run_bert.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "293077f9-46f1-4489-b8ac-b2de4bef194a",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/users/amarvenu/miniconda3/envs/nlp/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "import os\n",
    "import scipy\n",
    "\n",
    "import torch\n",
    "from torch import nn, optim\n",
    "from torch.nn import functional as F\n",
    "\n",
    "from src import data\n",
    "from src.detm import DETM\n",
    "\n",
    "from gensim.models import CoherenceModel\n",
    "from gensim.corpora import Dictionary\n",
    "\n",
    "from analysis_utils import get_detm_topics, topic_diversity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1309530f-89db-4946-b9ea-bce4cc36ea74",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "class AttrDict(dict):\n",
    "    def __init__(self, *args, **kwargs):\n",
    "        super(AttrDict, self).__init__(*args, **kwargs)\n",
    "        self.__dict__ = self"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "680dc701-a07d-46a5-9cc4-d351bc25fef3",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "### data and file related arguments\n",
    "arg_str = \"\"\"\n",
    "parser.add_argument('--dataset', type=str, default='un', help='name of corpus')\n",
    "parser.add_argument('--data_path', type=str, default='un/', help='directory containing data')\n",
    "parser.add_argument('--emb_path', type=str, default='skipgram/embeddings.txt', help='directory containing embeddings')\n",
    "parser.add_argument('--save_path', type=str, default='./results', help='path to save results')\n",
    "parser.add_argument('--batch_size', type=int, default=1000, help='number of documents in a batch for training')\n",
    "parser.add_argument('--min_df', type=int, default=100, help='to get the right data..minimum document frequency')\n",
    "\n",
    "### model-related arguments\n",
    "parser.add_argument('--num_topics', type=int, default=50, help='number of topics')\n",
    "parser.add_argument('--rho_size', type=int, default=300, help='dimension of rho')\n",
    "parser.add_argument('--emb_size', type=int, default=300, help='dimension of embeddings')\n",
    "parser.add_argument('--t_hidden_size', type=int, default=800, help='dimension of hidden space of q(theta)')\n",
    "parser.add_argument('--theta_act', type=str, default='relu', help='tanh, softplus, relu, rrelu, leakyrelu, elu, selu, glu)')\n",
    "parser.add_argument('--train_embeddings', type=int, default=1, help='whether to fix rho or train it')\n",
    "parser.add_argument('--eta_nlayers', type=int, default=3, help='number of layers for eta')\n",
    "parser.add_argument('--eta_hidden_size', type=int, default=200, help='number of hidden units for rnn')\n",
    "parser.add_argument('--delta', type=float, default=0.005, help='prior variance')\n",
    "\n",
    "### optimization-related arguments\n",
    "parser.add_argument('--lr', type=float, default=0.005, help='learning rate')\n",
    "parser.add_argument('--lr_factor', type=float, default=4.0, help='divide learning rate by this')\n",
    "parser.add_argument('--epochs', type=int, default=100, help='number of epochs to train')\n",
    "parser.add_argument('--mode', type=str, default='train', help='train or eval model')\n",
    "parser.add_argument('--optimizer', type=str, default='adam', help='choice of optimizer')\n",
    "parser.add_argument('--seed', type=int, default=2019, help='random seed (default: 1)')\n",
    "parser.add_argument('--enc_drop', type=float, default=0.0, help='dropout rate on encoder')\n",
    "parser.add_argument('--eta_dropout', type=float, default=0.0, help='dropout rate on rnn for eta')\n",
    "parser.add_argument('--clip', type=float, default=0.0, help='gradient clipping')\n",
    "parser.add_argument('--nonmono', type=int, default=10, help='number of bad hits allowed')\n",
    "parser.add_argument('--wdecay', type=float, default=1.2e-6, help='some l2 regularization')\n",
    "parser.add_argument('--anneal_lr', type=int, default=0, help='whether to anneal the learning rate or not')\n",
    "parser.add_argument('--bow_norm', type=int, default=1, help='normalize the bows or not')\n",
    "\n",
    "### evaluation, visualization, and logging-related arguments\n",
    "parser.add_argument('--num_words', type=int, default=20, help='number of words for topic viz')\n",
    "parser.add_argument('--log_interval', type=int, default=10, help='when to log training')\n",
    "parser.add_argument('--visualize_every', type=int, default=1, help='when to visualize results')\n",
    "parser.add_argument('--eval_batch_size', type=int, default=1000, help='input batch size for evaluation')\n",
    "parser.add_argument('--load_from', type=str, default='', help='the name of the ckpt to eval from')\n",
    "parser.add_argument('--tc', type=int, default=0, help='whether to compute tc or not')\n",
    "\"\"\".split('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4c037c86-db04-4fa8-8fc5-13f99332813d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "keys = [x.strip(\"parser.add_argument('\").split(',')[0].strip('--').strip(\"'\") for x in arg_str if (len(x) > 0) and (not x.startswith('#'))]\n",
    "values = [x.strip(\"parser.add_argument('\").split(',')[2].strip(\" default=\").strip(\"'\") for x in arg_str if (len(x) > 0) and (not x.startswith('#'))]\n",
    "tmp_dict = dict(zip(keys, values))\n",
    "\n",
    "for k, v in tmp_dict.items():\n",
    "    if v.isnumeric():\n",
    "        tmp_dict[k] = int(v)\n",
    "    elif ('.' in v) and (v[0].isnumeric()):\n",
    "        tmp_dict[k] = float(v)    \n",
    "\n",
    "args = AttrDict()\n",
    "args.update(tmp_dict)\n",
    "\n",
    "args.train_embeddings = 0\n",
    "args.rho_size = 768\n",
    "args.num_topics = 10\n",
    "args.batch_size = 100\n",
    "args.epochs = 150"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "574d5721-f212-49cf-afe9-dddf97ee08d8",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "train_arr = np.load('test_data.npz', allow_pickle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e20155d7-2fff-4d89-a127-ba89a66d01f9",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "train_tokens = train_arr['train_tokens']\n",
    "train_counts = train_arr['train_counts']\n",
    "train_times = train_arr['train_times']\n",
    "vocab = train_arr['vocab']\n",
    "embeddings = train_arr['embeddings']\n",
    "\n",
    "args.num_times = len(np.unique(train_times))\n",
    "args.num_docs_train = len(train_tokens)\n",
    "args.vocab_size = len(vocab)\n",
    "args.num_words = 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "21ada37c-a91d-4293-a147-7de92607ae8d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "idx: 0/2\n",
      "CPU times: user 2.92 s, sys: 543 ms, total: 3.46 s\n",
      "Wall time: 314 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "train_rnn_inp = data.get_rnn_input(train_tokens, train_counts, train_times, args.num_times, args.vocab_size, args.num_docs_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c6b0df7f-8ffb-4dea-84c5-3faa403ac5ee",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if not os.path.exists(args['save_path']):\n",
    "    os.makedirs(args['save_path'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "24c7d9fa-102e-4d7e-8734-53096816d0a5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if args.mode == 'eval':\n",
    "    ckpt = args.load_from\n",
    "else:\n",
    "    ckpt = os.path.join(args.save_path, \n",
    "        'bert_detm_{}_K_{}_Htheta_{}_Optim_{}_Clip_{}_ThetaAct_{}_Lr_{}_Bsz_{}_RhoSize_{}_L_{}_minDF_{}_trainEmbeddings_{}'.format(\n",
    "        args.dataset, args.num_topics, args.t_hidden_size, args.optimizer, args.clip, args.theta_act, \n",
    "            args.lr, args.batch_size, args.rho_size, args.eta_nlayers, args.min_df, args.train_embeddings))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "047e40c2-92aa-4707-a77c-3db9313831cf",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DETM(\n",
       "  (t_drop): Dropout(p=0.0, inplace=False)\n",
       "  (theta_act): ReLU()\n",
       "  (q_theta): Sequential(\n",
       "    (0): Linear(in_features=4948, out_features=800, bias=True)\n",
       "    (1): ReLU()\n",
       "    (2): Linear(in_features=800, out_features=800, bias=True)\n",
       "    (3): ReLU()\n",
       "  )\n",
       "  (mu_q_theta): Linear(in_features=800, out_features=10, bias=True)\n",
       "  (logsigma_q_theta): Linear(in_features=800, out_features=10, bias=True)\n",
       "  (q_eta_map): Linear(in_features=4938, out_features=200, bias=True)\n",
       "  (q_eta): LSTM(200, 200, num_layers=3)\n",
       "  (mu_q_eta): Linear(in_features=210, out_features=10, bias=True)\n",
       "  (logsigma_q_eta): Linear(in_features=210, out_features=10, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "embeddings = torch.from_numpy(embeddings).to(device)\n",
    "args.embeddings_dim = embeddings.size()\n",
    "\n",
    "model = DETM(args, embeddings)\n",
    "model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f54de551-16ca-419d-876e-1ba965d39d2d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if args.optimizer == 'adam':\n",
    "    optimizer = optim.Adam(model.parameters(), lr=args.lr, weight_decay=args.wdecay)\n",
    "elif args.optimizer == 'adagrad':\n",
    "    optimizer = optim.Adagrad(model.parameters(), lr=args.lr, weight_decay=args.wdecay)\n",
    "elif args.optimizer == 'adadelta':\n",
    "    optimizer = optim.Adadelta(model.parameters(), lr=args.lr, weight_decay=args.wdecay)\n",
    "elif args.optimizer == 'rmsprop':\n",
    "    optimizer = optim.RMSprop(model.parameters(), lr=args.lr, weight_decay=args.wdecay)\n",
    "elif args.optimizer == 'asgd':\n",
    "    optimizer = optim.ASGD(model.parameters(), lr=args.lr, t0=0, lambd=0., weight_decay=args.wdecay)\n",
    "else:\n",
    "    print('Defaulting to vanilla SGD')\n",
    "    optimizer = optim.SGD(model.parameters(), lr=args.lr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "71a92f90-5e83-4bf4-a3a0-2d9390061275",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def train(epoch):\n",
    "    \"\"\"Train DETM on data for one epoch.\n",
    "    \"\"\"\n",
    "    model.train()\n",
    "    acc_loss = 0\n",
    "    acc_nll = 0\n",
    "    acc_kl_theta_loss = 0\n",
    "    acc_kl_eta_loss = 0\n",
    "    acc_kl_alpha_loss = 0\n",
    "    cnt = 0\n",
    "    indices = torch.randperm(args.num_docs_train)\n",
    "    indices = torch.split(indices, args.batch_size) \n",
    "    for idx, ind in enumerate(indices):\n",
    "        optimizer.zero_grad()\n",
    "        model.zero_grad()\n",
    "        data_batch, times_batch = data.get_batch(\n",
    "            train_tokens, train_counts, ind, args.vocab_size, args.emb_size, temporal=True, times=train_times)\n",
    "        sums = data_batch.sum(1).unsqueeze(1)\n",
    "        if args.bow_norm:\n",
    "            normalized_data_batch = data_batch / sums\n",
    "        else:\n",
    "            normalized_data_batch = data_batch\n",
    "\n",
    "        loss, nll, kl_alpha, kl_eta, kl_theta = model(data_batch, normalized_data_batch, times_batch, train_rnn_inp, args.num_docs_train)\n",
    "        loss.backward()\n",
    "        if args.clip > 0:\n",
    "            torch.nn.utils.clip_grad_norm_(model.parameters(), args.clip)\n",
    "        optimizer.step()\n",
    "\n",
    "        acc_loss += torch.sum(loss).item()\n",
    "        acc_nll += torch.sum(nll).item()\n",
    "        acc_kl_theta_loss += torch.sum(kl_theta).item()\n",
    "        acc_kl_eta_loss += torch.sum(kl_eta).item()\n",
    "        acc_kl_alpha_loss += torch.sum(kl_alpha).item()\n",
    "        cnt += 1\n",
    "\n",
    "        if idx % args.log_interval == 0 and idx > 0:\n",
    "            cur_loss = round(acc_loss / cnt, 2) \n",
    "            cur_nll = round(acc_nll / cnt, 2) \n",
    "            cur_kl_theta = round(acc_kl_theta_loss / cnt, 2) \n",
    "            cur_kl_eta = round(acc_kl_eta_loss / cnt, 2) \n",
    "            cur_kl_alpha = round(acc_kl_alpha_loss / cnt, 2) \n",
    "            lr = optimizer.param_groups[0]['lr']\n",
    "            print('Epoch: {} .. batch: {}/{} .. LR: {} .. KL_theta: {} .. KL_eta: {} .. KL_alpha: {} .. Rec_loss: {} .. NELBO: {}'.format(\n",
    "                epoch, idx, len(indices), lr, cur_kl_theta, cur_kl_eta, cur_kl_alpha, cur_nll, cur_loss))\n",
    "    \n",
    "    cur_loss = round(acc_loss / cnt, 2) \n",
    "    cur_nll = round(acc_nll / cnt, 2) \n",
    "    cur_kl_theta = round(acc_kl_theta_loss / cnt, 2) \n",
    "    cur_kl_eta = round(acc_kl_eta_loss / cnt, 2) \n",
    "    cur_kl_alpha = round(acc_kl_alpha_loss / cnt, 2) \n",
    "    lr = optimizer.param_groups[0]['lr']\n",
    "    print('*'*100)\n",
    "    print('Epoch----->{} .. LR: {} .. KL_theta: {} .. KL_eta: {} .. KL_alpha: {} .. Rec_loss: {} .. NELBO: {}'.format(\n",
    "            epoch, lr, cur_kl_theta, cur_kl_eta, cur_kl_alpha, cur_nll, cur_loss))\n",
    "    print('*'*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "b51ea22c-39c7-4ece-a26c-2c97c817146f",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17713.67 .. KL_eta: 3175.82 .. KL_alpha: 11701875.27 .. Rec_loss: 20134262.18 .. NELBO: 31857027.27\n",
