0.99363009]\n",
      "diff signal variance:  0.8949817144533216\n",
      "diff lengthscales:     [0.88706898]\n",
      "RMSE over the dimensions is: 5.376546378262279\n",
      "**** greedy_obs_4 MSEs: [0.8640490660132585, 4.383611650907755, 5.376546378262279] Avg MSE: 3.5414023650610975 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.22214575 0.21806375]\n",
      "signal variance:       1.1466323274538124\n",
      "lengthscales:          [0.98187628 0.97106297]\n",
      "diff signal variance:  0.9116683523133863\n",
      "diff lengthscales:     [0.97137787]\n",
      "RMSE over the dimensions is: 9.569878122663559\n",
      "name:                  npsde\n",
      "noise variance:        [0.22214575 0.21806375]\n",
      "signal variance:       1.1466323274538124\n",
      "lengthscales:          [0.98187628 0.97106297]\n",
      "diff signal variance:  0.9116683523133863\n",
      "diff lengthscales:     [0.97137787]\n",
      "RMSE over the dimensions is: 0.35752572624779017\n",
      "name:                  npsde\n",
      "noise variance:        [0.22214575 0.21806375]\n",
      "signal variance:       1.1466323274538124\n",
      "lengthscales:          [0.98187628 0.97106297]\n",
      "diff signal variance:  0.9116683523133863\n",
      "diff lengthscales:     [0.97137787]\n",
      "RMSE over the dimensions is: 5.339784635000643\n",
      "**** greedy_sum_obs MSEs: [9.569878122663559, 0.35752572624779017, 5.339784635000643] Avg MSE: 5.089062827970664 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.4376177  0.29733927]\n",
      "signal variance:       1.0693757514743505\n",
      "lengthscales:          [0.95453221 0.97253413]\n",
      "diff signal variance:  0.9173397995006899\n",
      "diff lengthscales:     [0.93131455]\n",
      "RMSE over the dimensions is: 0.6110410492625463\n",
      "name:                  npsde\n",
      "noise variance:        [0.4376177  0.29733927]\n",
      "signal variance:       1.0693757514743505\n",
      "lengthscales:          [0.95453221 0.97253413]\n",
      "diff signal variance:  0.9173397995006899\n",
      "diff lengthscales:     [0.93131455]\n",
      "RMSE over the dimensions is: 3.047201460356969\n",
      "name:                  npsde\n",
      "noise variance:        [0.4376177  0.29733927]\n",
      "signal variance:       1.0693757514743505\n",
      "lengthscales:          [0.95453221 0.97253413]\n",
      "diff signal variance:  0.9173397995006899\n",
      "diff lengthscales:     [0.93131455]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 5.5149382389438895\n",
      "**** dynamic_beta_obs MSEs: [0.6110410492625463, 3.047201460356969, 5.5149382389438895] Avg MSE: 3.0577269161878013 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.44529949 0.22596995]\n",
      "signal variance:       1.0481407698694447\n",
      "lengthscales:          [1.01529337 0.95894985]\n",
      "diff signal variance:  0.9040212215862713\n",
      "diff lengthscales:     [0.9649967]\n",
      "RMSE over the dimensions is: 1.110862418175462\n",
      "name:                  npsde\n",
      "noise variance:        [0.44529949 0.22596995]\n",
      "signal variance:       1.0481407698694447\n",
      "lengthscales:          [1.01529337 0.95894985]\n",
      "diff signal variance:  0.9040212215862713\n",
      "diff lengthscales:     [0.9649967]\n",
      "RMSE over the dimensions is: 11.52559638870333\n",
      "name:                  npsde\n",
      "noise variance:        [0.44529949 0.22596995]\n",
      "signal variance:       1.0481407698694447\n",
      "lengthscales:          [1.01529337 0.95894985]\n",
      "diff signal variance:  0.9040212215862713\n",
      "diff lengthscales:     [0.9649967]\n",
      "RMSE over the dimensions is: 5.525628959314423\n",
      "**** joint_obs MSEs: [1.110862418175462, 11.52559638870333, 5.525628959314423] Avg MSE: 6.0540292553977375 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.23175254 0.20611855]\n",
      "signal variance:       1.1357973867002917\n",
      "lengthscales:          [0.98942398 0.99413138]\n",
      "diff signal variance:  0.875374647896645\n",
      "diff lengthscales:     [0.99533494]\n",
      "RMSE over the dimensions is: 1.2042883238886188\n",
      "name:                  npsde\n",
      "noise variance:        [0.23175254 0.20611855]\n",
      "signal variance:       1.1357973867002917\n",
      "lengthscales:          [0.98942398 0.99413138]\n",
      "diff signal variance:  0.875374647896645\n",
      "diff lengthscales:     [0.99533494]\n",
      "RMSE over the dimensions is: 3.6286560935167413\n",
      "name:                  npsde\n",
      "noise variance:        [0.23175254 0.20611855]\n",
      "signal variance:       1.1357973867002917\n",
      "lengthscales:          [0.98942398 0.99413138]\n",
      "diff signal variance:  0.875374647896645\n",
      "diff lengthscales:     [0.99533494]\n",
      "RMSE over the dimensions is: 5.310981714190368\n",
      "**** rand_obs MSEs: [1.2042883238886188, 3.6286560935167413, 5.310981714190368] Avg MSE: 3.3813087105319095 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.20863748 0.21725174]\n",
      "signal variance:       1.1477314635831344\n",
      "lengthscales:          [0.94050495 1.01581187]\n",
      "diff signal variance:  0.8889113278611778\n",
      "diff lengthscales:     [0.9523598]\n",
      "RMSE over the dimensions is: 0.622640541264697\n",
      "name:                  npsde\n",
      "noise variance:        [0.20863748 0.21725174]\n",
      "signal variance:       1.1477314635831344\n",
      "lengthscales:          [0.94050495 1.01581187]\n",
      "diff signal variance:  0.8889113278611778\n",
      "diff lengthscales:     [0.9523598]\n",
      "RMSE over the dimensions is: 1.1762722185470935\n",
      "name:                  npsde\n",
      "noise variance:        [0.20863748 0.21725174]\n",
      "signal variance:       1.1477314635831344\n",
      "lengthscales:          [0.94050495 1.01581187]\n",
      "diff signal variance:  0.8889113278611778\n",
      "diff lengthscales:     [0.9523598]\n",
      "RMSE over the dimensions is: 3.7313875435135078\n",
      "**** entropy_obs MSEs: [0.622640541264697, 1.1762722185470935, 3.7313875435135078] Avg MSE: 1.843433434441766 ****\n",
      "**** trial index 3 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.34830995 0.33638878]\n",
      "signal variance:       1.0980732521910053\n",
      "lengthscales:          [0.95229214 0.97894917]\n",
      "diff signal variance:  0.8952080521730112\n",
      "diff lengthscales:     [0.89566725]\n",
      "RMSE over the dimensions is: 3.536814682517542\n",
      "name:                  npsde\n",
      "noise variance:        [0.34830995 0.33638878]\n",
      "signal variance:       1.0980732521910053\n",
      "lengthscales:          [0.95229214 0.97894917]\n",
      "diff signal variance:  0.8952080521730112\n",
      "diff lengthscales:     [0.89566725]\n",
      "RMSE over the dimensions is: 4.866610861566275\n",
      "name:                  npsde\n",
      "noise variance:        [0.34830995 0.33638878]\n",
      "signal variance:       1.0980732521910053\n",
      "lengthscales:          [0.95229214 0.97894917]\n",
      "diff signal variance:  0.8952080521730112\n",
      "diff lengthscales:     [0.89566725]\n",
      "RMSE over the dimensions is: 6.088167390640395\n",
      "**** greedy_obs_1 MSEs: [3.536814682517542, 4.866610861566275, 6.088167390640395] Avg MSE: 4.830530978241404 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.22469092 0.26696415]\n",
      "signal variance:       1.0736921332455671\n",
      "lengthscales:          [1.00227351 1.00944793]\n",
      "diff signal variance:  0.8897039796546168\n",
      "diff lengthscales:     [0.93801175]\n",
      "RMSE over the dimensions is: 4.017329541211567\n",
      "name:                  npsde\n",
      "noise variance:        [0.22469092 0.26696415]\n",
      "signal variance:       1.0736921332455671\n",
      "lengthscales:          [1.00227351 1.00944793]\n",
      "diff signal variance:  0.8897039796546168\n",
      "diff lengthscales:     [0.93801175]\n",
      "RMSE over the dimensions is: 3.706809095418585\n",
      "name:                  npsde\n",
      "noise variance:        [0.22469092 0.26696415]\n",
      "signal variance:       1.0736921332455671\n",
      "lengthscales:          [1.00227351 1.00944793]\n",
      "diff signal variance:  0.8897039796546168\n",
      "diff lengthscales:     [0.93801175]\n",
      "RMSE over the dimensions is: 5.814971388846773\n",
      "**** greedy_obs_2 MSEs: [4.017329541211567, 3.706809095418585, 5.814971388846773] Avg MSE: 4.513036675158975 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.22949028 0.25756147]\n",
      "signal variance:       1.1141679271697797\n",
      "lengthscales:          [0.94484716 0.98434759]\n",
      "diff signal variance:  0.893923429776239\n",
      "diff lengthscales:     [0.88703754]\n",
      "RMSE over the dimensions is: 3.46011290857738\n",
