Example_main.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Example of running with hyperparameters tuning\n",
    "### Kirill Zakharov\n",
    "2024"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import copy\n",
    "%matplotlib inline\n",
    "\n",
    "import torch\n",
    "from torch import optim\n",
    "from torch.autograd import Variable, grad\n",
    "from torch.autograd.functional import jacobian\n",
    "from torch.utils.data import DataLoader, Dataset\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torch.distributions.multivariate_normal import MultivariateNormal\n",
    "\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "from sklearn.preprocessing import MinMaxScaler, StandardScaler\n",
    "from scipy.special import rel_entr, kl_div, factorial\n",
    "from scipy.spatial.distance import jensenshannon\n",
    "from scipy.stats import kstest, ks_2samp, wasserstein_distance\n",
    "from scipy.optimize import minimize\n",
    "import scipy.stats as sts\n",
    "from tqdm import tqdm\n",
    "import itertools\n",
    "\n",
    "\n",
    "from JDFLOW.intensity_optimization import *\n",
    "from JDFLOW.signature_computation import *\n",
    "from JDFLOW.stochastic_processes import *\n",
    "from JDFLOW.nsde_functions import *\n",
    "from JDFLOW.nsde_solvers import *\n",
    "from JDFLOW.jdflow import *\n",
    "from JDFLOW.evaluate.metrics import *\n",
    "from JDFLOW.evaluate.monte_carlo_sim import monte_carlo_train, monte_carlo_sample\n",
    "from JDFLOW.FourierFlows import FourierFlow\n",
    "from JDFLOW.FSDE import train_fsde\n",
    "from JDFLOW.evaluate.metrics import *\n",
    "from Model import *\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Choose any type process or upload your data. The main rule that length of each series must be the same. Available: 'DCL', 'GBM', 'Merton'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data, dt, n = choose_data('Merton', s0=2, xiP=7, muj=0, sigmaj=0.2, r=0.04, sigmad=1, T=1)\n",
    "# data, dt, n = choose_data('DCL', theta=1, delta=2, T=1)\n",
    "# data, dt, n = choose_data('GBM', s0=1, mu=0.2, sigma=0.9, T=1)\n",
    "# data, dt, n = choose_data('Stock', filename='Data/gen.us.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(data.T)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Loss: -1360.6332: 100%|██████████| 1000/1000 [01:55<00:00,  8.65it/s]\n"
     ]
    }
   ],
   "source": [
    "jdflow, M, scaler, losses = train(data, dt, n, n_flows=10, epochs=1000, time_steps=100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Generate synthetic time series."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "synth_data = sample(jdflow, n_samples=5, M=M, scaler=scaler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.plot(data.T, '--', color='black', alpha=0.8, label='Real')\n",
    "plt.plot(synth_data.T)\n",
    "\n",
    "# plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_synth_data = monte_carlo_sample(100, len(data), jdflow, M, scaler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(avg_synth_data.T)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# avg_tr_synth_data = monte_carlo_train(data, 2, dt, n, n_flows=8, epochs=1000, time_steps=125)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plt.plot(avg_tr_synth_data.T)\n",
    "\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hyperparameters tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ray import tune, train\n",