      "****************************************************************************************************\n",
      "Epoch----->1 .. LR: 0.005 .. KL_theta: 21019.38 .. KL_eta: 2439.99 .. KL_alpha: 11547434.12 .. Rec_loss: 20008464.47 .. NELBO: 31579358.24\n",
      "****************************************************************************************************\n",
      "Epoch: 2 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19563.38 .. KL_eta: 400.39 .. KL_alpha: 10816257.91 .. Rec_loss: 19639478.36 .. NELBO: 30475700.36\n",
      "****************************************************************************************************\n",
      "Epoch----->2 .. LR: 0.005 .. KL_theta: 19550.35 .. KL_eta: 325.96 .. KL_alpha: 10650328.94 .. Rec_loss: 19664786.59 .. NELBO: 30334992.0\n",
      "****************************************************************************************************\n",
      "Epoch: 3 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21565.01 .. KL_eta: 145.17 .. KL_alpha: 9937373.0 .. Rec_loss: 19598484.0 .. NELBO: 29557566.91\n",
      "****************************************************************************************************\n",
      "Epoch----->3 .. LR: 0.005 .. KL_theta: 20742.04 .. KL_eta: 140.33 .. KL_alpha: 9808933.06 .. Rec_loss: 19383946.0 .. NELBO: 29213761.18\n",
      "****************************************************************************************************\n",
      "Epoch: 4 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19964.81 .. KL_eta: 309.56 .. KL_alpha: 9177961.27 .. Rec_loss: 19104056.91 .. NELBO: 28302292.91\n",
      "****************************************************************************************************\n",
      "Epoch----->4 .. LR: 0.005 .. KL_theta: 20517.65 .. KL_eta: 293.12 .. KL_alpha: 9051338.76 .. Rec_loss: 19215480.0 .. NELBO: 28287629.53\n",
      "****************************************************************************************************\n",
      "Epoch: 5 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17652.5 .. KL_eta: 203.82 .. KL_alpha: 8518063.09 .. Rec_loss: 19092434.18 .. NELBO: 27628353.64\n",
      "****************************************************************************************************\n",
      "Epoch----->5 .. LR: 0.005 .. KL_theta: 18854.28 .. KL_eta: 231.36 .. KL_alpha: 8396860.03 .. Rec_loss: 19037126.12 .. NELBO: 27453071.88\n",
      "****************************************************************************************************\n",
      "Epoch: 6 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19484.74 .. KL_eta: 243.01 .. KL_alpha: 7914257.14 .. Rec_loss: 19036170.0 .. NELBO: 26970155.09\n",
      "****************************************************************************************************\n",
      "Epoch----->6 .. LR: 0.005 .. KL_theta: 19213.31 .. KL_eta: 230.94 .. KL_alpha: 7800786.41 .. Rec_loss: 18874839.06 .. NELBO: 26695069.76\n",
      "****************************************************************************************************\n",
      "Epoch: 7 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19233.25 .. KL_eta: 204.11 .. KL_alpha: 7379479.23 .. Rec_loss: 18855245.64 .. NELBO: 26254162.73\n",
      "****************************************************************************************************\n",
      "Epoch----->7 .. LR: 0.005 .. KL_theta: 18860.49 .. KL_eta: 204.78 .. KL_alpha: 7287636.71 .. Rec_loss: 18693633.12 .. NELBO: 26000335.41\n",
      "****************************************************************************************************\n",
      "Epoch: 8 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18425.77 .. KL_eta: 256.97 .. KL_alpha: 6816493.18 .. Rec_loss: 18376334.55 .. NELBO: 25211510.55\n",
      "****************************************************************************************************\n",
      "Epoch----->8 .. LR: 0.005 .. KL_theta: 18261.16 .. KL_eta: 239.37 .. KL_alpha: 6738225.15 .. Rec_loss: 18616772.0 .. NELBO: 25373497.76\n",
      "****************************************************************************************************\n",
      "Epoch: 9 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18898.76 .. KL_eta: 338.23 .. KL_alpha: 6399100.59 .. Rec_loss: 18514103.27 .. NELBO: 24932440.55\n",
      "****************************************************************************************************\n",
      "Epoch----->9 .. LR: 0.005 .. KL_theta: 19234.01 .. KL_eta: 363.76 .. KL_alpha: 6320045.18 .. Rec_loss: 18481424.47 .. NELBO: 24821067.06\n",
      "****************************************************************************************************\n",
      "Epoch: 10 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17714.99 .. KL_eta: 429.67 .. KL_alpha: 5982666.82 .. Rec_loss: 18233319.73 .. NELBO: 24234130.91\n",
      "****************************************************************************************************\n",
      "Epoch----->10 .. LR: 0.005 .. KL_theta: 17509.02 .. KL_eta: 422.99 .. KL_alpha: 5906043.09 .. Rec_loss: 18459105.47 .. NELBO: 24383080.59\n",
      "****************************************************************************************************\n",
      "Epoch: 11 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19412.07 .. KL_eta: 362.11 .. KL_alpha: 5595940.45 .. Rec_loss: 18288946.18 .. NELBO: 23904660.91\n",
      "****************************************************************************************************\n",
      "Epoch----->11 .. LR: 0.005 .. KL_theta: 18439.5 .. KL_eta: 367.09 .. KL_alpha: 5546749.41 .. Rec_loss: 18366153.29 .. NELBO: 23931709.29\n",
      "****************************************************************************************************\n",
      "Epoch: 12 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18415.52 .. KL_eta: 387.92 .. KL_alpha: 5298673.27 .. Rec_loss: 18274289.27 .. NELBO: 23591765.45\n",
      "****************************************************************************************************\n",
      "Epoch----->12 .. LR: 0.005 .. KL_theta: 18816.1 .. KL_eta: 457.4 .. KL_alpha: 5243335.74 .. Rec_loss: 18304325.18 .. NELBO: 23566934.35\n",
      "****************************************************************************************************\n",
      "Epoch: 13 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17974.3 .. KL_eta: 499.24 .. KL_alpha: 4970964.73 .. Rec_loss: 18101204.91 .. NELBO: 23090643.64\n",
      "****************************************************************************************************\n",
      "Epoch----->13 .. LR: 0.005 .. KL_theta: 18367.37 .. KL_eta: 541.17 .. KL_alpha: 4934128.24 .. Rec_loss: 18191160.59 .. NELBO: 23144197.76\n",
      "****************************************************************************************************\n",
      "Epoch: 14 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17346.66 .. KL_eta: 408.36 .. KL_alpha: 4681789.55 .. Rec_loss: 17981029.0 .. NELBO: 22680573.82\n",
      "****************************************************************************************************\n",
      "Epoch----->14 .. LR: 0.005 .. KL_theta: 18146.86 .. KL_eta: 394.98 .. KL_alpha: 4648479.71 .. Rec_loss: 18137712.18 .. NELBO: 22804734.0\n",
      "****************************************************************************************************\n",
      "Epoch: 15 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18455.42 .. KL_eta: 351.15 .. KL_alpha: 4443592.23 .. Rec_loss: 17982933.82 .. NELBO: 22445332.55\n",
      "****************************************************************************************************\n",
      "Epoch----->15 .. LR: 0.005 .. KL_theta: 17561.04 .. KL_eta: 356.14 .. KL_alpha: 4403916.12 .. Rec_loss: 18044325.29 .. NELBO: 22466158.47\n",
      "****************************************************************************************************\n",
      "Epoch: 16 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17766.39 .. KL_eta: 378.13 .. KL_alpha: 4205186.23 .. Rec_loss: 17926308.91 .. NELBO: 22149639.82\n",
      "****************************************************************************************************\n",
      "Epoch----->16 .. LR: 0.005 .. KL_theta: 18230.3 .. KL_eta: 407.17 .. KL_alpha: 4174293.28 .. Rec_loss: 17971118.12 .. NELBO: 22164048.82\n",
      "****************************************************************************************************\n",
      "Epoch: 17 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17416.64 .. KL_eta: 408.09 .. KL_alpha: 4013113.36 .. Rec_loss: 18011337.09 .. NELBO: 22042275.09\n",
      "****************************************************************************************************\n",
      "Epoch----->17 .. LR: 0.005 .. KL_theta: 17649.16 .. KL_eta: 390.04 .. KL_alpha: 3982488.94 .. Rec_loss: 17912066.71 .. NELBO: 21912594.82\n",
      "****************************************************************************************************\n",
      "Epoch: 18 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 16988.37 .. KL_eta: 375.18 .. KL_alpha: 3805582.93 .. Rec_loss: 17622691.36 .. NELBO: 21445638.0\n",
      "****************************************************************************************************\n",
      "Epoch----->18 .. LR: 0.005 .. KL_theta: 17629.21 .. KL_eta: 378.17 .. KL_alpha: 3777328.32 .. Rec_loss: 17880684.06 .. NELBO: 21676020.12\n",
      "****************************************************************************************************\n",
      "Epoch: 19 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18543.24 .. KL_eta: 358.31 .. KL_alpha: 3623996.89 .. Rec_loss: 17747706.0 .. NELBO: 21390604.18\n",
      "****************************************************************************************************\n",
      "Epoch----->19 .. LR: 0.005 .. KL_theta: 17977.1 .. KL_eta: 371.14 .. KL_alpha: 3596217.49 .. Rec_loss: 17818385.76 .. NELBO: 21432951.29\n",
      "****************************************************************************************************\n",
      "Epoch: 20 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17738.21 .. KL_eta: 446.27 .. KL_alpha: 3456789.57 .. Rec_loss: 17916205.45 .. NELBO: 21391179.27\n",
      "****************************************************************************************************\n",
      "Epoch----->20 .. LR: 0.005 .. KL_theta: 17863.63 .. KL_eta: 438.38 .. KL_alpha: 3429913.15 .. Rec_loss: 17739439.24 .. NELBO: 21187654.24\n",
      "****************************************************************************************************\n",
      "Epoch: 21 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 16201.69 .. KL_eta: 381.33 .. KL_alpha: 3305826.89 .. Rec_loss: 17586353.18 .. NELBO: 20908763.09\n",
      "****************************************************************************************************\n",
      "Epoch----->21 .. LR: 0.005 .. KL_theta: 17567.05 .. KL_eta: 394.99 .. KL_alpha: 3278200.85 .. Rec_loss: 17784141.35 .. NELBO: 21080304.47\n",
      "****************************************************************************************************\n",
      "Epoch: 22 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17797.5 .. KL_eta: 392.45 .. KL_alpha: 3162747.14 .. Rec_loss: 17753589.09 .. NELBO: 20934526.18\n",
      "****************************************************************************************************\n",
      "Epoch----->22 .. LR: 0.005 .. KL_theta: 17643.63 .. KL_eta: 385.64 .. KL_alpha: 3142356.56 .. Rec_loss: 17677535.94 .. NELBO: 20837921.76\n",
      "****************************************************************************************************\n",
      "Epoch: 23 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19223.76 .. KL_eta: 347.3 .. KL_alpha: 3022822.23 .. Rec_loss: 17990242.82 .. NELBO: 21032636.0\n",
      "****************************************************************************************************\n",
      "Epoch----->23 .. LR: 0.005 .. KL_theta: 18408.2 .. KL_eta: 356.16 .. KL_alpha: 2999212.49 .. Rec_loss: 17674648.29 .. NELBO: 20692624.82\n",
      "****************************************************************************************************\n",
      "Epoch: 24 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17959.15 .. KL_eta: 463.05 .. KL_alpha: 2902549.68 .. Rec_loss: 17606777.18 .. NELBO: 20527749.09\n",
      "****************************************************************************************************\n",
      "Epoch----->24 .. LR: 0.005 .. KL_theta: 18033.17 .. KL_eta: 484.91 .. KL_alpha: 2882129.54 .. Rec_loss: 17663471.71 .. NELBO: 20564119.41\n",
      "****************************************************************************************************\n",
      "Epoch: 25 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17597.6 .. KL_eta: 455.11 .. KL_alpha: 2789332.43 .. Rec_loss: 17601157.73 .. NELBO: 20408542.73\n",