      "name:                  npsde\n",
      "noise variance:        [0.22949028 0.25756147]\n",
      "signal variance:       1.1141679271697797\n",
      "lengthscales:          [0.94484716 0.98434759]\n",
      "diff signal variance:  0.893923429776239\n",
      "diff lengthscales:     [0.88703754]\n",
      "RMSE over the dimensions is: 4.460269604746667\n",
      "name:                  npsde\n",
      "noise variance:        [0.22949028 0.25756147]\n",
      "signal variance:       1.1141679271697797\n",
      "lengthscales:          [0.94484716 0.98434759]\n",
      "diff signal variance:  0.893923429776239\n",
      "diff lengthscales:     [0.88703754]\n",
      "RMSE over the dimensions is: 5.562736658266122\n",
      "**** greedy_obs_3 MSEs: [3.46011290857738, 4.460269604746667, 5.562736658266122] Avg MSE: 4.494373057196722 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21702823 0.24140233]\n",
      "signal variance:       1.0729666978254884\n",
      "lengthscales:          [0.98787918 1.00036054]\n",
      "diff signal variance:  0.9097305235075503\n",
      "diff lengthscales:     [0.91328077]\n",
      "RMSE over the dimensions is: 3.661954815938614\n",
      "name:                  npsde\n",
      "noise variance:        [0.21702823 0.24140233]\n",
      "signal variance:       1.0729666978254884\n",
      "lengthscales:          [0.98787918 1.00036054]\n",
      "diff signal variance:  0.9097305235075503\n",
      "diff lengthscales:     [0.91328077]\n",
      "RMSE over the dimensions is: 4.891401275367892\n",
      "name:                  npsde\n",
      "noise variance:        [0.21702823 0.24140233]\n",
      "signal variance:       1.0729666978254884\n",
      "lengthscales:          [0.98787918 1.00036054]\n",
      "diff signal variance:  0.9097305235075503\n",
      "diff lengthscales:     [0.91328077]\n",
      "RMSE over the dimensions is: 4.952816314789905\n",
      "**** greedy_obs_4 MSEs: [3.661954815938614, 4.891401275367892, 4.952816314789905] Avg MSE: 4.502057468698804 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.28958001 0.31864997]\n",
      "signal variance:       1.0493953926052533\n",
      "lengthscales:          [0.97301814 0.954996  ]\n",
      "diff signal variance:  0.895229121453946\n",
      "diff lengthscales:     [0.82685987]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 1.8862600200566044\n",
      "name:                  npsde\n",
      "noise variance:        [0.28958001 0.31864997]\n",
      "signal variance:       1.0493953926052533\n",
      "lengthscales:          [0.97301814 0.954996  ]\n",
      "diff signal variance:  0.895229121453946\n",
      "diff lengthscales:     [0.82685987]\n",
      "RMSE over the dimensions is: 3.9162657728227326\n",
      "name:                  npsde\n",
      "noise variance:        [0.28958001 0.31864997]\n",
      "signal variance:       1.0493953926052533\n",
      "lengthscales:          [0.97301814 0.954996  ]\n",
      "diff signal variance:  0.895229121453946\n",
      "diff lengthscales:     [0.82685987]\n",
      "RMSE over the dimensions is: 5.778977474660696\n",
      "**** greedy_sum_obs MSEs: [1.8862600200566044, 3.9162657728227326, 5.778977474660696] Avg MSE: 3.8605010891800107 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.36564136 0.33380581]\n",
      "signal variance:       1.0016019082474852\n",
      "lengthscales:          [0.98228447 0.94484393]\n",
      "diff signal variance:  0.8682850818432226\n",
      "diff lengthscales:     [0.90449128]\n",
      "RMSE over the dimensions is: 1.8835956378644427\n",
      "name:                  npsde\n",
      "noise variance:        [0.36564136 0.33380581]\n",
      "signal variance:       1.0016019082474852\n",
      "lengthscales:          [0.98228447 0.94484393]\n",
      "diff signal variance:  0.8682850818432226\n",
      "diff lengthscales:     [0.90449128]\n",
      "RMSE over the dimensions is: 3.6380733601024495\n",
      "name:                  npsde\n",
      "noise variance:        [0.36564136 0.33380581]\n",
      "signal variance:       1.0016019082474852\n",
      "lengthscales:          [0.98228447 0.94484393]\n",
      "diff signal variance:  0.8682850818432226\n",
      "diff lengthscales:     [0.90449128]\n",
      "RMSE over the dimensions is: 5.582340428690979\n",
      "**** dynamic_beta_obs MSEs: [1.8835956378644427, 3.6380733601024495, 5.582340428690979] Avg MSE: 3.701336475552624 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.26467205 0.28521398]\n",
      "signal variance:       1.1122449052235805\n",
      "lengthscales:          [0.97895001 0.97658206]\n",
      "diff signal variance:  0.8529458197565757\n",
      "diff lengthscales:     [0.91671591]\n",
      "RMSE over the dimensions is: 4.55427859985833\n",
      "name:                  npsde\n",
      "noise variance:        [0.26467205 0.28521398]\n",
      "signal variance:       1.1122449052235805\n",
      "lengthscales:          [0.97895001 0.97658206]\n",
      "diff signal variance:  0.8529458197565757\n",
      "diff lengthscales:     [0.91671591]\n",
      "RMSE over the dimensions is: 4.12596764301938\n",
      "name:                  npsde\n",
      "noise variance:        [0.26467205 0.28521398]\n",
      "signal variance:       1.1122449052235805\n",
      "lengthscales:          [0.97895001 0.97658206]\n",
      "diff signal variance:  0.8529458197565757\n",
      "diff lengthscales:     [0.91671591]\n",
      "RMSE over the dimensions is: 5.611719589349503\n",
      "**** joint_obs MSEs: [4.55427859985833, 4.12596764301938, 5.611719589349503] Avg MSE: 4.7639886107424045 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.31221131 0.37734023]\n",
      "signal variance:       0.9964744622198347\n",
      "lengthscales:          [0.94512368 0.94964972]\n",
      "diff signal variance:  0.9122568155837013\n",
      "diff lengthscales:     [0.90405052]\n",
      "RMSE over the dimensions is: 1.4777401199103668\n",
      "name:                  npsde\n",
      "noise variance:        [0.31221131 0.37734023]\n",
      "signal variance:       0.9964744622198347\n",
      "lengthscales:          [0.94512368 0.94964972]\n",
      "diff signal variance:  0.9122568155837013\n",
      "diff lengthscales:     [0.90405052]\n",
      "RMSE over the dimensions is: 2.5590892468007413\n",
      "name:                  npsde\n",
      "noise variance:        [0.31221131 0.37734023]\n",
      "signal variance:       0.9964744622198347\n",
      "lengthscales:          [0.94512368 0.94964972]\n",
      "diff signal variance:  0.9122568155837013\n",
      "diff lengthscales:     [0.90405052]\n",
      "RMSE over the dimensions is: 3.5465794434146707\n",
      "**** rand_obs MSEs: [1.4777401199103668, 2.5590892468007413, 3.5465794434146707] Avg MSE: 2.5278029367085932 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.3713195  0.42399128]\n",
      "signal variance:       1.012821240672298\n",
      "lengthscales:          [0.93168664 0.97824248]\n",
      "diff signal variance:  0.9130676070568078\n",
      "diff lengthscales:     [0.90404079]\n",
      "RMSE over the dimensions is: 4.4576424309071\n",
      "name:                  npsde\n",
      "noise variance:        [0.3713195  0.42399128]\n",
      "signal variance:       1.012821240672298\n",
      "lengthscales:          [0.93168664 0.97824248]\n",
      "diff signal variance:  0.9130676070568078\n",
      "diff lengthscales:     [0.90404079]\n",
      "RMSE over the dimensions is: 4.330893446777028\n",
      "name:                  npsde\n",
      "noise variance:        [0.3713195  0.42399128]\n",
      "signal variance:       1.012821240672298\n",
      "lengthscales:          [0.93168664 0.97824248]\n",
      "diff signal variance:  0.9130676070568078\n",
      "diff lengthscales:     [0.90404079]\n",
      "RMSE over the dimensions is: 2.6271151259490195\n",
      "**** entropy_obs MSEs: [4.4576424309071, 4.330893446777028, 2.6271151259490195] Avg MSE: 3.8052170012110498 ****\n",
      "**** trial index 4 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.31586665 0.23657201]\n",
      "signal variance:       1.0631392381499847\n",
      "lengthscales:          [0.97240095 1.03004017]\n",
      "diff signal variance:  0.9040449583799945\n",
      "diff lengthscales:     [0.94185912]\n",
      "RMSE over the dimensions is: 1.2948294571940437\n",
      "name:                  npsde\n",
      "noise variance:        [0.31586665 0.23657201]\n",
      "signal variance:       1.0631392381499847\n",
      "lengthscales:          [0.97240095 1.03004017]\n",
      "diff signal variance:  0.9040449583799945\n",
      "diff lengthscales:     [0.94185912]\n",
      "RMSE over the dimensions is: 2.4876888400853177\n",
      "name:                  npsde\n",
      "noise variance:        [0.31586665 0.23657201]\n",
      "signal variance:       1.0631392381499847\n",
      "lengthscales:          [0.97240095 1.03004017]\n",
      "diff signal variance:  0.9040449583799945\n",