    "import ray\n",
    "from ray.tune.schedulers import ASHAScheduler\n",
    "from ray.tune.search.optuna import OptunaSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def objective_fun(z, data, dt, n):\n",
    "    n_flows, epochs, time_steps = z['n_flows'], z['epochs'], z['time_steps']\n",
    "    jdflow, M, scaler, losses = train(data, dt, n, n_flows, epochs, time_steps)\n",
    "    \n",
    "    synth_data = sample(jdflow, n_samples=len(data), M=M, scaler=scaler)\n",
    "    \n",
    "    ray.train.report({'loss': w_dist_calc(synth_data, data)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = {\n",
    "        \"n_flows\": tune.choice(np.arange(4, 12, 2)),\n",
    "        \"epochs\": tune.choice(np.arange(700, 1200, 100)),\n",
    "        \"time_steps\": tune.choice(np.arange(5, 130, 20)),\n",
    "    }\n",
    "\n",
    "optuna_search = OptunaSearch(\n",
    "    metric='loss',\n",
    "    mode=\"min\")\n",
    "\n",
    "scheduler = ASHAScheduler(\n",
    "        metric=\"loss\",\n",
    "        mode=\"min\",\n",
    "        max_t=10,\n",
    "        grace_period=1,\n",
    "        reduction_factor=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "tuner = tune.Tuner(partial(objective_fun, data=data, dt=dt, n=n),\\\n",
    "                param_space=config,\\\n",
    "                tune_config=tune.TuneConfig(num_samples=20, search_alg=optuna_search, scheduler=scheduler),)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2024-01-10 22:00:51</td></tr>\n",
       "<tr><td>Running for: </td><td>00:19:22.24        </td></tr>\n",
       "<tr><td>Memory:      </td><td>5.6/8.0 GiB        </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      Using AsyncHyperBand: num_stopped=9<br>Bracket: Iter 8.000: None | Iter 4.000: None | Iter 2.000: None | Iter 1.000: -0.006903638585523021<br>Logical resource usage: 1.0/8 CPUs, 0/0 GPUs\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name            </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">  epochs</th><th style=\"text-align: right;\">  n_flows</th><th style=\"text-align: right;\">  time_steps</th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">      loss</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>objective_fun_39bc151d</td><td>TERMINATED</td><td>127.0.0.1:58661</td><td style=\"text-align: right;\">    1100</td><td style=\"text-align: right;\">        8</td><td style=\"text-align: right;\">         125</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         628.701</td><td style=\"text-align: right;\">0.00322762</td></tr>\n",
       "<tr><td>objective_fun_485d52a5</td><td>TERMINATED</td><td>127.0.0.1:58668</td><td style=\"text-align: right;\">     900</td><td style=\"text-align: right;\">        6</td><td style=\"text-align: right;\">         125</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         450.05 </td><td style=\"text-align: right;\">0.0112339 </td></tr>\n",
       "<tr><td>objective_fun_ff1ac5d2</td><td>TERMINATED</td><td>127.0.0.1:58670</td><td style=\"text-align: right;\">     800</td><td style=\"text-align: right;\">        8</td><td style=\"text-align: right;\">           5</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         251.922</td><td style=\"text-align: right;\">0.0108874 </td></tr>\n",