      "****************************************************************************************************\n",
      "Epoch----->25 .. LR: 0.005 .. KL_theta: 17290.9 .. KL_eta: 419.83 .. KL_alpha: 2770174.9 .. Rec_loss: 17599992.53 .. NELBO: 20387878.0\n",
      "****************************************************************************************************\n",
      "Epoch: 26 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19153.03 .. KL_eta: 417.86 .. KL_alpha: 2673617.05 .. Rec_loss: 17805107.36 .. NELBO: 20498295.45\n",
      "****************************************************************************************************\n",
      "Epoch----->26 .. LR: 0.005 .. KL_theta: 19159.42 .. KL_eta: 409.24 .. KL_alpha: 2663517.78 .. Rec_loss: 17526808.0 .. NELBO: 20209894.59\n",
      "****************************************************************************************************\n",
      "Epoch: 27 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18607.25 .. KL_eta: 456.22 .. KL_alpha: 2580353.0 .. Rec_loss: 17496052.91 .. NELBO: 20095469.27\n",
      "****************************************************************************************************\n",
      "Epoch----->27 .. LR: 0.005 .. KL_theta: 18738.08 .. KL_eta: 481.38 .. KL_alpha: 2557792.1 .. Rec_loss: 17544737.41 .. NELBO: 20121748.82\n",
      "****************************************************************************************************\n",
      "Epoch: 28 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17995.23 .. KL_eta: 459.22 .. KL_alpha: 2485133.5 .. Rec_loss: 17695490.73 .. NELBO: 20199078.91\n",
      "****************************************************************************************************\n",
      "Epoch----->28 .. LR: 0.005 .. KL_theta: 18301.67 .. KL_eta: 432.66 .. KL_alpha: 2469189.43 .. Rec_loss: 17579107.71 .. NELBO: 20067031.65\n",
      "****************************************************************************************************\n",
      "Epoch: 29 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20070.84 .. KL_eta: 313.12 .. KL_alpha: 2389357.3 .. Rec_loss: 17428924.64 .. NELBO: 19838666.36\n",
      "****************************************************************************************************\n",
      "Epoch----->29 .. LR: 0.005 .. KL_theta: 19559.83 .. KL_eta: 317.78 .. KL_alpha: 2371409.31 .. Rec_loss: 17568822.65 .. NELBO: 19960110.0\n",
      "****************************************************************************************************\n",
      "Epoch: 30 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18337.61 .. KL_eta: 433.65 .. KL_alpha: 2310204.55 .. Rec_loss: 17805113.82 .. NELBO: 20134089.64\n",
      "****************************************************************************************************\n",
      "Epoch----->30 .. LR: 0.005 .. KL_theta: 18311.47 .. KL_eta: 449.7 .. KL_alpha: 2292782.65 .. Rec_loss: 17467972.29 .. NELBO: 19779516.12\n",
      "****************************************************************************************************\n",
      "Epoch: 31 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18203.96 .. KL_eta: 532.13 .. KL_alpha: 2225784.27 .. Rec_loss: 17428956.73 .. NELBO: 19673477.27\n",
      "****************************************************************************************************\n",
      "Epoch----->31 .. LR: 0.005 .. KL_theta: 18550.35 .. KL_eta: 528.77 .. KL_alpha: 2208899.46 .. Rec_loss: 17475047.82 .. NELBO: 19703026.59\n",
      "****************************************************************************************************\n",
      "Epoch: 32 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18245.9 .. KL_eta: 462.42 .. KL_alpha: 2145174.36 .. Rec_loss: 17228038.45 .. NELBO: 19391921.09\n",
      "****************************************************************************************************\n",
      "Epoch----->32 .. LR: 0.005 .. KL_theta: 19469.01 .. KL_eta: 452.61 .. KL_alpha: 2135044.66 .. Rec_loss: 17503196.82 .. NELBO: 19658163.06\n",
      "****************************************************************************************************\n",
      "Epoch: 33 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19131.55 .. KL_eta: 388.67 .. KL_alpha: 2079739.25 .. Rec_loss: 17605321.45 .. NELBO: 19704581.27\n",
      "****************************************************************************************************\n",
      "Epoch----->33 .. LR: 0.005 .. KL_theta: 19402.26 .. KL_eta: 403.82 .. KL_alpha: 2068727.01 .. Rec_loss: 17456113.29 .. NELBO: 19544646.71\n",
      "****************************************************************************************************\n",
      "Epoch: 34 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18504.23 .. KL_eta: 429.66 .. KL_alpha: 2008982.33 .. Rec_loss: 17581726.27 .. NELBO: 19609642.36\n",
      "****************************************************************************************************\n",
      "Epoch----->34 .. LR: 0.005 .. KL_theta: 19231.37 .. KL_eta: 438.05 .. KL_alpha: 1994711.54 .. Rec_loss: 17465672.24 .. NELBO: 19480052.94\n",
      "****************************************************************************************************\n",
      "Epoch: 35 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19008.0 .. KL_eta: 432.86 .. KL_alpha: 1947227.82 .. Rec_loss: 17293488.09 .. NELBO: 19260156.55\n",
      "****************************************************************************************************\n",
      "Epoch----->35 .. LR: 0.005 .. KL_theta: 19041.42 .. KL_eta: 438.9 .. KL_alpha: 1935828.35 .. Rec_loss: 17391397.71 .. NELBO: 19346706.12\n",
      "****************************************************************************************************\n",
      "Epoch: 36 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 18776.34 .. KL_eta: 473.44 .. KL_alpha: 1879891.58 .. Rec_loss: 17313582.09 .. NELBO: 19212722.91\n",
      "****************************************************************************************************\n",
      "Epoch----->36 .. LR: 0.005 .. KL_theta: 19419.96 .. KL_eta: 494.67 .. KL_alpha: 1871150.18 .. Rec_loss: 17412584.0 .. NELBO: 19303648.59\n",
      "****************************************************************************************************\n",
      "Epoch: 37 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 17875.25 .. KL_eta: 485.88 .. KL_alpha: 1825153.92 .. Rec_loss: 17075233.27 .. NELBO: 18918748.18\n",
      "****************************************************************************************************\n",
      "Epoch----->37 .. LR: 0.005 .. KL_theta: 18069.34 .. KL_eta: 461.43 .. KL_alpha: 1818311.51 .. Rec_loss: 17322659.76 .. NELBO: 19159501.88\n",
      "****************************************************************************************************\n",
      "Epoch: 38 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19898.62 .. KL_eta: 398.87 .. KL_alpha: 1771899.22 .. Rec_loss: 17549244.91 .. NELBO: 19341441.09\n",
      "****************************************************************************************************\n",
      "Epoch----->38 .. LR: 0.005 .. KL_theta: 19754.95 .. KL_eta: 399.3 .. KL_alpha: 1763245.95 .. Rec_loss: 17353315.82 .. NELBO: 19136715.53\n",
      "****************************************************************************************************\n",
      "Epoch: 39 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19831.21 .. KL_eta: 359.67 .. KL_alpha: 1719428.69 .. Rec_loss: 17255639.64 .. NELBO: 18995258.91\n",
      "****************************************************************************************************\n",
      "Epoch----->39 .. LR: 0.005 .. KL_theta: 20342.43 .. KL_eta: 344.64 .. KL_alpha: 1710462.35 .. Rec_loss: 17344251.59 .. NELBO: 19075400.76\n",
      "****************************************************************************************************\n",
      "Epoch: 40 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19491.28 .. KL_eta: 352.42 .. KL_alpha: 1672672.78 .. Rec_loss: 17400766.55 .. NELBO: 19093283.09\n",
      "****************************************************************************************************\n",
      "Epoch----->40 .. LR: 0.005 .. KL_theta: 19703.02 .. KL_eta: 348.25 .. KL_alpha: 1660667.78 .. Rec_loss: 17252619.41 .. NELBO: 18933338.59\n",
      "****************************************************************************************************\n",
      "Epoch: 41 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20089.24 .. KL_eta: 279.6 .. KL_alpha: 1622847.12 .. Rec_loss: 17268208.73 .. NELBO: 18911425.09\n",
      "****************************************************************************************************\n",
      "Epoch----->41 .. LR: 0.005 .. KL_theta: 20347.74 .. KL_eta: 272.79 .. KL_alpha: 1615259.99 .. Rec_loss: 17233992.0 .. NELBO: 18869872.82\n",
      "****************************************************************************************************\n",
      "Epoch: 42 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20240.45 .. KL_eta: 277.54 .. KL_alpha: 1574382.38 .. Rec_loss: 17047783.36 .. NELBO: 18642683.27\n",
      "****************************************************************************************************\n",
      "Epoch----->42 .. LR: 0.005 .. KL_theta: 20098.48 .. KL_eta: 298.3 .. KL_alpha: 1565251.65 .. Rec_loss: 17330484.0 .. NELBO: 18916132.12\n",
      "****************************************************************************************************\n",
      "Epoch: 43 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 19590.12 .. KL_eta: 353.78 .. KL_alpha: 1528614.38 .. Rec_loss: 16822015.09 .. NELBO: 18370573.36\n",
      "****************************************************************************************************\n",
      "Epoch----->43 .. LR: 0.005 .. KL_theta: 19916.76 .. KL_eta: 360.75 .. KL_alpha: 1521286.17 .. Rec_loss: 17265250.24 .. NELBO: 18806813.94\n",
      "****************************************************************************************************\n",
      "Epoch: 44 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20555.74 .. KL_eta: 386.65 .. KL_alpha: 1490341.02 .. Rec_loss: 17138501.27 .. NELBO: 18649784.55\n",
      "****************************************************************************************************\n",
      "Epoch----->44 .. LR: 0.005 .. KL_theta: 20322.29 .. KL_eta: 396.94 .. KL_alpha: 1483589.42 .. Rec_loss: 17204456.65 .. NELBO: 18708765.18\n",
      "****************************************************************************************************\n",
      "Epoch: 45 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20613.5 .. KL_eta: 415.79 .. KL_alpha: 1452462.22 .. Rec_loss: 17470799.55 .. NELBO: 18944290.36\n",
      "****************************************************************************************************\n",
      "Epoch----->45 .. LR: 0.005 .. KL_theta: 20080.3 .. KL_eta: 408.03 .. KL_alpha: 1446415.91 .. Rec_loss: 17162064.41 .. NELBO: 18628968.24\n",
      "****************************************************************************************************\n",
      "Epoch: 46 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20628.88 .. KL_eta: 349.65 .. KL_alpha: 1415196.99 .. Rec_loss: 17159023.36 .. NELBO: 18595198.91\n",
      "****************************************************************************************************\n",
      "Epoch----->46 .. LR: 0.005 .. KL_theta: 21003.7 .. KL_eta: 339.48 .. KL_alpha: 1407589.76 .. Rec_loss: 17165905.82 .. NELBO: 18594838.88\n",
      "****************************************************************************************************\n",
      "Epoch: 47 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21419.54 .. KL_eta: 390.07 .. KL_alpha: 1377055.85 .. Rec_loss: 17328654.45 .. NELBO: 18727519.45\n",
      "****************************************************************************************************\n",
      "Epoch----->47 .. LR: 0.005 .. KL_theta: 20992.94 .. KL_eta: 401.03 .. KL_alpha: 1367897.27 .. Rec_loss: 17132119.24 .. NELBO: 18521410.24\n",
      "****************************************************************************************************\n",
      "Epoch: 48 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20055.95 .. KL_eta: 341.86 .. KL_alpha: 1338996.89 .. Rec_loss: 17073745.64 .. NELBO: 18433140.36\n",
      "****************************************************************************************************\n",
      "Epoch----->48 .. LR: 0.005 .. KL_theta: 20702.6 .. KL_eta: 332.38 .. KL_alpha: 1336267.89 .. Rec_loss: 17219417.0 .. NELBO: 18576720.12\n",
      "****************************************************************************************************\n",
      "Epoch: 49 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 20913.89 .. KL_eta: 377.64 .. KL_alpha: 1304778.55 .. Rec_loss: 17204381.55 .. NELBO: 18530451.36\n",
      "****************************************************************************************************\n",
      "Epoch----->49 .. LR: 0.005 .. KL_theta: 20918.18 .. KL_eta: 376.26 .. KL_alpha: 1300400.52 .. Rec_loss: 17150499.47 .. NELBO: 18472194.65\n",