      "diff lengthscales:     [0.94185912]\n",
      "RMSE over the dimensions is: 5.567451333658423\n",
      "**** greedy_obs_1 MSEs: [1.2948294571940437, 2.4876888400853177, 5.567451333658423] Avg MSE: 3.1166565436459286 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.25947956 0.23508892]\n",
      "signal variance:       1.0768442539421483\n",
      "lengthscales:          [0.99286872 1.00233859]\n",
      "diff signal variance:  0.8906127161804316\n",
      "diff lengthscales:     [0.89455494]\n",
      "RMSE over the dimensions is: 1.4844439985077853\n",
      "name:                  npsde\n",
      "noise variance:        [0.25947956 0.23508892]\n",
      "signal variance:       1.0768442539421483\n",
      "lengthscales:          [0.99286872 1.00233859]\n",
      "diff signal variance:  0.8906127161804316\n",
      "diff lengthscales:     [0.89455494]\n",
      "RMSE over the dimensions is: 5.029718454541945\n",
      "name:                  npsde\n",
      "noise variance:        [0.25947956 0.23508892]\n",
      "signal variance:       1.0768442539421483\n",
      "lengthscales:          [0.99286872 1.00233859]\n",
      "diff signal variance:  0.8906127161804316\n",
      "diff lengthscales:     [0.89455494]\n",
      "RMSE over the dimensions is: 5.477704251460688\n",
      "**** greedy_obs_2 MSEs: [1.4844439985077853, 5.029718454541945, 5.477704251460688] Avg MSE: 3.9972889015034725 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.30234864 0.25267731]\n",
      "signal variance:       1.064161226478547\n",
      "lengthscales:          [0.98033565 1.0862334 ]\n",
      "diff signal variance:  0.8984846858525904\n",
      "diff lengthscales:     [0.93914007]\n",
      "RMSE over the dimensions is: 2.1752160222346886\n",
      "name:                  npsde\n",
      "noise variance:        [0.30234864 0.25267731]\n",
      "signal variance:       1.064161226478547\n",
      "lengthscales:          [0.98033565 1.0862334 ]\n",
      "diff signal variance:  0.8984846858525904\n",
      "diff lengthscales:     [0.93914007]\n",
      "RMSE over the dimensions is: 3.853646194984521\n",
      "name:                  npsde\n",
      "noise variance:        [0.30234864 0.25267731]\n",
      "signal variance:       1.064161226478547\n",
      "lengthscales:          [0.98033565 1.0862334 ]\n",
      "diff signal variance:  0.8984846858525904\n",
      "diff lengthscales:     [0.93914007]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 5.574434913676644\n",
      "**** greedy_obs_3 MSEs: [2.1752160222346886, 3.853646194984521, 5.574434913676644] Avg MSE: 3.867765710298618 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.37924809 0.28931748]\n",
      "signal variance:       1.0527942617252333\n",
      "lengthscales:          [0.97798583 1.07208792]\n",
      "diff signal variance:  0.901964461684449\n",
      "diff lengthscales:     [0.92870097]\n",
      "RMSE over the dimensions is: 3.9039624129638466\n",
      "name:                  npsde\n",
      "noise variance:        [0.37924809 0.28931748]\n",
      "signal variance:       1.0527942617252333\n",
      "lengthscales:          [0.97798583 1.07208792]\n",
      "diff signal variance:  0.901964461684449\n",
      "diff lengthscales:     [0.92870097]\n",
      "RMSE over the dimensions is: 2.3784628341723892\n",
      "name:                  npsde\n",
      "noise variance:        [0.37924809 0.28931748]\n",
      "signal variance:       1.0527942617252333\n",
      "lengthscales:          [0.97798583 1.07208792]\n",
      "diff signal variance:  0.901964461684449\n",
      "diff lengthscales:     [0.92870097]\n",
      "RMSE over the dimensions is: 5.706474087001787\n",
      "**** greedy_obs_4 MSEs: [3.9039624129638466, 2.3784628341723892, 5.706474087001787] Avg MSE: 3.9962997780460072 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.29103029 0.2492158 ]\n",
      "signal variance:       1.0603603986755608\n",
      "lengthscales:          [0.97195122 1.05409362]\n",
      "diff signal variance:  0.9129509088126603\n",
      "diff lengthscales:     [0.93035488]\n",
      "RMSE over the dimensions is: 1.4334227816767913\n",
      "name:                  npsde\n",
      "noise variance:        [0.29103029 0.2492158 ]\n",
      "signal variance:       1.0603603986755608\n",
      "lengthscales:          [0.97195122 1.05409362]\n",
      "diff signal variance:  0.9129509088126603\n",
      "diff lengthscales:     [0.93035488]\n",
      "RMSE over the dimensions is: 1.4440884530002591\n",
      "name:                  npsde\n",
      "noise variance:        [0.29103029 0.2492158 ]\n",
      "signal variance:       1.0603603986755608\n",
      "lengthscales:          [0.97195122 1.05409362]\n",
      "diff signal variance:  0.9129509088126603\n",
      "diff lengthscales:     [0.93035488]\n",
      "RMSE over the dimensions is: 5.589315402117044\n",
      "**** greedy_sum_obs MSEs: [1.4334227816767913, 1.4440884530002591, 5.589315402117044] Avg MSE: 2.8222755455980315 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.2636935  0.21829031]\n",
      "signal variance:       1.111580468796749\n",
      "lengthscales:          [0.97448388 1.01671911]\n",
      "diff signal variance:  0.8911571777919253\n",
      "diff lengthscales:     [0.90656495]\n",
      "RMSE over the dimensions is: 1.3683429054025158\n",
      "name:                  npsde\n",
      "noise variance:        [0.2636935  0.21829031]\n",
      "signal variance:       1.111580468796749\n",
      "lengthscales:          [0.97448388 1.01671911]\n",
      "diff signal variance:  0.8911571777919253\n",
      "diff lengthscales:     [0.90656495]\n",
      "RMSE over the dimensions is: 3.3205586063073094\n",
      "name:                  npsde\n",
      "noise variance:        [0.2636935  0.21829031]\n",
      "signal variance:       1.111580468796749\n",
      "lengthscales:          [0.97448388 1.01671911]\n",
      "diff signal variance:  0.8911571777919253\n",
      "diff lengthscales:     [0.90656495]\n",
      "RMSE over the dimensions is: 5.516102015650242\n",
      "**** dynamic_beta_obs MSEs: [1.3683429054025158, 3.3205586063073094, 5.516102015650242] Avg MSE: 3.4016678424533553 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.46384484 0.30355221]\n",
      "signal variance:       0.9166913005312602\n",
      "lengthscales:          [0.98964753 0.9971699 ]\n",
      "diff signal variance:  1.026737525331274\n",
      "diff lengthscales:     [1.00446588]\n",
      "RMSE over the dimensions is: 0.8454079174503026\n",
      "name:                  npsde\n",
      "noise variance:        [0.46384484 0.30355221]\n",
      "signal variance:       0.9166913005312602\n",
      "lengthscales:          [0.98964753 0.9971699 ]\n",
      "diff signal variance:  1.026737525331274\n",
      "diff lengthscales:     [1.00446588]\n",
      "RMSE over the dimensions is: 19.51113572605341\n",
      "name:                  npsde\n",
      "noise variance:        [0.46384484 0.30355221]\n",
      "signal variance:       0.9166913005312602\n",
      "lengthscales:          [0.98964753 0.9971699 ]\n",
      "diff signal variance:  1.026737525331274\n",
      "diff lengthscales:     [1.00446588]\n",
      "RMSE over the dimensions is: 5.1647216463036125\n",
      "**** joint_obs MSEs: [0.8454079174503026, 19.51113572605341, 5.1647216463036125] Avg MSE: 8.507088429935775 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21152    0.21261237]\n",
      "signal variance:       1.2412632683532248\n",
      "lengthscales:          [1.06043996 1.11968605]\n",
      "diff signal variance:  0.8598871896898969\n",
      "diff lengthscales:     [0.96209487]\n",
      "RMSE over the dimensions is: 2.4756183413369635\n",
      "name:                  npsde\n",
      "noise variance:        [0.21152    0.21261237]\n",
      "signal variance:       1.2412632683532248\n",
      "lengthscales:          [1.06043996 1.11968605]\n",
      "diff signal variance:  0.8598871896898969\n",
      "diff lengthscales:     [0.96209487]\n",
      "RMSE over the dimensions is: 5.037075537332197\n",
      "name:                  npsde\n",
      "noise variance:        [0.21152    0.21261237]\n",
      "signal variance:       1.2412632683532248\n",
      "lengthscales:          [1.06043996 1.11968605]\n",
      "diff signal variance:  0.8598871896898969\n",
      "diff lengthscales:     [0.96209487]\n",
      "RMSE over the dimensions is: 5.530625965655741\n",
      "**** rand_obs MSEs: [2.4756183413369635, 5.037075537332197, 5.530625965655741] Avg MSE: 4.347773281441634 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.3208733  0.22656095]\n",
      "signal variance:       1.217216366037427\n",
      "lengthscales:          [0.95599974 1.16689972]\n",
      "diff signal variance:  0.8571701706738243\n",
      "diff lengthscales:     [0.89393564]\n",
      "RMSE over the dimensions is: 2.438476112528729\n",
      "name:                  npsde\n",
      "noise variance:        [0.3208733  0.22656095]\n",