       "<tr><td>objective_fun_586db584</td><td>TERMINATED</td><td>127.0.0.1:58673</td><td style=\"text-align: right;\">     800</td><td style=\"text-align: right;\">        8</td><td style=\"text-align: right;\">          45</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         367.846</td><td style=\"text-align: right;\">0.00643524</td></tr>\n",
       "<tr><td>objective_fun_1e8dd32d</td><td>TERMINATED</td><td>127.0.0.1:58678</td><td style=\"text-align: right;\">     700</td><td style=\"text-align: right;\">       10</td><td style=\"text-align: right;\">           5</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         303.613</td><td style=\"text-align: right;\">0.00918304</td></tr>\n",
       "<tr><td>objective_fun_1e38b96c</td><td>TERMINATED</td><td>127.0.0.1:58681</td><td style=\"text-align: right;\">     800</td><td style=\"text-align: right;\">        8</td><td style=\"text-align: right;\">         105</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         499.215</td><td style=\"text-align: right;\">0.00720973</td></tr>\n",
       "<tr><td>objective_fun_9b744f06</td><td>TERMINATED</td><td>127.0.0.1:58684</td><td style=\"text-align: right;\">    1000</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">          85</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         409.633</td><td style=\"text-align: right;\">0.0233961 </td></tr>\n",
       "<tr><td>objective_fun_c6c843b2</td><td>TERMINATED</td><td>127.0.0.1:58686</td><td style=\"text-align: right;\">     800</td><td style=\"text-align: right;\">        6</td><td style=\"text-align: right;\">           5</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         222.525</td><td style=\"text-align: right;\">0.00480901</td></tr>\n",
       "<tr><td>objective_fun_6f4ae881</td><td>TERMINATED</td><td>127.0.0.1:58686</td><td style=\"text-align: right;\">    1100</td><td style=\"text-align: right;\">       10</td><td style=\"text-align: right;\">         105</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         726.974</td><td style=\"text-align: right;\">0.00864103</td></tr>\n",
       "<tr><td>objective_fun_e4136afb</td><td>TERMINATED</td><td>127.0.0.1:58670</td><td style=\"text-align: right;\">    1000</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">         125</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         425.597</td><td style=\"text-align: right;\">0.0104963 </td></tr>\n",
       "<tr><td>objective_fun_3bf56851</td><td>TERMINATED</td><td>127.0.0.1:58678</td><td style=\"text-align: right;\">    1000</td><td style=\"text-align: right;\">       10</td><td style=\"text-align: right;\">          85</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         634.252</td><td style=\"text-align: right;\">0.00605884</td></tr>\n",
       "<tr><td>objective_fun_17b9489e</td><td>TERMINATED</td><td>127.0.0.1:58673</td><td style=\"text-align: right;\">     800</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">          25</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         202.256</td><td style=\"text-align: right;\">0.00651542</td></tr>\n",
       "<tr><td>objective_fun_571128e0</td><td>TERMINATED</td><td>127.0.0.1:58684</td><td style=\"text-align: right;\">     700</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">         105</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         322.472</td><td style=\"text-align: right;\">0.00847192</td></tr>\n",