      "****************************************************************************************************\n",
      "Epoch: 50 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21012.53 .. KL_eta: 334.81 .. KL_alpha: 1271868.5 .. Rec_loss: 16997691.27 .. NELBO: 18290907.45\n",
      "****************************************************************************************************\n",
      "Epoch----->50 .. LR: 0.005 .. KL_theta: 21268.98 .. KL_eta: 313.91 .. KL_alpha: 1265796.65 .. Rec_loss: 17164125.71 .. NELBO: 18451505.41\n",
      "****************************************************************************************************\n",
      "Epoch: 51 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22026.62 .. KL_eta: 299.78 .. KL_alpha: 1244066.86 .. Rec_loss: 17399651.73 .. NELBO: 18666045.09\n",
      "****************************************************************************************************\n",
      "Epoch----->51 .. LR: 0.005 .. KL_theta: 21653.66 .. KL_eta: 315.26 .. KL_alpha: 1237438.31 .. Rec_loss: 17089586.65 .. NELBO: 18348994.18\n",
      "****************************************************************************************************\n",
      "Epoch: 52 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21002.33 .. KL_eta: 291.14 .. KL_alpha: 1211429.66 .. Rec_loss: 17214277.45 .. NELBO: 18447000.36\n",
      "****************************************************************************************************\n",
      "Epoch----->52 .. LR: 0.005 .. KL_theta: 21199.08 .. KL_eta: 277.16 .. KL_alpha: 1206959.06 .. Rec_loss: 17100058.82 .. NELBO: 18328493.94\n",
      "****************************************************************************************************\n",
      "Epoch: 53 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22607.46 .. KL_eta: 238.45 .. KL_alpha: 1175728.09 .. Rec_loss: 17293688.64 .. NELBO: 18492262.55\n",
      "****************************************************************************************************\n",
      "Epoch----->53 .. LR: 0.005 .. KL_theta: 21835.56 .. KL_eta: 248.86 .. KL_alpha: 1174281.55 .. Rec_loss: 17060557.0 .. NELBO: 18256923.0\n",
      "****************************************************************************************************\n",
      "Epoch: 54 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22515.98 .. KL_eta: 273.99 .. KL_alpha: 1153305.43 .. Rec_loss: 17110348.45 .. NELBO: 18286443.82\n",
      "****************************************************************************************************\n",
      "Epoch----->54 .. LR: 0.005 .. KL_theta: 22184.03 .. KL_eta: 271.25 .. KL_alpha: 1150644.87 .. Rec_loss: 17114317.41 .. NELBO: 18287417.53\n",
      "****************************************************************************************************\n",
      "Epoch: 55 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21765.74 .. KL_eta: 288.43 .. KL_alpha: 1122524.61 .. Rec_loss: 17224766.09 .. NELBO: 18369345.09\n",
      "****************************************************************************************************\n",
      "Epoch----->55 .. LR: 0.005 .. KL_theta: 21865.78 .. KL_eta: 286.72 .. KL_alpha: 1120600.3 .. Rec_loss: 17061763.41 .. NELBO: 18204516.24\n",
      "****************************************************************************************************\n",
      "Epoch: 56 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21812.53 .. KL_eta: 238.5 .. KL_alpha: 1102623.59 .. Rec_loss: 17060603.18 .. NELBO: 18185277.45\n",
      "****************************************************************************************************\n",
      "Epoch----->56 .. LR: 0.005 .. KL_theta: 21925.15 .. KL_eta: 236.72 .. KL_alpha: 1096292.28 .. Rec_loss: 17061484.41 .. NELBO: 18179938.24\n",
      "****************************************************************************************************\n",
      "Epoch: 57 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22257.34 .. KL_eta: 227.41 .. KL_alpha: 1080330.36 .. Rec_loss: 17273139.82 .. NELBO: 18375955.09\n",
      "****************************************************************************************************\n",
      "Epoch----->57 .. LR: 0.005 .. KL_theta: 22869.75 .. KL_eta: 235.8 .. KL_alpha: 1073094.41 .. Rec_loss: 17041154.35 .. NELBO: 18137354.24\n",
      "****************************************************************************************************\n",
      "Epoch: 58 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 21664.33 .. KL_eta: 229.78 .. KL_alpha: 1050415.23 .. Rec_loss: 16832020.73 .. NELBO: 17904329.64\n",
      "****************************************************************************************************\n",
      "Epoch----->58 .. LR: 0.005 .. KL_theta: 22411.61 .. KL_eta: 239.0 .. KL_alpha: 1046133.39 .. Rec_loss: 17068241.24 .. NELBO: 18137025.0\n",
      "****************************************************************************************************\n",
      "Epoch: 59 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22573.64 .. KL_eta: 263.71 .. KL_alpha: 1028995.23 .. Rec_loss: 16822311.91 .. NELBO: 17874144.55\n",
      "****************************************************************************************************\n",
      "Epoch----->59 .. LR: 0.005 .. KL_theta: 22707.31 .. KL_eta: 254.1 .. KL_alpha: 1024441.99 .. Rec_loss: 17067004.29 .. NELBO: 18114407.76\n",
      "****************************************************************************************************\n",
      "Epoch: 60 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22693.9 .. KL_eta: 232.02 .. KL_alpha: 1003369.75 .. Rec_loss: 16957527.09 .. NELBO: 17983823.36\n",
      "****************************************************************************************************\n",
      "Epoch----->60 .. LR: 0.005 .. KL_theta: 22360.61 .. KL_eta: 230.99 .. KL_alpha: 1000993.73 .. Rec_loss: 17024958.88 .. NELBO: 18048544.53\n",
      "****************************************************************************************************\n",
      "Epoch: 61 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 23500.1 .. KL_eta: 228.59 .. KL_alpha: 983001.06 .. Rec_loss: 17014814.18 .. NELBO: 18021544.45\n",
      "****************************************************************************************************\n",
      "Epoch----->61 .. LR: 0.005 .. KL_theta: 23280.62 .. KL_eta: 234.53 .. KL_alpha: 978481.46 .. Rec_loss: 17068350.0 .. NELBO: 18070346.88\n",
      "****************************************************************************************************\n",
      "Epoch: 62 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 23368.03 .. KL_eta: 240.27 .. KL_alpha: 962793.42 .. Rec_loss: 17127462.82 .. NELBO: 18113864.45\n",
      "****************************************************************************************************\n",
      "Epoch----->62 .. LR: 0.005 .. KL_theta: 23054.26 .. KL_eta: 239.09 .. KL_alpha: 957107.47 .. Rec_loss: 17024410.18 .. NELBO: 18004811.0\n",
      "****************************************************************************************************\n",
      "Epoch: 63 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 23475.85 .. KL_eta: 229.6 .. KL_alpha: 939188.94 .. Rec_loss: 17205600.36 .. NELBO: 18168495.09\n",
      "****************************************************************************************************\n",
      "Epoch----->63 .. LR: 0.005 .. KL_theta: 23231.14 .. KL_eta: 225.84 .. KL_alpha: 934562.61 .. Rec_loss: 16987335.59 .. NELBO: 17945355.18\n",
      "****************************************************************************************************\n",
      "Epoch: 64 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 22951.43 .. KL_eta: 218.22 .. KL_alpha: 920725.89 .. Rec_loss: 17381069.36 .. NELBO: 18324964.45\n",
      "****************************************************************************************************\n",
      "Epoch----->64 .. LR: 0.005 .. KL_theta: 23316.78 .. KL_eta: 210.33 .. KL_alpha: 916387.64 .. Rec_loss: 16953073.53 .. NELBO: 17892988.35\n",
      "****************************************************************************************************\n",
      "Epoch: 65 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 23529.3 .. KL_eta: 206.39 .. KL_alpha: 901017.76 .. Rec_loss: 16827329.82 .. NELBO: 17752083.18\n",
      "****************************************************************************************************\n",
      "Epoch----->65 .. LR: 0.005 .. KL_theta: 23150.77 .. KL_eta: 199.67 .. KL_alpha: 898854.04 .. Rec_loss: 16957148.06 .. NELBO: 17879352.41\n",
      "****************************************************************************************************\n",
      "Epoch: 66 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24160.66 .. KL_eta: 183.86 .. KL_alpha: 880710.84 .. Rec_loss: 17115698.64 .. NELBO: 18020753.64\n",
      "****************************************************************************************************\n",
      "Epoch----->66 .. LR: 0.005 .. KL_theta: 23948.15 .. KL_eta: 184.13 .. KL_alpha: 879122.84 .. Rec_loss: 16986978.71 .. NELBO: 17890233.65\n",
      "****************************************************************************************************\n",
      "Epoch: 67 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25064.59 .. KL_eta: 188.23 .. KL_alpha: 866788.7 .. Rec_loss: 17019523.18 .. NELBO: 17911564.64\n",
      "****************************************************************************************************\n",
      "Epoch----->67 .. LR: 0.005 .. KL_theta: 24182.65 .. KL_eta: 197.51 .. KL_alpha: 864486.62 .. Rec_loss: 16965776.76 .. NELBO: 17854643.71\n",
      "****************************************************************************************************\n",
      "Epoch: 68 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 23574.3 .. KL_eta: 194.08 .. KL_alpha: 848631.13 .. Rec_loss: 16826966.82 .. NELBO: 17699366.64\n",
      "****************************************************************************************************\n",
      "Epoch----->68 .. LR: 0.005 .. KL_theta: 23701.37 .. KL_eta: 191.53 .. KL_alpha: 845355.95 .. Rec_loss: 16951006.18 .. NELBO: 17820255.0\n",
      "****************************************************************************************************\n",
      "Epoch: 69 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 23881.28 .. KL_eta: 173.93 .. KL_alpha: 830144.44 .. Rec_loss: 16943233.55 .. NELBO: 17797432.73\n",
      "****************************************************************************************************\n",
      "Epoch----->69 .. LR: 0.005 .. KL_theta: 24362.48 .. KL_eta: 170.41 .. KL_alpha: 826653.81 .. Rec_loss: 16893925.35 .. NELBO: 17745112.06\n",
      "****************************************************************************************************\n",
      "Epoch: 70 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24868.4 .. KL_eta: 181.94 .. KL_alpha: 810765.16 .. Rec_loss: 17010770.0 .. NELBO: 17846586.18\n",
      "****************************************************************************************************\n",
      "Epoch----->70 .. LR: 0.005 .. KL_theta: 24352.0 .. KL_eta: 181.17 .. KL_alpha: 808214.42 .. Rec_loss: 16930164.06 .. NELBO: 17762912.29\n",
      "****************************************************************************************************\n",
      "Epoch: 71 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24655.3 .. KL_eta: 170.32 .. KL_alpha: 795919.19 .. Rec_loss: 16969024.55 .. NELBO: 17789769.73\n",
      "****************************************************************************************************\n",
      "Epoch----->71 .. LR: 0.005 .. KL_theta: 24332.46 .. KL_eta: 174.57 .. KL_alpha: 795400.55 .. Rec_loss: 16899935.76 .. NELBO: 17719843.82\n",
      "****************************************************************************************************\n",
      "Epoch: 72 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24630.99 .. KL_eta: 179.26 .. KL_alpha: 779705.19 .. Rec_loss: 16947419.0 .. NELBO: 17751934.73\n",
      "****************************************************************************************************\n",
      "Epoch----->72 .. LR: 0.005 .. KL_theta: 24230.78 .. KL_eta: 178.1 .. KL_alpha: 778690.02 .. Rec_loss: 16881023.65 .. NELBO: 17684122.59\n",
      "****************************************************************************************************\n",
      "Epoch: 73 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24840.59 .. KL_eta: 153.57 .. KL_alpha: 769863.88 .. Rec_loss: 17039941.45 .. NELBO: 17834799.18\n",
      "****************************************************************************************************\n",
      "Epoch----->73 .. LR: 0.005 .. KL_theta: 24891.91 .. KL_eta: 142.29 .. KL_alpha: 765280.03 .. Rec_loss: 16913419.53 .. NELBO: 17703733.47\n",
      "****************************************************************************************************\n",