      "signal variance:       1.217216366037427\n",
      "lengthscales:          [0.95599974 1.16689972]\n",
      "diff signal variance:  0.8571701706738243\n",
      "diff lengthscales:     [0.89393564]\n",
      "RMSE over the dimensions is: 3.577536412795877\n",
      "name:                  npsde\n",
      "noise variance:        [0.3208733  0.22656095]\n",
      "signal variance:       1.217216366037427\n",
      "lengthscales:          [0.95599974 1.16689972]\n",
      "diff signal variance:  0.8571701706738243\n",
      "diff lengthscales:     [0.89393564]\n",
      "RMSE over the dimensions is: 2.8017639251073163\n",
      "**** entropy_obs MSEs: [2.438476112528729, 3.577536412795877, 2.8017639251073163] Avg MSE: 2.9392588168106406 ****\n",
      "**** trial index 5 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.2327096  0.23431887]\n",
      "signal variance:       1.0424569690214485\n",
      "lengthscales:          [0.98459676 1.08246447]\n",
      "diff signal variance:  0.9126663280814592\n",
      "diff lengthscales:     [0.936379]\n",
      "RMSE over the dimensions is: 4.494106614129199\n",
      "name:                  npsde\n",
      "noise variance:        [0.2327096  0.23431887]\n",
      "signal variance:       1.0424569690214485\n",
      "lengthscales:          [0.98459676 1.08246447]\n",
      "diff signal variance:  0.9126663280814592\n",
      "diff lengthscales:     [0.936379]\n",
      "RMSE over the dimensions is: 2.0943330872399697\n",
      "name:                  npsde\n",
      "noise variance:        [0.2327096  0.23431887]\n",
      "signal variance:       1.0424569690214485\n",
      "lengthscales:          [0.98459676 1.08246447]\n",
      "diff signal variance:  0.9126663280814592\n",
      "diff lengthscales:     [0.936379]\n",
      "RMSE over the dimensions is: 5.80130404724847\n",
      "**** greedy_obs_1 MSEs: [4.494106614129199, 2.0943330872399697, 5.80130404724847] Avg MSE: 4.129914582872547 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.22666727 0.22609526]\n",
      "signal variance:       1.0534070303713188\n",
      "lengthscales:          [0.98333247 1.18563279]\n",
      "diff signal variance:  0.9081762087603383\n",
      "diff lengthscales:     [0.93018924]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 4.516787138281382\n",
      "name:                  npsde\n",
      "noise variance:        [0.22666727 0.22609526]\n",
      "signal variance:       1.0534070303713188\n",
      "lengthscales:          [0.98333247 1.18563279]\n",
      "diff signal variance:  0.9081762087603383\n",
      "diff lengthscales:     [0.93018924]\n",
      "RMSE over the dimensions is: 2.563893611961305\n",
      "name:                  npsde\n",
      "noise variance:        [0.22666727 0.22609526]\n",
      "signal variance:       1.0534070303713188\n",
      "lengthscales:          [0.98333247 1.18563279]\n",
      "diff signal variance:  0.9081762087603383\n",
      "diff lengthscales:     [0.93018924]\n",
      "RMSE over the dimensions is: 5.523109053964598\n",
      "**** greedy_obs_2 MSEs: [4.516787138281382, 2.563893611961305, 5.523109053964598] Avg MSE: 4.201263268069095 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.20700499 0.20856676]\n",
      "signal variance:       1.1252748549023353\n",
      "lengthscales:          [0.96814028 1.16878999]\n",
      "diff signal variance:  0.8783338550923392\n",
      "diff lengthscales:     [0.89316548]\n",
      "RMSE over the dimensions is: 4.46682684440744\n",
      "name:                  npsde\n",
      "noise variance:        [0.20700499 0.20856676]\n",
      "signal variance:       1.1252748549023353\n",
      "lengthscales:          [0.96814028 1.16878999]\n",
      "diff signal variance:  0.8783338550923392\n",
      "diff lengthscales:     [0.89316548]\n",
      "RMSE over the dimensions is: 2.4023331740922123\n",
      "name:                  npsde\n",
      "noise variance:        [0.20700499 0.20856676]\n",
      "signal variance:       1.1252748549023353\n",
      "lengthscales:          [0.96814028 1.16878999]\n",
      "diff signal variance:  0.8783338550923392\n",
      "diff lengthscales:     [0.89316548]\n",
      "RMSE over the dimensions is: 5.6401787145279965\n",
      "**** greedy_obs_3 MSEs: [4.46682684440744, 2.4023331740922123, 5.6401787145279965] Avg MSE: 4.169779577675883 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21967694 0.22667691]\n",
      "signal variance:       1.0443692774680156\n",
      "lengthscales:          [0.92933876 1.11158683]\n",
      "diff signal variance:  0.9006970515636502\n",
      "diff lengthscales:     [0.94058741]\n",
      "RMSE over the dimensions is: 4.793669504809245\n",
      "name:                  npsde\n",
      "noise variance:        [0.21967694 0.22667691]\n",
      "signal variance:       1.0443692774680156\n",
      "lengthscales:          [0.92933876 1.11158683]\n",
      "diff signal variance:  0.9006970515636502\n",
      "diff lengthscales:     [0.94058741]\n",
      "RMSE over the dimensions is: 3.012583876776012\n",
      "name:                  npsde\n",
      "noise variance:        [0.21967694 0.22667691]\n",
      "signal variance:       1.0443692774680156\n",
      "lengthscales:          [0.92933876 1.11158683]\n",
      "diff signal variance:  0.9006970515636502\n",
      "diff lengthscales:     [0.94058741]\n",
      "RMSE over the dimensions is: 4.703343297543731\n",
      "**** greedy_obs_4 MSEs: [4.793669504809245, 3.012583876776012, 4.703343297543731] Avg MSE: 4.169865559709662 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.2136696  0.21281645]\n",
      "signal variance:       1.2489845530355632\n",
      "lengthscales:          [1.0015429  1.06996734]\n",
      "diff signal variance:  0.8063622994251127\n",
      "diff lengthscales:     [0.91361652]\n",
      "RMSE over the dimensions is: 3.9621754653047887\n",
      "name:                  npsde\n",
      "noise variance:        [0.2136696  0.21281645]\n",
      "signal variance:       1.2489845530355632\n",
      "lengthscales:          [1.0015429  1.06996734]\n",
      "diff signal variance:  0.8063622994251127\n",
      "diff lengthscales:     [0.91361652]\n",
      "RMSE over the dimensions is: 2.865779493798144\n",
      "name:                  npsde\n",
      "noise variance:        [0.2136696  0.21281645]\n",
      "signal variance:       1.2489845530355632\n",
      "lengthscales:          [1.0015429  1.06996734]\n",
      "diff signal variance:  0.8063622994251127\n",
      "diff lengthscales:     [0.91361652]\n",
      "RMSE over the dimensions is: 4.960114907337439\n",
      "**** greedy_sum_obs MSEs: [3.9621754653047887, 2.865779493798144, 4.960114907337439] Avg MSE: 3.92935662214679 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.20587352 0.20636785]\n",
      "signal variance:       1.124123374975039\n",
      "lengthscales:          [1.00480704 1.11457941]\n",
      "diff signal variance:  0.8457641374205185\n",
      "diff lengthscales:     [0.89365682]\n",
      "RMSE over the dimensions is: 4.430388157457864\n",
      "name:                  npsde\n",
      "noise variance:        [0.20587352 0.20636785]\n",
      "signal variance:       1.124123374975039\n",
      "lengthscales:          [1.00480704 1.11457941]\n",
      "diff signal variance:  0.8457641374205185\n",
      "diff lengthscales:     [0.89365682]\n",
      "RMSE over the dimensions is: 2.471906799321776\n",
      "name:                  npsde\n",
      "noise variance:        [0.20587352 0.20636785]\n",
      "signal variance:       1.124123374975039\n",
      "lengthscales:          [1.00480704 1.11457941]\n",
      "diff signal variance:  0.8457641374205185\n",
      "diff lengthscales:     [0.89365682]\n",
      "RMSE over the dimensions is: 5.268602359497146\n",
      "**** dynamic_beta_obs MSEs: [4.430388157457864, 2.471906799321776, 5.268602359497146] Avg MSE: 4.056965772092262 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.2261563  0.23203902]\n",
      "signal variance:       1.034864065278786\n",
      "lengthscales:          [0.99269104 1.12650685]\n",
      "diff signal variance:  0.9017543625438613\n",
      "diff lengthscales:     [0.92166591]\n",
      "RMSE over the dimensions is: 4.4893426421978715\n",
      "name:                  npsde\n",
      "noise variance:        [0.2261563  0.23203902]\n",
      "signal variance:       1.034864065278786\n",
      "lengthscales:          [0.99269104 1.12650685]\n",
      "diff signal variance:  0.9017543625438613\n",
      "diff lengthscales:     [0.92166591]\n",
      "RMSE over the dimensions is: 2.5308548094227037\n",
      "name:                  npsde\n",
      "noise variance:        [0.2261563  0.23203902]\n",
      "signal variance:       1.034864065278786\n",
      "lengthscales:          [0.99269104 1.12650685]\n",
      "diff signal variance:  0.9017543625438613\n",
      "diff lengthscales:     [0.92166591]\n",
      "RMSE over the dimensions is: 5.410478244493175\n",