       "<tr><td>objective_fun_de0ae5f8</td><td>TERMINATED</td><td>127.0.0.1:58668</td><td style=\"text-align: right;\">    1100</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">         125</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         452.676</td><td style=\"text-align: right;\">0.00333464</td></tr>\n",
       "<tr><td>objective_fun_a6fbbfce</td><td>TERMINATED</td><td>127.0.0.1:58681</td><td style=\"text-align: right;\">    1100</td><td style=\"text-align: right;\">        8</td><td style=\"text-align: right;\">         125</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         582.814</td><td style=\"text-align: right;\">0.00541151</td></tr>\n",
       "<tr><td>objective_fun_4cb4be19</td><td>TERMINATED</td><td>127.0.0.1:58673</td><td style=\"text-align: right;\">     900</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">          25</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         223.965</td><td style=\"text-align: right;\">0.0146601 </td></tr>\n",
       "<tr><td>objective_fun_97d06347</td><td>TERMINATED</td><td>127.0.0.1:58661</td><td style=\"text-align: right;\">    1000</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">          65</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         290.236</td><td style=\"text-align: right;\">0.00659754</td></tr>\n",
       "<tr><td>objective_fun_0dde9108</td><td>TERMINATED</td><td>127.0.0.1:58670</td><td style=\"text-align: right;\">     700</td><td style=\"text-align: right;\">        4</td><td style=\"text-align: right;\">          45</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         186.425</td><td style=\"text-align: right;\">0.0109274 </td></tr>\n",
       "<tr><td>objective_fun_ec6db7f8</td><td>TERMINATED</td><td>127.0.0.1:58684</td><td style=\"text-align: right;\">    1100</td><td style=\"text-align: right;\">       10</td><td style=\"text-align: right;\">          65</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         411.24 </td><td style=\"text-align: right;\">0.00472326</td></tr>\n",
       "<tr><td>objective_fun_83d62da9</td><td>TERMINATED</td><td>127.0.0.1:58673</td><td style=\"text-align: right;\">    1100</td><td style=\"text-align: right;\">        6</td><td style=\"text-align: right;\">          65</td><td style=\"text-align: right;\">     1</td><td style=\"text-align: right;\">         307.722</td><td style=\"text-align: right;\">0.00421945</td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/1100 [00:00<?, ?it/s]\n",
      "Loss: 5271.2754:   0%|          | 1/1100 [00:00<05:40,  3.22it/s]\n",
      "Loss: 1324.9431:   0%|          | 2/1100 [00:00<04:55,  3.72it/s]\n",
      "Loss: 1073.4806:   0%|          | 3/1100 [00:00<04:43,  3.87it/s]\n",
      "Loss: 838.6102:   0%|          | 4/1100 [00:01<04:49,  3.78it/s] \n",
      "Loss: 814.0424:   0%|          | 5/1100 [00:01<05:35,  3.26it/s]\n",
      "Loss: 833.4789:   1%|          | 6/1100 [00:01<06:15,  2.91it/s]\n",
      "Loss: 793.8875:   1%|          | 7/1100 [00:02<05:44,  3.17it/s]\n",
      "Loss: 750.123:   1%|          | 8/1100 [00:02<05:46,  3.15it/s] \n",
      "Loss: 643.6787:   1%|          | 9/1100 [00:02<05:38,  3.23it/s]\n",
      "  0%|          | 0/800 [00:00<?, ?it/s]\u001b[32m [repeated 2x across cluster] (Ray deduplicates logs by default. Set RAY_DEDUP_LOGS=0 to disable log deduplication, or see https://docs.ray.io/en/master/ray-observability/ray-logging.html#log-deduplication for more options.)\u001b[0m\n",