      "Epoch: 74 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24980.81 .. KL_eta: 130.01 .. KL_alpha: 749439.22 .. Rec_loss: 17038402.91 .. NELBO: 17812952.82\n",
      "****************************************************************************************************\n",
      "Epoch----->74 .. LR: 0.005 .. KL_theta: 24805.78 .. KL_eta: 137.21 .. KL_alpha: 745637.25 .. Rec_loss: 16896608.41 .. NELBO: 17667188.59\n",
      "****************************************************************************************************\n",
      "Epoch: 75 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24459.67 .. KL_eta: 149.13 .. KL_alpha: 737902.38 .. Rec_loss: 17028785.0 .. NELBO: 17791295.91\n",
      "****************************************************************************************************\n",
      "Epoch----->75 .. LR: 0.005 .. KL_theta: 24475.74 .. KL_eta: 151.66 .. KL_alpha: 736164.25 .. Rec_loss: 16845984.12 .. NELBO: 17606775.41\n",
      "****************************************************************************************************\n",
      "Epoch: 76 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24850.43 .. KL_eta: 152.72 .. KL_alpha: 721809.83 .. Rec_loss: 17015828.27 .. NELBO: 17762641.27\n",
      "****************************************************************************************************\n",
      "Epoch----->76 .. LR: 0.005 .. KL_theta: 24902.24 .. KL_eta: 152.93 .. KL_alpha: 719931.27 .. Rec_loss: 16925414.71 .. NELBO: 17670401.47\n",
      "****************************************************************************************************\n",
      "Epoch: 77 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24816.14 .. KL_eta: 140.22 .. KL_alpha: 707429.84 .. Rec_loss: 16829759.27 .. NELBO: 17562145.09\n",
      "****************************************************************************************************\n",
      "Epoch----->77 .. LR: 0.005 .. KL_theta: 24799.69 .. KL_eta: 137.91 .. KL_alpha: 706862.02 .. Rec_loss: 16905871.53 .. NELBO: 17637670.94\n",
      "****************************************************************************************************\n",
      "Epoch: 78 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24749.74 .. KL_eta: 146.99 .. KL_alpha: 696829.82 .. Rec_loss: 17125350.64 .. NELBO: 17847077.18\n",
      "****************************************************************************************************\n",
      "Epoch----->78 .. LR: 0.005 .. KL_theta: 24806.95 .. KL_eta: 148.69 .. KL_alpha: 693564.46 .. Rec_loss: 16869692.47 .. NELBO: 17588212.53\n",
      "****************************************************************************************************\n",
      "Epoch: 79 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25234.15 .. KL_eta: 146.71 .. KL_alpha: 686014.95 .. Rec_loss: 17004222.45 .. NELBO: 17715618.64\n",
      "****************************************************************************************************\n",
      "Epoch----->79 .. LR: 0.005 .. KL_theta: 25275.05 .. KL_eta: 141.36 .. KL_alpha: 683102.18 .. Rec_loss: 16831973.24 .. NELBO: 17540492.0\n",
      "****************************************************************************************************\n",
      "Epoch: 80 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 24573.39 .. KL_eta: 133.17 .. KL_alpha: 670855.1 .. Rec_loss: 16897623.91 .. NELBO: 17593186.0\n",
      "****************************************************************************************************\n",
      "Epoch----->80 .. LR: 0.005 .. KL_theta: 24897.4 .. KL_eta: 133.31 .. KL_alpha: 668735.0 .. Rec_loss: 16811704.24 .. NELBO: 17505470.12\n",
      "****************************************************************************************************\n",
      "Epoch: 81 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26022.97 .. KL_eta: 137.48 .. KL_alpha: 656151.61 .. Rec_loss: 17014942.55 .. NELBO: 17697254.36\n",
      "****************************************************************************************************\n",
      "Epoch----->81 .. LR: 0.005 .. KL_theta: 25432.95 .. KL_eta: 134.34 .. KL_alpha: 653556.03 .. Rec_loss: 16856380.29 .. NELBO: 17535503.41\n",
      "****************************************************************************************************\n",
      "Epoch: 82 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25932.79 .. KL_eta: 130.48 .. KL_alpha: 645158.14 .. Rec_loss: 16820493.45 .. NELBO: 17491714.82\n",
      "****************************************************************************************************\n",
      "Epoch----->82 .. LR: 0.005 .. KL_theta: 25451.74 .. KL_eta: 129.95 .. KL_alpha: 644252.81 .. Rec_loss: 16840071.18 .. NELBO: 17509905.82\n",
      "****************************************************************************************************\n",
      "Epoch: 83 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25534.34 .. KL_eta: 123.98 .. KL_alpha: 631942.32 .. Rec_loss: 16718107.45 .. NELBO: 17375708.09\n",
      "****************************************************************************************************\n",
      "Epoch----->83 .. LR: 0.005 .. KL_theta: 25871.99 .. KL_eta: 123.46 .. KL_alpha: 631122.89 .. Rec_loss: 16852643.0 .. NELBO: 17509761.53\n",
      "****************************************************************************************************\n",
      "Epoch: 84 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25699.33 .. KL_eta: 120.16 .. KL_alpha: 622648.6 .. Rec_loss: 16776086.64 .. NELBO: 17424554.73\n",
      "****************************************************************************************************\n",
      "Epoch----->84 .. LR: 0.005 .. KL_theta: 25640.33 .. KL_eta: 119.18 .. KL_alpha: 621220.65 .. Rec_loss: 16845673.24 .. NELBO: 17492653.29\n",
      "****************************************************************************************************\n",
      "Epoch: 85 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25515.44 .. KL_eta: 113.47 .. KL_alpha: 613848.02 .. Rec_loss: 16822958.91 .. NELBO: 17462435.91\n",
      "****************************************************************************************************\n",
      "Epoch----->85 .. LR: 0.005 .. KL_theta: 25448.11 .. KL_eta: 111.51 .. KL_alpha: 610497.43 .. Rec_loss: 16850589.0 .. NELBO: 17486645.94\n",
      "****************************************************************************************************\n",
      "Epoch: 86 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26086.1 .. KL_eta: 106.08 .. KL_alpha: 602800.54 .. Rec_loss: 16711977.09 .. NELBO: 17340969.91\n",
      "****************************************************************************************************\n",
      "Epoch----->86 .. LR: 0.005 .. KL_theta: 25897.53 .. KL_eta: 106.63 .. KL_alpha: 600265.67 .. Rec_loss: 16788109.94 .. NELBO: 17414379.88\n",
      "****************************************************************************************************\n",
      "Epoch: 87 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26322.37 .. KL_eta: 124.11 .. KL_alpha: 591526.9 .. Rec_loss: 17028549.45 .. NELBO: 17646522.64\n",
      "****************************************************************************************************\n",
      "Epoch----->87 .. LR: 0.005 .. KL_theta: 25970.9 .. KL_eta: 123.15 .. KL_alpha: 588745.64 .. Rec_loss: 16747003.0 .. NELBO: 17361842.65\n",
      "****************************************************************************************************\n",
      "Epoch: 88 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25973.58 .. KL_eta: 113.9 .. KL_alpha: 578711.12 .. Rec_loss: 16962704.0 .. NELBO: 17567502.82\n",
      "****************************************************************************************************\n",
      "Epoch----->88 .. LR: 0.005 .. KL_theta: 25761.18 .. KL_eta: 114.4 .. KL_alpha: 577382.31 .. Rec_loss: 16795765.47 .. NELBO: 17399023.41\n",
      "****************************************************************************************************\n",
      "Epoch: 89 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26285.16 .. KL_eta: 116.07 .. KL_alpha: 570565.27 .. Rec_loss: 16996406.55 .. NELBO: 17593373.09\n",
      "****************************************************************************************************\n",
      "Epoch----->89 .. LR: 0.005 .. KL_theta: 26010.62 .. KL_eta: 113.51 .. KL_alpha: 569029.01 .. Rec_loss: 16754983.47 .. NELBO: 17350136.65\n",
      "****************************************************************************************************\n",
      "Epoch: 90 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26029.81 .. KL_eta: 107.55 .. KL_alpha: 562604.81 .. Rec_loss: 16917508.0 .. NELBO: 17506250.09\n",
      "****************************************************************************************************\n",
      "Epoch----->90 .. LR: 0.005 .. KL_theta: 26028.95 .. KL_eta: 107.53 .. KL_alpha: 559714.4 .. Rec_loss: 16744169.59 .. NELBO: 17330020.65\n",
      "****************************************************************************************************\n",
      "Epoch: 91 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26505.57 .. KL_eta: 98.39 .. KL_alpha: 551063.59 .. Rec_loss: 16687661.27 .. NELBO: 17265328.18\n",
      "****************************************************************************************************\n",
      "Epoch----->91 .. LR: 0.005 .. KL_theta: 26491.21 .. KL_eta: 99.68 .. KL_alpha: 550564.47 .. Rec_loss: 16847161.29 .. NELBO: 17424316.0\n",
      "****************************************************************************************************\n",
      "Epoch: 92 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25960.67 .. KL_eta: 107.59 .. KL_alpha: 545186.72 .. Rec_loss: 16769449.55 .. NELBO: 17340704.36\n",
      "****************************************************************************************************\n",
      "Epoch----->92 .. LR: 0.005 .. KL_theta: 25995.9 .. KL_eta: 107.87 .. KL_alpha: 542027.69 .. Rec_loss: 16844637.18 .. NELBO: 17412768.53\n",
      "****************************************************************************************************\n",
      "Epoch: 93 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 25931.12 .. KL_eta: 113.88 .. KL_alpha: 531744.52 .. Rec_loss: 16913661.27 .. NELBO: 17471451.0\n",
      "****************************************************************************************************\n",
      "Epoch----->93 .. LR: 0.005 .. KL_theta: 26078.59 .. KL_eta: 114.38 .. KL_alpha: 530429.72 .. Rec_loss: 16757832.71 .. NELBO: 17314455.65\n",
      "****************************************************************************************************\n",
      "Epoch: 94 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26910.46 .. KL_eta: 95.67 .. KL_alpha: 523398.96 .. Rec_loss: 16882155.27 .. NELBO: 17432560.64\n",
      "****************************************************************************************************\n",
      "Epoch----->94 .. LR: 0.005 .. KL_theta: 26879.59 .. KL_eta: 92.96 .. KL_alpha: 522774.38 .. Rec_loss: 16808857.47 .. NELBO: 17358604.53\n",
      "****************************************************************************************************\n",
      "Epoch: 95 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26653.45 .. KL_eta: 96.25 .. KL_alpha: 516077.8 .. Rec_loss: 16971031.91 .. NELBO: 17513860.0\n",
      "****************************************************************************************************\n",
      "Epoch----->95 .. LR: 0.005 .. KL_theta: 26452.01 .. KL_eta: 96.97 .. KL_alpha: 514045.97 .. Rec_loss: 16725341.41 .. NELBO: 17265937.0\n",
      "****************************************************************************************************\n",
      "Epoch: 96 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26490.26 .. KL_eta: 99.02 .. KL_alpha: 507497.03 .. Rec_loss: 16928484.0 .. NELBO: 17462570.64\n",
      "****************************************************************************************************\n",
      "Epoch----->96 .. LR: 0.005 .. KL_theta: 26382.92 .. KL_eta: 98.83 .. KL_alpha: 506302.15 .. Rec_loss: 16683523.65 .. NELBO: 17216307.76\n",
      "****************************************************************************************************\n",
      "Epoch: 97 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27074.81 .. KL_eta: 102.18 .. KL_alpha: 498307.52 .. Rec_loss: 16793616.55 .. NELBO: 17319101.18\n",
      "****************************************************************************************************\n",
      "Epoch----->97 .. LR: 0.005 .. KL_theta: 26640.95 .. KL_eta: 102.49 .. KL_alpha: 496904.33 .. Rec_loss: 16702043.35 .. NELBO: 17225691.29\n",
      "****************************************************************************************************\n",
      "Epoch: 98 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26550.17 .. KL_eta: 104.92 .. KL_alpha: 490907.09 .. Rec_loss: 16935412.55 .. NELBO: 17452974.82\n",
      "****************************************************************************************************\n",