      "**** joint_obs MSEs: [4.4893426421978715, 2.5308548094227037, 5.410478244493175] Avg MSE: 4.14355856537125 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.21211275 0.21415108]\n",
      "signal variance:       1.1164535684099994\n",
      "lengthscales:          [0.95538238 1.20317678]\n",
      "diff signal variance:  0.894982999118357\n",
      "diff lengthscales:     [0.90436158]\n",
      "RMSE over the dimensions is: 3.974530994283291\n",
      "name:                  npsde\n",
      "noise variance:        [0.21211275 0.21415108]\n",
      "signal variance:       1.1164535684099994\n",
      "lengthscales:          [0.95538238 1.20317678]\n",
      "diff signal variance:  0.894982999118357\n",
      "diff lengthscales:     [0.90436158]\n",
      "RMSE over the dimensions is: 3.071304769114967\n",
      "name:                  npsde\n",
      "noise variance:        [0.21211275 0.21415108]\n",
      "signal variance:       1.1164535684099994\n",
      "lengthscales:          [0.95538238 1.20317678]\n",
      "diff signal variance:  0.894982999118357\n",
      "diff lengthscales:     [0.90436158]\n",
      "RMSE over the dimensions is: 4.962796759807645\n",
      "**** rand_obs MSEs: [3.974530994283291, 3.071304769114967, 4.962796759807645] Avg MSE: 4.002877507735302 ****\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npsde\n",
      "noise variance:        [0.20884595 0.22924298]\n",
      "signal variance:       1.1908735498131662\n",
      "lengthscales:          [0.98995211 1.24006819]\n",
      "diff signal variance:  0.7624261714358949\n",
      "diff lengthscales:     [0.99855377]\n",
      "RMSE over the dimensions is: 4.951289621187042\n",
      "name:                  npsde\n",
      "noise variance:        [0.20884595 0.22924298]\n",
      "signal variance:       1.1908735498131662\n",
      "lengthscales:          [0.98995211 1.24006819]\n",
      "diff signal variance:  0.7624261714358949\n",
      "diff lengthscales:     [0.99855377]\n",
      "RMSE over the dimensions is: 3.5623577652919303\n",
      "name:                  npsde\n",
      "noise variance:        [0.20884595 0.22924298]\n",
      "signal variance:       1.1908735498131662\n",
      "lengthscales:          [0.98995211 1.24006819]\n",
      "diff signal variance:  0.7624261714358949\n",
      "diff lengthscales:     [0.99855377]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 5.077343687171046\n",
      "**** entropy_obs MSEs: [4.951289621187042, 3.5623577652919303, 5.077343687171046] Avg MSE: 4.530330357883339 ****\n"
     ]
    }
   ],
   "source": [
    "n_trials = 5\n",
    "n = 3\n",
    "de_type = 'SDE'\n",
    "trajectory_index = 2\n",
    "\n",
    "results_dir = oj('results', 'DE')\n",
    "\n",
    "with cwd(results_dir):\n",
    "    result_datas = []\n",
    "    for file in os.listdir():\n",
    "        if file.endswith('.npz') and de_type.upper() in file:\n",
    "            result_datas.append(np.load(file, allow_pickle=True))\n",
    "\n",
    "    mse_results = defaultdict(list)\n",
    "\n",
    "    for trial_i in range(n_trials):\n",
    "        print(f'**** trial index {trial_i+1} **** ')\n",
    "        print()\n",
    "        obs = result_datas[trial_i]['obs'].item()\n",
    "        times = result_datas[trial_i]['times']\n",
    "\n",
    "        overall_t = result_datas[trial_i]['t'][trajectory_index]\n",
    "        overall_Y = result_datas[trial_i]['Y'][trajectory_index]\n",
    "\n",
    "        Ts = result_datas[trial_i]['Ts']\n",
    "        for collab_type, baseline_obs in obs.items():   \n",
    "                \n",
    "                if 'indiv' in collab_type:continue \n",
    "                try:\n",
    "                    mses = get_mses(collab_type, n, overall_t, overall_Y, baseline_obs, Ts, de_type)\n",
    "                    mse = np.mean(mses)\n",
    "\n",
    "                    mse_results[collab_type+'-avg-mses'].append(mse)        \n",
    "                    mse_results[collab_type+'-mses'].append(mses)\n",
    "\n",
    "                    print(f'**** {collab_type} MSEs: {mses} Avg MSE: {mse} ****')\n",
    "                except Exception as inst:\n",
    "                    print(type(inst))    # the exception instance\n",
    "                    print(inst.args)     # arguments stored in .args\n",
    "                    print(inst)          # __str__ allows args to be printed directly,\n",
    "                                         # but may be overridden in exception subclasses\n",
    "\n",
    "    regression_df = get_regression_df(mse_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "23036292",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>greedy_1</td>\n",
       "      <td>3.961355</td>\n",
       "      <td>0.415532</td>\n",
       "      <td>1.509374</td>\n",
       "      <td>0.133572</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>greedy_2</td>\n",
       "      <td>5.113194</td>\n",
       "      <td>0.781788</td>\n",
       "      <td>2.020337</td>\n",
       "      <td>0.742366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>greedy_3</td>\n",
       "      <td>4.138763</td>\n",
       "      <td>0.298056</td>\n",
       "      <td>1.320298</td>\n",
       "      <td>0.130503</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>greedy_4</td>\n",
       "      <td>4.198453</td>\n",
       "      <td>0.212938</td>\n",
       "      <td>1.214258</td>\n",
       "      <td>0.234954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>greedy_sum</td>\n",
       "      <td>4.096240</td>\n",
       "      <td>0.397469</td>\n",
       "      <td>1.898504</td>\n",
       "      <td>0.499957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>dynamic_beta</td>\n",
       "      <td>3.792370</td>\n",
       "      <td>0.289529</td>\n",
       "      <td>1.553035</td>\n",
       "      <td>0.140785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>joint</td>\n",
       "      <td>5.679079</td>\n",
       "      <td>0.771345</td>\n",
       "      <td>3.095353</td>\n",
       "      <td>1.374249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>rand</td>\n",
       "      <td>3.836869</td>\n",
       "      <td>0.411879</td>\n",
       "      <td>1.220882</td>\n",
       "      <td>0.177586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>entropy</td>\n",
       "      <td>3.361926</td>\n",
       "      <td>0.455779</td>\n",
       "      <td>0.692293</td>\n",
       "      <td>0.205279</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines   Avg MSE    Stderr   Std MSE  Stderr Std\n",
       "0      greedy_1  3.961355  0.415532  1.509374    0.133572\n",
       "1      greedy_2  5.113194  0.781788  2.020337    0.742366\n",
       "2      greedy_3  4.138763  0.298056  1.320298    0.130503\n",
       "3      greedy_4  4.198453  0.212938  1.214258    0.234954\n",
       "4    greedy_sum  4.096240  0.397469  1.898504    0.499957\n",
       "5  dynamic_beta  3.792370  0.289529  1.553035    0.140785\n",
       "6         joint  5.679079  0.771345  3.095353    1.374249\n",
       "7          rand  3.836869  0.411879  1.220882    0.177586\n",
       "8       entropy  3.361926  0.455779  0.692293    0.205279"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regression_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d80ccd1d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5a5a6486",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ffae4f00",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "**** trial index 1 **** \n",
      "\n",
      "Model being initialized...\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde.py:83: The name tf.cholesky is deprecated. Please use tf.linalg.cholesky instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde.py:87: The name tf.matrix_triangular_solve is deprecated. Please use tf.linalg.triangular_solve instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde_helper.py:95: The name tf.global_variables_initializer is deprecated. Please use tf.compat.v1.global_variables_initializer instead.\n",
      "\n",