      "Loss: 4880.7114:   0%|          | 1/800 [00:00<03:23,  3.92it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
      "Loss: 213.3025:   1%|          | 11/900 [00:03<04:42,  3.15it/s]\u001b[32m [repeated 20x across cluster]\u001b[0m\n",
      "  0%|          | 0/800 [00:00<?, ?it/s]\n",
      "Loss: 5091.5854:   0%|          | 1/800 [00:00<05:47,  2.30it/s]\n",
      "Loss: 24.0991:   3%|▎         | 25/800 [00:05<03:19,  3.88it/s]\u001b[32m [repeated 52x across cluster]\u001b[0m\n",
      "  0%|          | 0/700 [00:00<?, ?it/s]\n",
      "Loss: 6849.835:   0%|          | 1/700 [00:00<06:00,  1.94it/s]\n",
      "Loss: -112.1358:   4%|▍         | 36/900 [00:13<06:21,  2.27it/s]\u001b[32m [repeated 53x across cluster]\u001b[0m\n",
      "  0%|          | 0/800 [00:00<?, ?it/s]\n",
      "Loss: 4930.5503:   0%|          | 1/800 [00:00<10:55,  1.22it/s]\n",
      "Loss: -130.2879:   7%|▋         | 56/800 [00:16<05:18,  2.34it/s]\u001b[32m [repeated 53x across cluster]\u001b[0m\n",
      "  0%|          | 0/1000 [00:00<?, ?it/s]\n",
      "Loss: 1277.2213:   0%|          | 1/1000 [00:00<08:37,  1.93it/s]\n",
      "Loss: -49.4355:   4%|▍         | 34/800 [00:17<06:45,  1.89it/s]\u001b[32m [repeated 58x across cluster]\u001b[0m\n",
      "  0%|          | 0/800 [00:00<?, ?it/s]\n",
      "Loss: 3236.3711:   0%|          | 1/800 [00:00<08:09,  1.63it/s]\n",
      "Loss: -40.5193:   5%|▌         | 37/700 [00:19<06:24,  1.73it/s]\u001b[32m [repeated 64x across cluster]\u001b[0m\n",
      "Loss: -27.7529:   2%|▏         | 23/1000 [00:13<12:05,  1.35it/s]\u001b[32m [repeated 60x across cluster]\u001b[0m\n",
      "Loss: -209.993:  13%|█▎        | 104/800 [00:36<04:19,  2.68it/s]\u001b[32m [repeated 70x across cluster]\u001b[0m\n",
      "Loss: -119.8167:   9%|▊         | 60/700 [00:34<08:18,  1.28it/s]\u001b[32m [repeated 57x across cluster]\u001b[0m\n",
      "Loss: -243.0836:  10%|▉         | 87/900 [00:48<06:32,  2.07it/s]\u001b[32m [repeated 101x across cluster]\u001b[0m\n",
      "Loss: -279.576:  18%|█▊        | 146/800 [00:51<04:11,  2.61it/s]\u001b[32m [repeated 88x across cluster]\u001b[0m\n",
      "Loss: -304.308:  20%|██        | 160/800 [00:56<04:15,  2.50it/s]\u001b[32m [repeated 80x across cluster]\u001b[0m\n",
      "Loss: -324.8947:  22%|██▏       | 177/800 [01:01<02:43,  3.81it/s]\u001b[32m [repeated 104x across cluster]\u001b[0m\n",
      "Loss: -356.9496:  24%|██▍       | 196/800 [01:06<02:26,  4.13it/s]\u001b[32m [repeated 121x across cluster]\u001b[0m\n",
      "Loss: -255.2344:  17%|█▋        | 133/800 [01:08<04:28,  2.48it/s]\u001b[32m [repeated 109x across cluster]\u001b[0m\n",
      "Loss: -381.42:  17%|█▋        | 150/900 [01:19<05:13,  2.39it/s]\u001b[32m [repeated 106x across cluster]\u001b[0m\n",
      "Loss: -415.6037:  18%|█▊        | 162/900 [01:24<05:07,  2.40it/s]\u001b[32m [repeated 110x across cluster]\u001b[0m\n",
      "Loss: -406.7772:  19%|█▉        | 174/900 [01:29<04:46,  2.54it/s]\u001b[32m [repeated 109x across cluster]\u001b[0m\n",
      "Loss: -528.5609:  30%|███       | 243/800 [01:09<03:04,  3.02it/s]\u001b[32m [repeated 85x across cluster]\u001b[0m\n",
      "Loss: -506.4839:  37%|███▋      | 296/800 [01:36<02:25,  3.46it/s]\u001b[32m [repeated 100x across cluster]\u001b[0m\n",