      "Epoch----->98 .. LR: 0.005 .. KL_theta: 26529.47 .. KL_eta: 103.83 .. KL_alpha: 489465.66 .. Rec_loss: 16688498.76 .. NELBO: 17204597.76\n",
      "****************************************************************************************************\n",
      "Epoch: 99 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26751.6 .. KL_eta: 107.17 .. KL_alpha: 481780.02 .. Rec_loss: 16491684.55 .. NELBO: 17000323.82\n",
      "****************************************************************************************************\n",
      "Epoch----->99 .. LR: 0.005 .. KL_theta: 27039.38 .. KL_eta: 107.77 .. KL_alpha: 480417.61 .. Rec_loss: 16715977.76 .. NELBO: 17223542.76\n",
      "****************************************************************************************************\n",
      "Epoch: 100 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26598.92 .. KL_eta: 102.94 .. KL_alpha: 474843.99 .. Rec_loss: 16689941.09 .. NELBO: 17191486.64\n",
      "****************************************************************************************************\n",
      "Epoch----->100 .. LR: 0.005 .. KL_theta: 26713.38 .. KL_eta: 100.38 .. KL_alpha: 472973.36 .. Rec_loss: 16656726.94 .. NELBO: 17156513.76\n",
      "****************************************************************************************************\n",
      "Epoch: 101 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27021.24 .. KL_eta: 97.56 .. KL_alpha: 469374.66 .. Rec_loss: 17031255.09 .. NELBO: 17527748.36\n",
      "****************************************************************************************************\n",
      "Epoch----->101 .. LR: 0.005 .. KL_theta: 27061.43 .. KL_eta: 98.59 .. KL_alpha: 467037.51 .. Rec_loss: 16735917.12 .. NELBO: 17230114.53\n",
      "****************************************************************************************************\n",
      "Epoch: 102 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26306.47 .. KL_eta: 103.97 .. KL_alpha: 461284.17 .. Rec_loss: 16791600.09 .. NELBO: 17279294.45\n",
      "****************************************************************************************************\n",
      "Epoch----->102 .. LR: 0.005 .. KL_theta: 26563.0 .. KL_eta: 103.89 .. KL_alpha: 459899.71 .. Rec_loss: 16695577.88 .. NELBO: 17182144.41\n",
      "****************************************************************************************************\n",
      "Epoch: 103 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27258.49 .. KL_eta: 92.82 .. KL_alpha: 453877.1 .. Rec_loss: 16716589.18 .. NELBO: 17197817.82\n",
      "****************************************************************************************************\n",
      "Epoch----->103 .. LR: 0.005 .. KL_theta: 27064.23 .. KL_eta: 93.07 .. KL_alpha: 452216.09 .. Rec_loss: 16703822.82 .. NELBO: 17183196.41\n",
      "****************************************************************************************************\n",
      "Epoch: 104 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27434.15 .. KL_eta: 95.21 .. KL_alpha: 444584.7 .. Rec_loss: 16581628.73 .. NELBO: 17053743.36\n",
      "****************************************************************************************************\n",
      "Epoch----->104 .. LR: 0.005 .. KL_theta: 26957.4 .. KL_eta: 95.77 .. KL_alpha: 443877.07 .. Rec_loss: 16688320.47 .. NELBO: 17159251.29\n",
      "****************************************************************************************************\n",
      "Epoch: 105 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26732.1 .. KL_eta: 93.1 .. KL_alpha: 439981.39 .. Rec_loss: 16614761.55 .. NELBO: 17081568.27\n",
      "****************************************************************************************************\n",
      "Epoch----->105 .. LR: 0.005 .. KL_theta: 27250.27 .. KL_eta: 90.75 .. KL_alpha: 438381.69 .. Rec_loss: 16687984.0 .. NELBO: 17153706.65\n",
      "****************************************************************************************************\n",
      "Epoch: 106 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27189.35 .. KL_eta: 87.0 .. KL_alpha: 434524.88 .. Rec_loss: 16576541.09 .. NELBO: 17038342.18\n",
      "****************************************************************************************************\n",
      "Epoch----->106 .. LR: 0.005 .. KL_theta: 27447.63 .. KL_eta: 89.63 .. KL_alpha: 432549.77 .. Rec_loss: 16672809.12 .. NELBO: 17132895.88\n",
      "****************************************************************************************************\n",
      "Epoch: 107 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26712.86 .. KL_eta: 97.99 .. KL_alpha: 426457.6 .. Rec_loss: 16427383.18 .. NELBO: 16880651.27\n",
      "****************************************************************************************************\n",
      "Epoch----->107 .. LR: 0.005 .. KL_theta: 27275.57 .. KL_eta: 99.72 .. KL_alpha: 425940.66 .. Rec_loss: 16682410.88 .. NELBO: 17135726.65\n",
      "****************************************************************************************************\n",
      "Epoch: 108 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26919.49 .. KL_eta: 88.63 .. KL_alpha: 420278.12 .. Rec_loss: 16795193.91 .. NELBO: 17242480.09\n",
      "****************************************************************************************************\n",
      "Epoch----->108 .. LR: 0.005 .. KL_theta: 27042.41 .. KL_eta: 88.48 .. KL_alpha: 419724.49 .. Rec_loss: 16659202.0 .. NELBO: 17106057.47\n",
      "****************************************************************************************************\n",
      "Epoch: 109 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27205.18 .. KL_eta: 98.64 .. KL_alpha: 414837.83 .. Rec_loss: 16464398.36 .. NELBO: 16906540.0\n",
      "****************************************************************************************************\n",
      "Epoch----->109 .. LR: 0.005 .. KL_theta: 27183.89 .. KL_eta: 100.62 .. KL_alpha: 413640.6 .. Rec_loss: 16695688.76 .. NELBO: 17136613.65\n",
      "****************************************************************************************************\n",
      "Epoch: 110 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27402.8 .. KL_eta: 99.19 .. KL_alpha: 409959.89 .. Rec_loss: 16941847.73 .. NELBO: 17379309.82\n",
      "****************************************************************************************************\n",
      "Epoch----->110 .. LR: 0.005 .. KL_theta: 27689.59 .. KL_eta: 97.96 .. KL_alpha: 407991.95 .. Rec_loss: 16648206.24 .. NELBO: 17083985.94\n",
      "****************************************************************************************************\n",
      "Epoch: 111 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27086.1 .. KL_eta: 95.37 .. KL_alpha: 404906.89 .. Rec_loss: 16710564.91 .. NELBO: 17142653.36\n",
      "****************************************************************************************************\n",
      "Epoch----->111 .. LR: 0.005 .. KL_theta: 27233.83 .. KL_eta: 95.95 .. KL_alpha: 403130.14 .. Rec_loss: 16681604.35 .. NELBO: 17112064.41\n",
      "****************************************************************************************************\n",
      "Epoch: 112 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 26989.46 .. KL_eta: 86.2 .. KL_alpha: 397539.33 .. Rec_loss: 16681531.27 .. NELBO: 17106146.55\n",
      "****************************************************************************************************\n",
      "Epoch----->112 .. LR: 0.005 .. KL_theta: 27457.12 .. KL_eta: 84.59 .. KL_alpha: 395994.46 .. Rec_loss: 16653319.12 .. NELBO: 17076855.47\n",
      "****************************************************************************************************\n",
      "Epoch: 113 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27332.62 .. KL_eta: 88.09 .. KL_alpha: 391031.08 .. Rec_loss: 16687493.36 .. NELBO: 17105944.91\n",
      "****************************************************************************************************\n",
      "Epoch----->113 .. LR: 0.005 .. KL_theta: 27777.0 .. KL_eta: 90.38 .. KL_alpha: 390731.37 .. Rec_loss: 16695756.94 .. NELBO: 17114355.59\n",
      "****************************************************************************************************\n",
      "Epoch: 114 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27395.93 .. KL_eta: 95.39 .. KL_alpha: 386285.74 .. Rec_loss: 16821941.09 .. NELBO: 17235718.36\n",
      "****************************************************************************************************\n",
      "Epoch----->114 .. LR: 0.005 .. KL_theta: 27260.37 .. KL_eta: 95.82 .. KL_alpha: 384875.48 .. Rec_loss: 16577765.88 .. NELBO: 16989997.71\n",
      "****************************************************************************************************\n",
      "Epoch: 115 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27682.87 .. KL_eta: 99.75 .. KL_alpha: 382632.03 .. Rec_loss: 16622871.0 .. NELBO: 17033285.27\n",
      "****************************************************************************************************\n",
      "Epoch----->115 .. LR: 0.005 .. KL_theta: 27459.04 .. KL_eta: 99.01 .. KL_alpha: 380704.27 .. Rec_loss: 16653338.24 .. NELBO: 17061600.35\n",
      "****************************************************************************************************\n",
      "Epoch: 116 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27559.18 .. KL_eta: 92.08 .. KL_alpha: 374727.07 .. Rec_loss: 16813014.09 .. NELBO: 17215392.36\n",
      "****************************************************************************************************\n",
      "Epoch----->116 .. LR: 0.005 .. KL_theta: 27690.43 .. KL_eta: 92.4 .. KL_alpha: 374448.08 .. Rec_loss: 16591378.82 .. NELBO: 16993609.71\n",
      "****************************************************************************************************\n",
      "Epoch: 117 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27387.12 .. KL_eta: 89.48 .. KL_alpha: 370391.74 .. Rec_loss: 16277118.64 .. NELBO: 16674987.0\n",
      "****************************************************************************************************\n",
      "Epoch----->117 .. LR: 0.005 .. KL_theta: 27794.6 .. KL_eta: 93.85 .. KL_alpha: 370010.62 .. Rec_loss: 16693845.59 .. NELBO: 17091744.76\n",
      "****************************************************************************************************\n",
      "Epoch: 118 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27387.58 .. KL_eta: 100.46 .. KL_alpha: 365145.83 .. Rec_loss: 16762465.0 .. NELBO: 17155098.64\n",
      "****************************************************************************************************\n",
      "Epoch----->118 .. LR: 0.005 .. KL_theta: 27458.34 .. KL_eta: 100.19 .. KL_alpha: 363623.89 .. Rec_loss: 16662185.76 .. NELBO: 17053367.94\n",
      "****************************************************************************************************\n",
      "Epoch: 119 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27388.37 .. KL_eta: 94.93 .. KL_alpha: 359638.96 .. Rec_loss: 16622182.09 .. NELBO: 17009304.45\n",
      "****************************************************************************************************\n",
      "Epoch----->119 .. LR: 0.005 .. KL_theta: 27563.83 .. KL_eta: 95.28 .. KL_alpha: 359199.96 .. Rec_loss: 16713680.29 .. NELBO: 17100539.41\n",
      "****************************************************************************************************\n",
      "Epoch: 120 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27849.46 .. KL_eta: 93.79 .. KL_alpha: 353479.68 .. Rec_loss: 16309873.64 .. NELBO: 16691296.64\n",
      "****************************************************************************************************\n",
      "Epoch----->120 .. LR: 0.005 .. KL_theta: 27693.07 .. KL_eta: 94.15 .. KL_alpha: 353017.69 .. Rec_loss: 16674316.18 .. NELBO: 17055121.18\n",
      "****************************************************************************************************\n",
      "Epoch: 121 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27919.44 .. KL_eta: 98.63 .. KL_alpha: 349285.43 .. Rec_loss: 16518550.64 .. NELBO: 16895854.36\n",
      "****************************************************************************************************\n",
      "Epoch----->121 .. LR: 0.005 .. KL_theta: 27833.63 .. KL_eta: 99.61 .. KL_alpha: 348089.76 .. Rec_loss: 16662292.53 .. NELBO: 17038315.76\n",
      "****************************************************************************************************\n",
      "Epoch: 122 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28345.55 .. KL_eta: 103.01 .. KL_alpha: 346096.51 .. Rec_loss: 16650006.82 .. NELBO: 17024551.73\n",
      "****************************************************************************************************\n",
      "Epoch----->122 .. LR: 0.005 .. KL_theta: 27860.48 .. KL_eta: 103.34 .. KL_alpha: 345506.45 .. Rec_loss: 16684699.76 .. NELBO: 17058169.82\n",