      "Building loss function...\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde_helper.py:135: The name tf.diag is deprecated. Please use tf.linalg.tensor_diag instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde_helper.py:135: MultivariateNormalFullCovariance.__init__ (from tensorflow.contrib.distributions.python.ops.mvn_full_covariance) is deprecated and will be removed after 2018-10-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.contrib.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/contrib/distributions/python/ops/mvn_full_covariance.py:195: MultivariateNormalTriL.__init__ (from tensorflow.contrib.distributions.python.ops.mvn_tril) is deprecated and will be removed after 2018-10-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.contrib.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/contrib/distributions/python/ops/mvn_tril.py:222: MultivariateNormalLinearOperator.__init__ (from tensorflow.contrib.distributions.python.ops.mvn_linear_operator) is deprecated and will be removed after 2018-10-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.contrib.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/contrib/distributions/python/ops/mvn_linear_operator.py:199: Normal.__init__ (from tensorflow.python.ops.distributions.normal) is deprecated and will be removed after 2019-01-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/python/ops/distributions/normal.py:160: Distribution.__init__ (from tensorflow.python.ops.distributions.distribution) is deprecated and will be removed after 2019-01-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/contrib/distributions/python/ops/mvn_linear_operator.py:201: AffineLinearOperator.__init__ (from tensorflow.contrib.distributions.python.ops.bijectors.affine_linear_operator) is deprecated and will be removed after 2018-10-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.contrib.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/contrib/distributions/python/ops/bijectors/affine_linear_operator.py:158: _DistributionShape.__init__ (from tensorflow.contrib.distributions.python.ops.shape) is deprecated and will be removed after 2018-10-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.contrib.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/contrib/distributions/python/ops/mvn_linear_operator.py:205: TransformedDistribution.__init__ (from tensorflow.python.ops.distributions.transformed_distribution) is deprecated and will be removed after 2019-01-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.distributions`.\n",
      "WARNING:tensorflow:From /home/fengzi/anaconda3/envs/CAL-TF/lib/python3.6/site-packages/tensorflow_core/python/ops/math_ops.py:2509: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde.py:105: MultivariateNormalDiag.__init__ (from tensorflow.contrib.distributions.python.ops.mvn_diag) is deprecated and will be removed after 2018-10-01.\n",
      "Instructions for updating:\n",
      "The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.contrib.distributions`.\n",
      "Adam optimizer being initialized...\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde_helper.py:159: The name tf.train.exponential_decay is deprecated. Please use tf.compat.v1.train.exponential_decay instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/fengzi/Downloads/codebase/collaborative-active-learning/npde_master/npde_helper.py:160: The name tf.train.AdamOptimizer is deprecated. Please use tf.compat.v1.train.AdamOptimizer instead.\n",
      "\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00031422 0.01822301]\n",
      "signal variance:       3.0334046409120607\n",
      "lengthscales:          [0.94935922 1.32759265]\n",
      "RMSE over the dimensions is: 4.042510931096123\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00626058 0.00870978]\n",
      "signal variance:       2.257332460304492\n",
      "lengthscales:          [0.95269059 1.04950168]\n",
      "RMSE over the dimensions is: 7.004897877741255\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.01006496 0.00073809]\n",
      "signal variance:       2.3847369721498484\n",
      "lengthscales:          [0.76336902 1.03753903]\n",
      "RMSE over the dimensions is: 11.044435793342352\n",
      "**** indiv_greedy_obs MSEs: [4.042510931096123, 7.004897877741255, 11.044435793342352] Avg MSE: 7.363948200726576 ****\n",
      "**** trial index 2 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00567648 0.00135132]\n",
      "signal variance:       5.148087465216208\n",
      "lengthscales:          [1.29186751 1.10661647]\n",
      "RMSE over the dimensions is: 3.2090181592693146\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00442563 0.00352321]\n",
      "signal variance:       2.4652604146583434\n",
      "lengthscales:          [0.73773403 1.33580795]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE over the dimensions is: 8.265091394012495\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00081491 0.00385079]\n",
      "signal variance:       1.9007579267037944\n",
      "lengthscales:          [0.98594686 0.71797308]\n",
      "RMSE over the dimensions is: 10.750136520276772\n",
      "**** indiv_greedy_obs MSEs: [3.2090181592693146, 8.265091394012495, 10.750136520276772] Avg MSE: 7.408082024519527 ****\n",
      "**** trial index 3 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00556034 0.0033205 ]\n",
      "signal variance:       2.8660508982570696\n",
      "lengthscales:          [0.81520525 1.23615604]\n",
      "RMSE over the dimensions is: 3.7939577818486914\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00462282 0.00148981]\n",
      "signal variance:       2.558046928438533\n",
      "lengthscales:          [0.97755583 0.45113333]\n",
      "RMSE over the dimensions is: 5.567758228896983\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.01251133 0.01118068]\n",
      "signal variance:       1.760128476516213\n",
      "lengthscales:          [0.94723843 0.72402322]\n",
      "RMSE over the dimensions is: 10.630913816087624\n",
      "**** indiv_greedy_obs MSEs: [3.7939577818486914, 5.567758228896983, 10.630913816087624] Avg MSE: 6.664209942277766 ****\n",
      "**** trial index 4 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.0004016  0.00172953]\n",
      "signal variance:       5.325246871260713\n",
      "lengthscales:          [0.5573448  0.93855836]\n",
      "RMSE over the dimensions is: 4.394643220130628\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00334392 0.01910298]\n",
      "signal variance:       2.2200858061604465\n",
      "lengthscales:          [0.6618611  1.44182141]\n",
      "RMSE over the dimensions is: 8.368799096561261\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.01126923 0.00891699]\n",
      "signal variance:       1.7364342639192223\n",
      "lengthscales:          [1.10911949 0.88856991]\n",
      "RMSE over the dimensions is: 10.811629094213075\n",
      "**** indiv_greedy_obs MSEs: [4.394643220130628, 8.368799096561261, 10.811629094213075] Avg MSE: 7.858357136968322 ****\n",
      "**** trial index 5 **** \n",
      "\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00069826 0.01492975]\n",
      "signal variance:       4.018087720008793\n",
      "lengthscales:          [0.75631038 1.21031036]\n",
      "RMSE over the dimensions is: 5.531001000887238\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00058731 0.00310157]\n",
      "signal variance:       2.19760622501263\n",
      "lengthscales:          [0.60060364 1.21572585]\n",
      "RMSE over the dimensions is: 4.660090103096185\n",
      "Model being initialized...\n",
      "Building loss function...\n",
      "Adam optimizer being initialized...\n",
      "Optimization starts.\n",
      "Optimization ends.\n",
      "name:                  npode\n",
      "noise variance:        [0.00396563 0.00290032]\n",
      "signal variance:       1.7210353176622575\n",
      "lengthscales:          [0.93170573 0.92599737]\n",
      "RMSE over the dimensions is: 10.97789841000935\n",
      "**** indiv_greedy_obs MSEs: [5.531001000887238, 4.660090103096185, 10.97789841000935] Avg MSE: 7.056329837997591 ****\n"
     ]
    }
   ],
   "source": [
    "n_trials = 5\n",
    "n = 3\n",
    "de_type = 'ODE'\n",
    "trajectory_index = 2\n",
    "\n",
    "results_dir = oj('results', 'DE')\n",
    "\n",
    "with cwd(results_dir):\n",
    "    result_datas = []\n",
    "    for file in os.listdir():\n",
    "        if file.endswith('.npz') and de_type.upper() in file:\n",
    "            result_datas.append(np.load(file, allow_pickle=True))\n",
    "\n",
    "    mse_results = defaultdict(list)\n",
    "\n",
    "    for trial_i in range(n_trials):\n",
    "        print(f'**** trial index {trial_i+1} **** ')\n",
    "        print()\n",
    "        obs = result_datas[trial_i]['obs'].item()\n",
    "        times = result_datas[trial_i]['times']\n",
    "\n",
    "        overall_t = result_datas[trial_i]['t'][trajectory_index]\n",
    "        overall_Y = result_datas[trial_i]['Y'][trajectory_index]\n",
    "\n",
    "        Ts = result_datas[trial_i]['Ts']\n",
    "        for collab_type, baseline_obs in obs.items():   \n",
    "                \n",
    "            if 'indiv' in collab_type: \n",
    "                try:\n",
    "                    mses = get_mses(collab_type, n, overall_t, overall_Y, baseline_obs, Ts, de_type)\n",
    "                    mse = np.mean(mses)\n",
    "\n",
    "                    mse_results[collab_type+'-avg-mses'].append(mse)        \n",
    "                    mse_results[collab_type+'-mses'].append(mses)\n",
    "\n",
    "                    print(f'**** {collab_type} MSEs: {mses} Avg MSE: {mse} ****')\n",
    "                except Exception as inst:\n",
    "                    print(type(inst))    # the exception instance\n",
    "                    print(inst.args)     # arguments stored in .args\n",
    "                    print(inst)          # __str__ allows args to be printed directly,\n",