      "Loss: -521.1964:  23%|██▎       | 205/900 [01:44<05:35,  2.07it/s]\u001b[32m [repeated 106x across cluster]\u001b[0m\n",
      "Loss: -512.754:  40%|███▉      | 319/800 [01:43<02:42,  2.96it/s] \n",
      "Loss: -512.754:  40%|████      | 320/800 [01:43<02:51,  2.80it/s]\n",
      "Loss: -329.8108:  17%|█▋        | 190/1100 [01:51<10:52,  1.40it/s]\u001b[32m [repeated 82x across cluster]\u001b[0m\n",
      "Loss: -511.4834:  40%|███▉      | 316/800 [01:29<02:29,  3.25it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "Loss: -353.339:  18%|█▊        | 202/1100 [02:00<08:42,  1.72it/s] \n",
      "Loss: -353.339:  18%|█▊        | 203/1100 [02:00<10:06,  1.48it/s]\n",
      "Loss: -356.9504:  19%|█▊        | 204/1100 [02:01<11:41,  1.28it/s]\u001b[32m [repeated 85x across cluster]\u001b[0m\n",
      "Loss: -647.3301:  43%|████▎     | 346/800 [01:40<02:36,  2.90it/s]\u001b[32m [repeated 63x across cluster]\u001b[0m\n",
      "Loss: -552.8942:  27%|██▋       | 244/900 [02:09<06:18,  1.73it/s]\u001b[32m [repeated 70x across cluster]\u001b[0m\n",
      "Loss: -452.3401:  32%|███▏      | 258/800 [02:08<05:12,  1.74it/s]\u001b[32m [repeated 99x across cluster]\u001b[0m\n",
      "Loss: -385.9277:  21%|██▏       | 234/1100 [02:22<06:56,  2.08it/s]\u001b[32m [repeated 105x across cluster]\u001b[0m\n",
      "Loss: -396.14:  22%|██▏       | 240/1100 [02:25<06:26,  2.23it/s]  \n",
      "Loss: -396.14:  22%|██▏       | 241/1100 [02:25<06:42,  2.13it/s]\n",
      "Loss: -429.1524:  22%|██▏       | 245/1100 [02:27<06:27,  2.21it/s]\u001b[32m [repeated 118x across cluster]\u001b[0m\n",
      "Loss: -705.7298:  31%|███       | 306/1000 [02:09<03:37,  3.19it/s]\u001b[32m [repeated 122x across cluster]\u001b[0m\n",
      "Loss: -484.1709:  39%|███▉      | 310/800 [02:28<03:05,  2.64it/s]\u001b[32m [repeated 115x across cluster]\u001b[0m\n",
      "Loss: -818.4765:  60%|██████    | 481/800 [02:37<01:36,  3.31it/s]\u001b[32m [repeated 82x across cluster]\u001b[0m\n",
      "Loss: -605.1095:  36%|███▌      | 322/900 [02:45<04:06,  2.34it/s]\u001b[32m [repeated 101x across cluster]\u001b[0m\n",
      "Loss: -839.1083:  62%|██████▏   | 497/800 [02:42<01:33,  3.25it/s]\n",
      "Loss: -839.1083:  62%|██████▏   | 498/800 [02:42<01:34,  3.19it/s]\n",
      "Loss: -819.687:  62%|██████▏   | 498/800 [02:43<01:34,  3.19it/s] \n",
      "Loss: -819.687:  62%|██████▏   | 499/800 [02:43<01:37,  3.10it/s]\n",
      "Loss: -882.0131:  64%|██████▍   | 514/800 [02:47<01:23,  3.42it/s]\u001b[32m [repeated 102x across cluster]\u001b[0m\n",
      "Loss: -364.1494:  50%|████▉     | 349/700 [02:39<02:17,  2.55it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
      "Loss: -508.3637:  38%|███▊      | 345/900 [02:55<03:52,  2.39it/s]\u001b[32m [repeated 112x across cluster]\u001b[0m\n",
      "Loss: -557.8043:  44%|████▍     | 355/800 [02:49<02:57,  2.50it/s]\n",
      "Loss: -557.8043:  44%|████▍     | 356/800 [02:49<02:59,  2.47it/s]\n",
      "Loss: -650.2228:  40%|███▉      | 357/900 [03:00<03:48,  2.38it/s]\u001b[32m [repeated 109x across cluster]\u001b[0m\n",
      "Loss: -1046.2927:  72%|███████▏  | 573/800 [02:38<00:56,  4.04it/s]\n",
      "Loss: -1046.2927:  72%|███████▏  | 574/800 [02:38<00:59,  3.78it/s]\n",
      "Loss: -435.7507:  29%|██▉       | 323/1100 [03:07<08:07,  1.59it/s]\u001b[32m [repeated 94x across cluster]\u001b[0m\n",