      "****************************************************************************************************\n",
      "Epoch: 123 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27739.94 .. KL_eta: 108.07 .. KL_alpha: 341341.09 .. Rec_loss: 16880884.0 .. NELBO: 17250073.09\n",
      "****************************************************************************************************\n",
      "Epoch----->123 .. LR: 0.005 .. KL_theta: 27553.93 .. KL_eta: 109.62 .. KL_alpha: 340436.26 .. Rec_loss: 16609006.65 .. NELBO: 16977106.35\n",
      "****************************************************************************************************\n",
      "Epoch: 124 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27765.24 .. KL_eta: 98.43 .. KL_alpha: 336492.83 .. Rec_loss: 16512254.18 .. NELBO: 16876611.0\n",
      "****************************************************************************************************\n",
      "Epoch----->124 .. LR: 0.005 .. KL_theta: 27886.31 .. KL_eta: 96.82 .. KL_alpha: 335194.43 .. Rec_loss: 16627294.35 .. NELBO: 16990472.18\n",
      "****************************************************************************************************\n",
      "Epoch: 125 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27631.59 .. KL_eta: 95.37 .. KL_alpha: 329215.83 .. Rec_loss: 16790660.0 .. NELBO: 17147603.0\n",
      "****************************************************************************************************\n",
      "Epoch----->125 .. LR: 0.005 .. KL_theta: 27610.89 .. KL_eta: 95.93 .. KL_alpha: 329163.66 .. Rec_loss: 16587365.06 .. NELBO: 16944235.76\n",
      "****************************************************************************************************\n",
      "Epoch: 126 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28307.12 .. KL_eta: 98.93 .. KL_alpha: 326697.63 .. Rec_loss: 16435785.18 .. NELBO: 16790889.0\n",
      "****************************************************************************************************\n",
      "Epoch----->126 .. LR: 0.005 .. KL_theta: 27966.56 .. KL_eta: 98.33 .. KL_alpha: 325201.62 .. Rec_loss: 16569653.29 .. NELBO: 16922919.88\n",
      "****************************************************************************************************\n",
      "Epoch: 127 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27809.93 .. KL_eta: 95.23 .. KL_alpha: 323367.0 .. Rec_loss: 16625698.18 .. NELBO: 16976970.0\n",
      "****************************************************************************************************\n",
      "Epoch----->127 .. LR: 0.005 .. KL_theta: 28117.13 .. KL_eta: 96.05 .. KL_alpha: 322196.81 .. Rec_loss: 16606912.94 .. NELBO: 16957322.76\n",
      "****************************************************************************************************\n",
      "Epoch: 128 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27524.5 .. KL_eta: 95.87 .. KL_alpha: 318249.79 .. Rec_loss: 16623632.55 .. NELBO: 16969502.64\n",
      "****************************************************************************************************\n",
      "Epoch----->128 .. LR: 0.005 .. KL_theta: 27553.97 .. KL_eta: 95.27 .. KL_alpha: 317557.72 .. Rec_loss: 16597993.94 .. NELBO: 16943200.88\n",
      "****************************************************************************************************\n",
      "Epoch: 129 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27929.95 .. KL_eta: 92.92 .. KL_alpha: 314257.47 .. Rec_loss: 16838026.91 .. NELBO: 17180307.0\n",
      "****************************************************************************************************\n",
      "Epoch----->129 .. LR: 0.005 .. KL_theta: 27941.76 .. KL_eta: 94.31 .. KL_alpha: 314243.21 .. Rec_loss: 16597893.71 .. NELBO: 16940172.94\n",
      "****************************************************************************************************\n",
      "Epoch: 130 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28094.82 .. KL_eta: 103.06 .. KL_alpha: 311090.46 .. Rec_loss: 16445143.82 .. NELBO: 16784432.27\n",
      "****************************************************************************************************\n",
      "Epoch----->130 .. LR: 0.005 .. KL_theta: 27971.94 .. KL_eta: 100.33 .. KL_alpha: 310072.94 .. Rec_loss: 16638183.06 .. NELBO: 16976328.53\n",
      "****************************************************************************************************\n",
      "Epoch: 131 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28344.84 .. KL_eta: 97.52 .. KL_alpha: 305765.79 .. Rec_loss: 16580252.64 .. NELBO: 16914460.64\n",
      "****************************************************************************************************\n",
      "Epoch----->131 .. LR: 0.005 .. KL_theta: 28084.98 .. KL_eta: 95.83 .. KL_alpha: 306075.44 .. Rec_loss: 16574699.0 .. NELBO: 16908955.12\n",
      "****************************************************************************************************\n",
      "Epoch: 132 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28400.51 .. KL_eta: 93.17 .. KL_alpha: 300271.21 .. Rec_loss: 16814929.91 .. NELBO: 17143694.36\n",
      "****************************************************************************************************\n",
      "Epoch----->132 .. LR: 0.005 .. KL_theta: 28134.18 .. KL_eta: 96.22 .. KL_alpha: 300837.9 .. Rec_loss: 16562437.71 .. NELBO: 16891505.59\n",
      "****************************************************************************************************\n",
      "Epoch: 133 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27980.15 .. KL_eta: 106.19 .. KL_alpha: 298070.21 .. Rec_loss: 16401112.64 .. NELBO: 16727269.18\n",
      "****************************************************************************************************\n",
      "Epoch----->133 .. LR: 0.005 .. KL_theta: 27892.67 .. KL_eta: 103.26 .. KL_alpha: 297766.6 .. Rec_loss: 16599080.94 .. NELBO: 16924843.59\n",
      "****************************************************************************************************\n",
      "Epoch: 134 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28254.51 .. KL_eta: 103.16 .. KL_alpha: 294823.09 .. Rec_loss: 16538511.0 .. NELBO: 16861691.64\n",
      "****************************************************************************************************\n",
      "Epoch----->134 .. LR: 0.005 .. KL_theta: 28030.3 .. KL_eta: 101.44 .. KL_alpha: 293761.48 .. Rec_loss: 16582075.47 .. NELBO: 16903968.65\n",
      "****************************************************************************************************\n",
      "Epoch: 135 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28067.47 .. KL_eta: 95.24 .. KL_alpha: 291611.54 .. Rec_loss: 16259103.73 .. NELBO: 16578878.18\n",
      "****************************************************************************************************\n",
      "Epoch----->135 .. LR: 0.005 .. KL_theta: 28160.56 .. KL_eta: 96.77 .. KL_alpha: 291561.91 .. Rec_loss: 16605709.65 .. NELBO: 16925528.94\n",
      "****************************************************************************************************\n",
      "Epoch: 136 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28142.1 .. KL_eta: 99.9 .. KL_alpha: 287997.73 .. Rec_loss: 16552710.0 .. NELBO: 16868949.82\n",
      "****************************************************************************************************\n",
      "Epoch----->136 .. LR: 0.005 .. KL_theta: 28037.04 .. KL_eta: 97.51 .. KL_alpha: 287673.23 .. Rec_loss: 16590127.12 .. NELBO: 16905935.06\n",
      "****************************************************************************************************\n",
      "Epoch: 137 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28843.58 .. KL_eta: 93.45 .. KL_alpha: 283534.08 .. Rec_loss: 16704230.0 .. NELBO: 17016700.82\n",
      "****************************************************************************************************\n",
      "Epoch----->137 .. LR: 0.005 .. KL_theta: 28353.01 .. KL_eta: 94.34 .. KL_alpha: 283015.16 .. Rec_loss: 16560787.76 .. NELBO: 16872250.0\n",
      "****************************************************************************************************\n",
      "Epoch: 138 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28522.49 .. KL_eta: 95.5 .. KL_alpha: 280951.89 .. Rec_loss: 16517156.64 .. NELBO: 16826726.36\n",
      "****************************************************************************************************\n",
      "Epoch----->138 .. LR: 0.005 .. KL_theta: 28542.05 .. KL_eta: 94.62 .. KL_alpha: 280275.08 .. Rec_loss: 16641298.06 .. NELBO: 16950209.76\n",
      "****************************************************************************************************\n",
      "Epoch: 139 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 27952.76 .. KL_eta: 98.04 .. KL_alpha: 278275.34 .. Rec_loss: 16514667.27 .. NELBO: 16820993.55\n",
      "****************************************************************************************************\n",
      "Epoch----->139 .. LR: 0.005 .. KL_theta: 28128.52 .. KL_eta: 100.21 .. KL_alpha: 276789.35 .. Rec_loss: 16512808.29 .. NELBO: 16817826.29\n",
      "****************************************************************************************************\n",
      "Epoch: 140 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28034.62 .. KL_eta: 107.61 .. KL_alpha: 273837.02 .. Rec_loss: 16405676.64 .. NELBO: 16707655.82\n",
      "****************************************************************************************************\n",
      "Epoch----->140 .. LR: 0.005 .. KL_theta: 28135.33 .. KL_eta: 107.3 .. KL_alpha: 272933.16 .. Rec_loss: 16621409.65 .. NELBO: 16922585.41\n",
      "****************************************************************************************************\n",
      "Epoch: 141 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28057.38 .. KL_eta: 102.9 .. KL_alpha: 269775.07 .. Rec_loss: 16667539.82 .. NELBO: 16965475.64\n",
      "****************************************************************************************************\n",
      "Epoch----->141 .. LR: 0.005 .. KL_theta: 28208.03 .. KL_eta: 101.54 .. KL_alpha: 269497.35 .. Rec_loss: 16613400.65 .. NELBO: 16911207.82\n",
      "****************************************************************************************************\n",
      "Epoch: 142 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28425.78 .. KL_eta: 104.52 .. KL_alpha: 267107.02 .. Rec_loss: 16534635.64 .. NELBO: 16830273.09\n",
      "****************************************************************************************************\n",
      "Epoch----->142 .. LR: 0.005 .. KL_theta: 28297.23 .. KL_eta: 104.34 .. KL_alpha: 266976.85 .. Rec_loss: 16637585.06 .. NELBO: 16932963.47\n",
      "****************************************************************************************************\n",
      "Epoch: 143 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28222.96 .. KL_eta: 96.56 .. KL_alpha: 263729.13 .. Rec_loss: 16802629.45 .. NELBO: 17094677.82\n",
      "****************************************************************************************************\n",
      "Epoch----->143 .. LR: 0.005 .. KL_theta: 28055.04 .. KL_eta: 95.48 .. KL_alpha: 263482.81 .. Rec_loss: 16604259.71 .. NELBO: 16895892.88\n",
      "****************************************************************************************************\n",
      "Epoch: 144 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28103.75 .. KL_eta: 90.96 .. KL_alpha: 261182.21 .. Rec_loss: 16608526.45 .. NELBO: 16897903.45\n",
      "****************************************************************************************************\n",
      "Epoch----->144 .. LR: 0.005 .. KL_theta: 28399.94 .. KL_eta: 90.77 .. KL_alpha: 260940.25 .. Rec_loss: 16564403.12 .. NELBO: 16853833.94\n",
      "****************************************************************************************************\n",
      "Epoch: 145 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28000.73 .. KL_eta: 94.11 .. KL_alpha: 257720.2 .. Rec_loss: 16261702.09 .. NELBO: 16547517.09\n",
      "****************************************************************************************************\n",
      "Epoch----->145 .. LR: 0.005 .. KL_theta: 28435.07 .. KL_eta: 95.46 .. KL_alpha: 257197.56 .. Rec_loss: 16627696.06 .. NELBO: 16913424.24\n",
      "****************************************************************************************************\n",
      "Epoch: 146 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28164.49 .. KL_eta: 97.13 .. KL_alpha: 255387.99 .. Rec_loss: 16364737.0 .. NELBO: 16648386.55\n",
      "****************************************************************************************************\n",
      "Epoch----->146 .. LR: 0.005 .. KL_theta: 28117.2 .. KL_eta: 96.22 .. KL_alpha: 254513.57 .. Rec_loss: 16525241.06 .. NELBO: 16807968.12\n",
      "****************************************************************************************************\n",
      "Epoch: 147 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28277.33 .. KL_eta: 101.94 .. KL_alpha: 251947.88 .. Rec_loss: 16536093.55 .. NELBO: 16816420.82\n",