    "                                         # but may be overridden in exception subclasses\n",
    "\n",
    "    ode_ind_regression_df = get_regression_df(mse_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a173f356",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>indiv_greedy</td>\n",
       "      <td>7.270185</td>\n",
       "      <td>0.19833</td>\n",
       "      <td>2.868899</td>\n",
       "      <td>0.080243</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines   Avg MSE   Stderr   Std MSE  Stderr Std\n",
       "0  indiv_greedy  7.270185  0.19833  2.868899    0.080243"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ode_ind_regression_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a503a853",
   "metadata": {},
   "outputs": [],
   "source": [
    "with cwd(oj(results_dir, 'regression_results')):\n",
    "    ode_ind_regression_df.to_csv('ODE-regression-correction.csv', index=False)\n",
    "    ode_ind_regression_df.to_latex('ODE-regression-correction.tex', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "12f45d0b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b08ed3db",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e6f7d21d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "be293bc6",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8b99e448",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>entropy</td>\n",
       "      <td>2.981224</td>\n",
       "      <td>0.465341</td>\n",
       "      <td>1.157018</td>\n",
       "      <td>0.036469</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Baselines   Avg MSE    Stderr   Std MSE  Stderr Std\n",
       "0   entropy  2.981224  0.465341  1.157018    0.036469"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regression_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fadc90c1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "21374a41",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>entropy</td>\n",
       "      <td>2.864518</td>\n",
       "      <td>0.430831</td>\n",
       "      <td>1.104927</td>\n",
       "      <td>0.051972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>indiv_greedy</td>\n",
       "      <td>5.179316</td>\n",
       "      <td>0.802955</td>\n",
       "      <td>2.613171</td>\n",
       "      <td>0.137215</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines   Avg MSE    Stderr   Std MSE  Stderr Std\n",
       "0       entropy  2.864518  0.430831  1.104927    0.051972\n",
       "1  indiv_greedy  5.179316  0.802955  2.613171    0.137215"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regression_df # training iter 1000, for entropy and indi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7920d77b",
   "metadata": {},
   "outputs": [],
   "source": [
    "with cwd(oj(results_dir, 'regression_results')):\n",
    "    regression_df.to_csv('SDE-regression-correction.csv', index=False)\n",
    "    regression_df.to_latex('SDE-regression-correction.tex', index=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5d3aa393",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9541a3ce",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a97ea9a2",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>greedy_1</td>\n",
       "      <td>5.857746</td>\n",
       "      <td>0.397743</td>\n",
       "      <td>2.359273</td>\n",
       "      <td>0.266774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>greedy_2</td>\n",
       "      <td>6.725795</td>\n",
       "      <td>0.812513</td>\n",
       "      <td>3.805229</td>\n",
       "      <td>1.075850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>greedy_3</td>\n",
       "      <td>6.179537</td>\n",
       "      <td>0.635290</td>\n",
       "      <td>2.072478</td>\n",
       "      <td>0.665395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>greedy_4</td>\n",
       "      <td>5.476039</td>\n",
       "      <td>0.330058</td>\n",
       "      <td>1.987787</td>\n",
       "      <td>0.302376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>greedy_sum</td>\n",
       "      <td>6.212020</td>\n",
       "      <td>0.661704</td>\n",
       "      <td>2.767850</td>\n",
       "      <td>1.249287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>dynamic_beta</td>\n",
       "      <td>5.758070</td>\n",
       "      <td>0.424243</td>\n",
       "      <td>2.077605</td>\n",
       "      <td>0.373631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>joint</td>\n",
       "      <td>5.339084</td>\n",
       "      <td>0.474563</td>\n",
       "      <td>2.125629</td>\n",
       "      <td>0.436403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>rand</td>\n",
       "      <td>6.239905</td>\n",
       "      <td>0.495719</td>\n",
       "      <td>2.178888</td>\n",
       "      <td>0.432880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>entropy</td>\n",
       "      <td>5.171592</td>\n",
       "      <td>0.156638</td>\n",
       "      <td>1.920965</td>\n",
       "      <td>0.151457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>indiv_greedy</td>\n",
       "      <td>6.484443</td>\n",
       "      <td>0.374912</td>\n",
       "      <td>1.281067</td>\n",
       "      <td>0.136480</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines   Avg MSE    Stderr   Std MSE  Stderr Std\n",
       "0      greedy_1  5.857746  0.397743  2.359273    0.266774\n",
       "1      greedy_2  6.725795  0.812513  3.805229    1.075850\n",
       "2      greedy_3  6.179537  0.635290  2.072478    0.665395\n",
       "3      greedy_4  5.476039  0.330058  1.987787    0.302376\n",
       "4    greedy_sum  6.212020  0.661704  2.767850    1.249287\n",
       "5  dynamic_beta  5.758070  0.424243  2.077605    0.373631\n",
       "6         joint  5.339084  0.474563  2.125629    0.436403\n",
       "7          rand  6.239905  0.495719  2.178888    0.432880\n",
       "8       entropy  5.171592  0.156638  1.920965    0.151457\n",
       "9  indiv_greedy  6.484443  0.374912  1.281067    0.136480"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ODE_regression_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f17e622a",
   "metadata": {},
   "source": [
    "### VDP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "07612694",
   "metadata": {},
   "outputs": [],
   "source": [
    "indiv_rand_regression_for_VDP = indiv_rand_regression_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "30b85fe0",
   "metadata": {},
   "outputs": [],
   "source": [
    "VDP_mse_results = deepcopy(mse_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "b56ee1da",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>indiv_greedy</td>\n",
       "      <td>6.839020</td>\n",
       "      <td>0.741066</td>\n",
       "      <td>1.858412</td>\n",
       "      <td>0.265737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>rand</td>\n",
       "      <td>4.973591</td>\n",
       "      <td>1.254244</td>\n",
       "      <td>2.457056</td>\n",
       "      <td>0.895329</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines   Avg MSE    Stderr   Std MSE  Stderr Std\n",
       "0  indiv_greedy  6.839020  0.741066  1.858412    0.265737\n",
       "1          rand  4.973591  1.254244  2.457056    0.895329"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "indiv_rand_regression_for_VDP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "6bc3fb81",
   "metadata": {},
   "outputs": [],
   "source": [
    "with cwd(oj(local_dir, setting)):\n",
    "    indiv_rand_regression_df.to_csv('VDP_correction.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "00496bab",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>indiv_greedy</td>\n",
       "      <td>6.83902</td>\n",
       "      <td>0.741066</td>\n",
       "      <td>1.858412</td>\n",
       "      <td>0.265737</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines  Avg MSE    Stderr   Std MSE  Stderr Std\n",
       "0  indiv_greedy  6.83902  0.741066  1.858412    0.265737"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# indiv_regression_df = get_regression_df(mse_results)\n",
    "indiv_regression_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "16deb378",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "47e1191b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "defaultdict(list,\n",
       "            {'greedy_obs_1-avg-mses': [4.178536317003405,\n",
       "              4.305602157058645,\n",
       "              2.868588750859472,\n",
       "              3.851607596680812,\n",
       "              5.1480001667037785],\n",
       "             'greedy_obs_1-mses': [[2.2593988882213423,\n",
       "               7.560133496191136,\n",
       "               2.7160765665977378],\n",
       "              [3.655590891935502, 7.222758576876525, 2.038457002363908],\n",
       "              [2.2572134940223623, 3.8007284909631096, 2.5478242675929437],\n",
       "              [2.623772508328433, 3.4927824561882304, 5.438267825525773],\n",