      "Loss: -558.7137:  30%|███       | 333/1100 [03:12<06:26,  1.98it/s]\u001b[32m [repeated 108x across cluster]\u001b[0m\n",
      "Loss: -838.7938:  43%|████▎     | 431/1000 [02:54<03:08,  3.01it/s]\u001b[32m [repeated 116x across cluster]\u001b[0m\n",
      "Loss: -1004.0757:  78%|███████▊  | 620/800 [03:17<00:50,  3.60it/s]\u001b[32m [repeated 113x across cluster]\u001b[0m\n",
      "Loss: -653.3168:  46%|████▌     | 415/900 [03:25<03:12,  2.53it/s]\u001b[32m [repeated 117x across cluster]\u001b[0m\n",
      "Loss: -750.9539:  56%|█████▌    | 444/800 [03:24<02:14,  2.65it/s]\u001b[32m [repeated 115x across cluster]\u001b[0m\n",
      "Loss: -1041.4125:  84%|████████▍ | 676/800 [03:33<00:35,  3.45it/s]\u001b[32m [repeated 118x across cluster]\u001b[0m\n",
      "Loss: -1133.2123:  91%|█████████▏| 730/800 [03:16<00:17,  4.07it/s]\n",
      "Loss: -981.5027:  87%|████████▋ | 695/800 [03:38<00:29,  3.59it/s]\u001b[32m [repeated 116x across cluster]\u001b[0m\n",
      "Loss: -1134.0133:  91%|█████████▏| 731/800 [03:16<00:16,  4.09it/s]\n",
      "Loss: -1148.207:  92%|█████████▏| 732/800 [03:16<00:15,  4.27it/s] \n",
      "Loss: -1149.1084:  92%|█████████▏| 733/800 [03:16<00:15,  4.19it/s]\n",
      "Loss: -1154.8455:  92%|█████████▏| 734/800 [03:17<00:14,  4.44it/s]\n",
      "Loss: -1162.9106:  92%|█████████▏| 735/800 [03:17<00:15,  4.30it/s]\n",
      "Loss: -1161.5493:  92%|█████████▏| 736/800 [03:17<00:14,  4.29it/s]\n",
      "Loss: -1166.9016:  92%|█████████▏| 737/800 [03:17<00:15,  4.12it/s]\n",
      "Loss: -1172.5321:  92%|█████████▏| 738/800 [03:17<00:13,  4.46it/s]\n",
      "Loss: -1170.4016:  92%|█████████▏| 739/800 [03:18<00:14,  4.26it/s]\n",
      "Loss: -1180.3917:  92%|█████████▎| 740/800 [03:18<00:14,  4.20it/s]\n",
      "Loss: -1179.3264:  93%|█████████▎| 741/800 [03:18<00:14,  4.10it/s]\n",
      "Loss: -1177.8141:  93%|█████████▎| 742/800 [03:18<00:13,  4.26it/s]\n",
      "Loss: -1169.3013:  93%|█████████▎| 743/800 [03:19<00:14,  3.88it/s]\n",
      "Loss: -1165.5972:  93%|█████████▎| 744/800 [03:19<00:13,  4.07it/s]\n",
      "Loss: -1137.3148:  93%|█████████▎| 745/800 [03:19<00:13,  4.04it/s]\n",
      "Loss: -1083.1776:  93%|█████████▎| 746/800 [03:19<00:13,  4.06it/s]\n",
      "Loss: -869.1605:  93%|█████████▎| 747/800 [03:20<00:12,  4.13it/s] \n",
      "Loss: -707.6714:  94%|█████████▎| 748/800 [03:20<00:11,  4.35it/s]\n",
      "Loss: -731.8971:  94%|█████████▎| 749/800 [03:20<00:11,  4.32it/s]\n",
      "Loss: -1028.979:  94%|█████████▍| 750/800 [03:20<00:10,  4.63it/s]\n",
      "Loss: -1010.5399:  94%|█████████▍| 751/800 [03:21<00:11,  4.31it/s]\n",
      "Loss: -1054.8773:  94%|█████████▍| 752/800 [03:21<00:11,  4.30it/s]\n",
      "Loss: -534.4481:  44%|████▎     | 348/800 [03:31<03:41,  2.04it/s]\u001b[32m [repeated 97x across cluster]\u001b[0m\n",
      "Loss: -1081.2876:  94%|█████████▍| 753/800 [03:21<00:10,  4.30it/s]\n",
      "Loss: -1089.6494:  94%|█████████▍| 754/800 [03:21<00:10,  4.22it/s]\n",
      "Loss: -1120.877:  94%|█████████▍| 755/800 [03:22<00:10,  4.11it/s] \n",
      "Loss: -1136.7811:  94%|█████████▍| 756/800 [03:22<00:10,  4.35it/s]\n",
      "Loss: -1124.4189:  95%|█████████▍| 757/800 [03:22<00:09,  4.38it/s]\n",
      "Loss: -1140.1498:  95%|█████████▍| 758/800 [03:22<00:09,  4.31it/s]\n",
      "Loss: -1146.7227:  95%|█████████▍| 759/800 [03:23<00:10,  3.84it/s]\n",