      "****************************************************************************************************\n",
      "Epoch----->147 .. LR: 0.005 .. KL_theta: 28429.69 .. KL_eta: 102.9 .. KL_alpha: 251264.64 .. Rec_loss: 16660242.0 .. NELBO: 16940039.12\n",
      "****************************************************************************************************\n",
      "Epoch: 148 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28533.83 .. KL_eta: 96.62 .. KL_alpha: 249280.02 .. Rec_loss: 16545254.18 .. NELBO: 16823164.36\n",
      "****************************************************************************************************\n",
      "Epoch----->148 .. LR: 0.005 .. KL_theta: 28458.07 .. KL_eta: 95.17 .. KL_alpha: 248741.01 .. Rec_loss: 16561960.41 .. NELBO: 16839254.76\n",
      "****************************************************************************************************\n",
      "Epoch: 149 .. batch: 10/17 .. LR: 0.005 .. KL_theta: 28990.48 .. KL_eta: 96.05 .. KL_alpha: 246115.61 .. Rec_loss: 16660307.36 .. NELBO: 16935509.64\n",
      "****************************************************************************************************\n",
      "Epoch----->149 .. LR: 0.005 .. KL_theta: 28651.67 .. KL_eta: 99.39 .. KL_alpha: 245714.65 .. Rec_loss: 16552897.35 .. NELBO: 16827363.18\n",
      "****************************************************************************************************\n",
      "saving topic matrix beta...\n",
      "CPU times: user 18h 24min 38s, sys: 42min 26s, total: 19h 7min 5s\n",
      "Wall time: 23min 55s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "## train model on data by looping through multiple epochs\n",
    "best_epoch = 0\n",
    "best_val_ppl = 1e9\n",
    "all_val_ppls = []\n",
    "for epoch in range(1, args.epochs):\n",
    "    train(epoch)\n",
    "    ## check whether to anneal lr\n",
    "    lr = optimizer.param_groups[0]['lr']\n",
    "    if args.anneal_lr and (len(all_val_ppls) > args.nonmono and val_ppl > min(all_val_ppls[:-args.nonmono]) and lr > 1e-5):\n",
    "        optimizer.param_groups[0]['lr'] /= args.lr_factor\n",
    "    #all_val_ppls.append(val_ppl)\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print('saving topic matrix beta...')\n",
    "    alpha = model.mu_q_alpha\n",
    "    beta = model.get_beta(alpha).cpu().numpy()\n",
    "    scipy.io.savemat(ckpt+'_beta.mat', {'values': beta}, do_compression=True)\n",
    "    if args.train_embeddings:\n",
    "        print('saving word embedding matrix rho...')\n",
    "        rho = model.rho.weight.cpu().numpy()\n",
    "        scipy.io.savemat(ckpt+'_rho.mat', {'values': rho}, do_compression=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce60ce5b-33c9-46d5-8052-4011e8ff7559",
   "metadata": {},
   "source": [
    "## see if i can get a measure of topic coherence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9e47996a-e7d0-4d97-945c-6929968fd00f",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_parquet('data/combined_clean.parquet')\n",
    "id2word = Dictionary([vocab])\n",
    "split_text = df['filtered_text'].str.split().values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f1e2940c-e1ee-4610-87cf-c02b84767760",
   "metadata": {},
   "outputs": [],
   "source": [
    "args.num_words = 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "ff7e94ff-629a-49ca-b780-df360a861b91",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "beta:  torch.Size([10, 4, 4938])\n",
      "\n",
      "\n",
      "####################################################################################################\n",
      "Visualize topics...\n",
      "Topic 0 .. Time: 0 ===> ['financial', 'loan', 'system', 'credit', 'bank', 'asset', 'payment', 'security', 'reserve']\n",
      "Topic 0 .. Time: 2 ===> ['financial', 'asset', 'bank', 'fund', 'system', 'payment', 'security', 'credit', 'reserve']\n",
      "Topic 1 .. Time: 0 ===> ['underlie', 'remark', 'examiner', 'subprime', 'foreclosure', 'intermediary', 'clearing', 'induce', 'insure']\n",
      "Topic 1 .. Time: 2 ===> ['remark', 'subprime', 'foreclosure', 'examiner', 'underlie', 'clearing', 'intermediary', 'insure', 'nonprime']\n",
      "Topic 2 .. Time: 0 ===> ['productivity', 'business', 'growth', 'technology', 'investment', 'economy', 'labor', 'development', 'saving']\n",
      "Topic 2 .. Time: 2 ===> ['labor', 'growth', 'business', 'work', 'job', 'productivity', 'research', 'development', 'worker']\n",
      "Topic 3 .. Time: 0 ===> ['price', 'effect', 'demand', 'expectation', 'level', 'cost', 'interest', 'rate', 'expect']\n",
      "Topic 3 .. Time: 2 ===> ['price', 'interest', 'level', 'expectation', 'effect', 'expect', 'condition', 'vol', 'suggest']\n",
      "Topic 4 .. Time: 0 ===> ['percent', 'increase', 'rate', 'rise', 'growth', 'high', 'decline', 'home', 'year']\n",
      "Topic 4 .. Time: 2 ===> ['text', 'percent', 'federal', 'increase', 'decline', 'datum', 'pdf', 'household', 'crisis']\n",
      "Topic 5 .. Time: 0 ===> ['institution', 'provide', 'need', 'information', 'require', 'issue', 'risk', 'regulatory', 'include']\n",
      "Topic 5 .. Time: 2 ===> ['institution', 'regulatory', 'regulation', 'provide', 'need', 'requirement', 'risk', 'include', 'rule']\n",
      "Topic 6 .. Time: 0 ===> ['committee', 'board', 'action', 'federal', 'fomc', 'believe', 'objective', 'governor', 'decision']\n",
      "Topic 6 .. Time: 2 ===> ['committee', 'fomc', 'board', 'governor', 'objective', 'discussion', 'meeting', 'decision', 'action']\n",
      "Topic 7 .. Time: 0 ===> ['year', 'long', 'time', 'new', 'country', 'term', 'change', 'important', 'high']\n",
      "Topic 7 .. Time: 2 ===> ['return', 'long', 'time', 'year', 'low', 'term', 'important', 'run', 'new']\n",
      "Topic 8 .. Time: 0 ===> ['policy', 'inflation', 'rate', 'monetary', 'central', 'economy', 'growth', 'target', 'stability']\n",
      "Topic 8 .. Time: 2 ===> ['policy', 'rate', 'inflation', 'monetary', 'economy', 'central', 'stability', 'growth', 'unemployment']\n",
      "Topic 9 .. Time: 0 ===> ['bank', 'risk', 'banking', 'capital', 'financial', 'management', 'business', 'firm', 'community']\n",
      "Topic 9 .. Time: 2 ===> ['bank', 'capital', 'risk', 'firm', 'banking', 'community', 'system', 'stress', 'large']\n"
     ]
    }
   ],
   "source": [
    "with torch.no_grad():\n",
    "    alpha = model.mu_q_alpha\n",
    "    beta = model.get_beta(alpha) \n",
    "    print('beta: ', beta.size())\n",
    "    print('\\n')\n",
    "    print('#'*100)\n",
    "    print('Visualize topics...')\n",
    "    times = [0, 2]\n",
    "    topics_words = []\n",
    "    for k in range(args.num_topics):\n",
    "        for t in times:\n",
    "            gamma = beta[k, t, :]\n",
    "            top_words = list(gamma.cpu().numpy().argsort()[-args.num_words+1:][::-1])\n",
    "            topic_words = [id2word[a] for a in top_words]\n",
    "            topics_words.append(' '.join(topic_words))\n",
    "            print('Topic {} .. Time: {} ===> {}'.format(k, t, topic_words)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "55e5a6f0-9fec-4550-9b17-0e536c9595bd",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 4/4 [00:52<00:00, 13.07s/it]\n"
     ]
    }
   ],
   "source": [
    "coherences = []\n",
    "for t in tqdm(range(args.num_times)):\n",
    "    coherences.append(\n",
    "        CoherenceModel(\n",
    "            topics=get_detm_topics(beta=beta, time=t, num_words=20, vocab=id2word, num_topics=args.num_topics), # use 20 words to standardize with DTM\n",
    "            texts=split_text, \n",
    "            dictionary=id2word, \n",
    "            coherence='c_v'\n",
    "        ).get_coherence()\n",
    "    )\n",
    "\n",
    "coherences = np.array(coherences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "31337eeb-a48c-48af-9339-67e2ca107c90",
   "metadata": {},
   "outputs": [],
   "source": [
    "def topic_diversity(topics):\n",
    "    all_words = set(np.concatenate(topics))\n",
    "    return len(all_words) / (len(topics[0]) * len(topics))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "46a4b7b1-7a78-4d78-b42c-6845c032bb08",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 4/4 [00:00<00:00, 207.43it/s]\n"
     ]
    }
   ],
   "source": [
    "diversities = []\n",
    "for t in tqdm(range(args.num_times)):\n",
    "    diversities.append(\n",
    "        topic_diversity(topics=get_detm_topics(beta=beta, time=t, num_words=20, vocab=id2word, num_topics=args.num_topics))\n",
    "    )\n",
    "    \n",
    "diversities = np.array(diversities)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "a1a4184d-93eb-4844-87c7-174d5ca8f485",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.3795111696155748, 0.003672591640759064)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qualities = diversities * coherences\n",
    "qualities.mean(), qualities.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "6294257d-5fc9-45d2-aab7-c57b24a99af8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0.43669734, 0.43612565, 0.43312171, 0.42893962]),\n",
       " array([0.875, 0.875, 0.88 , 0.87 ]))"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "coherences, diversities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "01cf7948-c9df-454d-82e1-8c1b95b787ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.savez_compressed(\n",
    "    'detm_stats_bert.npz',\n",
    "    coherence=coherences,\n",
    "    diversity=diversities\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb012b8b-271c-456c-bbd7-7132a5861bdf",
   "metadata": {},
   "source": [
    "## try to dig into topic evolutions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "2cbb7f4a-17a7-4e93-b15a-c708218907ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "beta = scipy.io.loadmat('results/bert_detm_un_K_10_Htheta_800_Optim_adam_Clip_0.0_ThetaAct_relu_Lr_0.005_Bsz_100_RhoSize_768_L_3_minDF_100_trainEmbeddings_0_beta.mat')['values']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "438c420a-b45e-40c9-98df-1a65bdfa06a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_word_probs(word, topic, beta):\n",
    "    word_id = id2word.token2id[word]\n",
    "    probs = []\n",
    "    for t in range(4):\n",
    "        gamma = beta[topic, t, :]\n",
    "        probs.append(gamma[word_id])\n",
    "    return probs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9fa4e13d-0374-4dce-93e0-d31c01a9e085",
   "metadata": {},
   "outputs": [],
   "source": [
    "## make some plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1b09bec2-a8fd-441a-b1e9-b39056e4f6b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "55906ad9-0a14-4544-bd32-ce73ba002b0e",
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_df = pd.DataFrame({\n",
    "    'credit': get_word_probs('credit', 0, beta),\n",
    "    'security': get_word_probs('security', 0, beta),\n",
    "    'foreclosure': get_word_probs('foreclosure', 1, beta),\n",
    "    'subprime': get_word_probs('subprime', 1, beta),\n",
    "    'labor': get_word_probs('labor', 2, beta),\n",
    "    'technology': get_word_probs('technology', 2, beta),\n",
    "    'vol': get_word_probs('vol', 3, beta),\n",
    "    'demand': get_word_probs('demand', 3, beta),\n",
    "    'regulation': get_word_probs('regulation', 5, beta),\n",
    "    'requirement': get_word_probs('requirement', 5, beta),\n",
    "    'inflation': get_word_probs('inflation', 8, beta),\n",
    "    'unemployment': get_word_probs('unemployment', 8, beta),\n",
    "    'stress': get_word_probs('stress', 9, beta),\n",
    "})\n",
    "\n",
    "plot_df.to_csv('bert_evolve.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "4b23408b-e2ec-4286-b443-8605f3da815c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df[['credit', 'security']].plot(marker='X', markersize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "acad156b-6d72-4dd5-8a7c-2af4cc2b01b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df[['foreclosure', 'subprime']].plot(marker='X', markersize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "a666037e-01bc-43aa-9296-35b35956f748",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df[['labor', 'technology']].plot(marker='X', markersize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "d61cd528-a2b6-451c-8e94-c31ad576221f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df[['vol', 'demand']].plot(marker='X', markersize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "2944ca47-d428-4748-b2ce-53c0ebda9bae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df[['regulation', 'requirement']].plot(marker='X', markersize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "f669c813-9d07-49c0-922f-1ece3ba24b87",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df[['inflation', 'unemployment']].plot(marker='X', markersize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "52b2abfc-d9df-4da2-930a-264dc7dd4c13",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "nlp",
   "language": "python",
   "name": "nlp"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}