       "              [1.7120316230317485, 10.138245941169346, 3.5937229359102414]],\n",
       "             'greedy_obs_2-avg-mses': [4.1246115376366195,\n",
       "              4.643293169825471,\n",
       "              4.009888615452311,\n",
       "              4.359279892610691],\n",
       "             'greedy_obs_2-mses': [[2.2619341957503094,\n",
       "               7.441289090976445,\n",
       "               2.6706113261831033],\n",
       "              [3.106127457532123, 7.439960984941314, 3.3837910670029743],\n",
       "              [2.3094426003926403, 4.582739874122948, 5.137483371841343],\n",
       "              [4.2961310801406745, 4.784866494497222, 3.9968421031941777]],\n",
       "             'greedy_obs_3-avg-mses': [3.8803439368609682,\n",
       "              4.6542023256061835,\n",
       "              2.968789755816945,\n",
       "              4.3402821613232945],\n",
       "             'greedy_obs_3-mses': [[2.08755619267208,\n",
       "               6.655423864549584,\n",
       "               2.8980517533612415],\n",
       "              [4.076492983234241, 8.04369721126687, 1.8424167823174407],\n",
       "              [2.130140146565823, 4.13357255959057, 2.6426565612944426],\n",
       "              [3.283246782628636, 6.763101204866206, 2.9744984964750407]],\n",
       "             'greedy_obs_4-avg-mses': [3.9074934959009666,\n",
       "              4.730217787186983,\n",
       "              3.433268780269914,\n",
       "              4.720419260059275],\n",
       "             'greedy_obs_4-mses': [[2.1776609491636307,\n",
       "               6.867276390217125,\n",
       "               2.6775431483221452],\n",
       "              [3.791931844516783, 8.268486044014427, 2.1302354730297393],\n",
       "              [2.6776572087158317, 4.70651390229001, 2.915635229803901],\n",
       "              [2.700525103519845, 4.462675130516333, 6.998057546141647]],\n",
       "             'greedy_sum_obs-avg-mses': [4.061784173740402,\n",
       "              4.9740955661904644,\n",
       "              3.3312987179495694,\n",
       "              3.472606954814185],\n",
       "             'greedy_sum_obs-mses': [[2.2663611124122496,\n",
       "               7.132224499700959,\n",
       "               2.786766909107997],\n",
       "              [4.315617727925617, 6.979010762949061, 3.6276582076967157],\n",
       "              [2.406190418663604, 4.176289691902994, 3.4114160432821095],\n",
       "              [2.725462924543759, 3.9503289278101175, 3.742029012088679]],\n",
       "             'dynamic_beta_obs-avg-mses': [4.343215354787309,\n",
       "              3.819967169191913,\n",
       "              3.1676683165120703,\n",
       "              3.994408331985957],\n",
       "             'dynamic_beta_obs-mses': [[2.121749495292021,\n",
       "               8.146438332858562,\n",
       "               2.7614582362113436],\n",
       "              [3.6423720806244173, 5.453761924276403, 2.3637675026749196],\n",
       "              [2.2880615165396234, 4.4703108041624535, 2.7446326288341343],\n",
       "              [3.3868287317467485, 4.338878241469001, 4.257518022742121]],\n",
       "             'joint_obs-avg-mses': [4.142191588448568,\n",
       "              4.156182681275967,\n",
       "              2.9960771459035205,\n",
       "              4.262828942106008],\n",
       "             'joint_obs-mses': [[2.247868456899366,\n",
       "               7.50485897321648,\n",
       "               2.673847335229857],\n",
       "              [3.0736898070658016, 6.493790407707974, 2.901067829054123],\n",
       "              [2.345002158071919, 4.17597109252603, 2.4672581871126122],\n",
       "              [5.016506501893241, 5.021465734597145, 2.750514589827638]],\n",
       "             'rand_obs-avg-mses': [4.220151103473373,\n",
       "              3.5617876620526436,\n",
       "              2.9597699315835513,\n",
       "              2.912154925624779],\n",
       "             'rand_obs-mses': [[2.077718080982347,\n",
       "               8.35578392859609,\n",
       "               2.226951300841684],\n",
       "              [3.4504448895337854, 5.1086831521416025, 2.1262349444825435],\n",
       "              [2.2451121399962144, 3.871478436038023, 2.7627192187164153],\n",
       "              [2.580703960798584, 4.203933367478642, 1.9518274485971112]],\n",
       "             'entropy_obs-avg-mses': [4.649901106905428,\n",
       "              4.649300127104264,\n",
       "              3.098202044917953,\n",
       "              5.493464080509275],\n",
       "             'entropy_obs-mses': [[2.213033030039715,\n",
       "               8.47031768434467,\n",
       "               3.2663526063318993],\n",
       "              [3.92318309284247, 7.888649143791107, 2.136068144679217],\n",
       "              [2.3565896613038664, 4.3367588907504855, 2.6012575826995077],\n",
       "              [4.127992066459512, 4.902020109131356, 7.450380065936957]],\n",
       "             'indiv_greedy_obs-avg-mses': [5.276575478681833,\n",
       "              7.245973460827789,\n",
       "              6.780558548921639,\n",
       "              7.882738649452617],\n",
       "             'indiv_greedy_obs-mses': [[4.507582228421915,\n",
       "               8.895293119214099,\n",
       "               2.426851088409484],\n",
       "              [12.105060058101998, 7.4870672502492575, 2.145793074132109],\n",
       "              [10.697877550566734, 7.192418273727554, 2.4513798224706314],\n",
       "              [14.015990238258247, 6.126535891141112, 3.50568981895849]]})"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mse_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e467c1e5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "25fa17a8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "c0a7a19b",
   "metadata": {},
   "source": [
    "#### all the share the same target - time [0,8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "8c1f515f",
   "metadata": {},
   "outputs": [
    {
     "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>Baselines</th>\n",
       "      <th>Avg MSE</th>\n",
       "      <th>Stderr</th>\n",
       "      <th>Std MSE</th>\n",
       "      <th>Stderr Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>greedy_1</td>\n",
       "      <td>2.756342</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.756342</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>greedy_2</td>\n",
       "      <td>2.398316</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.398316</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>greedy_3</td>\n",
       "      <td>3.438041</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.438041</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>greedy_4</td>\n",
       "      <td>3.627456</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.627456</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>greedy_sum</td>\n",
       "      <td>2.944234</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.944234</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>dynamic_beta</td>\n",
       "      <td>3.803732</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.803732</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>joint</td>\n",
       "      <td>2.785771</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.785771</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>rand</td>\n",
       "      <td>2.650305</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.650305</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>entropy</td>\n",
       "      <td>2.376445</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.376445</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Baselines   Avg MSE  Stderr   Std MSE  Stderr Std\n",
       "0      greedy_1  2.756342     NaN  2.756342         NaN\n",
       "1      greedy_2  2.398316     NaN  2.398316         NaN\n",
       "2      greedy_3  3.438041     NaN  3.438041         NaN\n",
       "3      greedy_4  3.627456     NaN  3.627456         NaN\n",
       "4    greedy_sum  2.944234     NaN  2.944234         NaN\n",
       "5  dynamic_beta  3.803732     NaN  3.803732         NaN\n",
       "6         joint  2.785771     NaN  2.785771         NaN\n",
       "7          rand  2.650305     NaN  2.650305         NaN\n",
       "8       entropy  2.376445     NaN  2.376445         NaN"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "de_regression_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "067b2ed9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c3e91f38",
   "metadata": {},
   "outputs": [],
   "source": [
    "# with cwd(oj(results_dir, setting)):\n",
    "#     de_regression_df.to_latex('regression_results.tex',index=False)\n",
    "#     de_regression_df.to_csv('regression_results.csv',index=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "823a5320",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}