      "Loss: -1159.7351:  95%|█████████▌| 760/800 [03:23<00:10,  3.89it/s]\n",
      "Loss: -1159.5687:  95%|█████████▌| 761/800 [03:23<00:09,  3.94it/s]\n",
      "Loss: -1167.8042:  95%|█████████▌| 762/800 [03:23<00:09,  4.10it/s]\n",
      "Loss: -1169.6135:  95%|█████████▌| 763/800 [03:23<00:08,  4.15it/s]\n",
      "Loss: -1177.2229:  96%|█████████▌| 764/800 [03:24<00:09,  3.87it/s]\n",
      "Loss: -1166.6669:  96%|█████████▌| 765/800 [03:24<00:09,  3.87it/s]\n",
      "Loss: -1169.1593:  96%|█████████▌| 766/800 [03:24<00:08,  3.89it/s]\n",
      "Loss: -1152.2914:  96%|█████████▌| 767/800 [03:25<00:08,  3.78it/s]\n",
      "Loss: -1150.6564:  96%|█████████▌| 768/800 [03:25<00:08,  3.95it/s]\n",
      "Loss: -1154.1322:  96%|█████████▌| 769/800 [03:25<00:07,  4.27it/s]\n",
      "Loss: -1119.0579:  96%|█████████▋| 770/800 [03:25<00:07,  4.01it/s]\n",
      "Loss: -1110.932:  96%|█████████▋| 771/800 [03:26<00:07,  3.96it/s] \n",
      "Loss: -1050.5742:  91%|█████████▏| 730/800 [03:48<00:20,  3.42it/s]\n",
      "Loss: -1052.265:  96%|█████████▋| 772/800 [03:26<00:07,  3.72it/s]\n",
      "Loss: -631.5317:  38%|███▊      | 418/1100 [03:53<05:48,  1.96it/s]\u001b[32m [repeated 84x across cluster]\u001b[0m\n",
      "Loss: -954.9961:  97%|█████████▋| 773/800 [03:26<00:06,  3.93it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
      "Loss: -906.0775:  54%|█████▍    | 489/900 [03:55<02:47,  2.45it/s]\u001b[32m [repeated 76x across cluster]\u001b[0m\n",
      "Loss: -1065.7889:  94%|█████████▎| 749/800 [03:53<00:14,  3.54it/s]\u001b[32m [repeated 39x across cluster]\u001b[0m\n",
      "Loss: -1231.39: 100%|██████████| 800/800 [03:32<00:00,  3.76it/s]  \n",
      "2024-01-10 21:45:35,334\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', 'ebf8d860'), 'epochs': ('__ref_ph', 'cfeef121'), 'time_steps': ('__ref_ph', '761202b5')}\n",
      "Loss: -711.07:  40%|████      | 440/1100 [04:03<05:09,  2.14it/s]\u001b[32m [repeated 74x across cluster]\u001b[0m\n",
      "Loss: -1096.9473:  96%|█████████▌| 768/800 [03:58<00:08,  3.71it/s]\u001b[32m [repeated 24x across cluster]\u001b[0m\n",
      "  0%|          | 0/1100 [00:00<?, ?it/s]\n",
      "Loss: 12127.3564:   0%|          | 1/1100 [00:00<16:02,  1.14it/s]\n",
      "Loss: -737.6343:  41%|████      | 450/1100 [04:08<05:33,  1.95it/s]\u001b[32m [repeated 76x across cluster]\u001b[0m\n",
      "Loss: -1140.5298:  98%|█████████▊| 786/800 [04:03<00:03,  4.06it/s]\u001b[32m [repeated 18x across cluster]\u001b[0m\n",
      "Loss: -900.2884: 100%|██████████| 800/800 [04:07<00:00,  3.23it/s]\n",
      "2024-01-10 21:45:48,243\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '674f5ac8'), 'epochs': ('__ref_ph', 'cfeef121'), 'time_steps': ('__ref_ph', '761202b5')}\n",
      "Loss: -1078.176:  60%|█████▉    | 599/1000 [03:50<02:12,  3.02it/s]\u001b[32m [repeated 85x across cluster]\u001b[0m\n",
      "Loss: -786.3461: 100%|█████████▉| 799/800 [04:07<00:00,  3.61it/s]\u001b[32m [repeated 13x across cluster]\u001b[0m\n",
      "  0%|          | 0/1000 [00:00<?, ?it/s]\n",
      "Loss: 1857.1732:   0%|          | 1/1000 [00:00<05:16,  3.16it/s]\n",
      "Loss: 838.1516:   0%|          | 2/1000 [00:00<05:36,  2.97it/s]\u001b[32m [repeated 84x across cluster]\u001b[0m\n",
      "Loss: 98.3725:   2%|▏         | 17/1000 [00:05<05:08,  3.19it/s]\u001b[32m [repeated 98x across cluster]\u001b[0m\n",
      "Loss: -827.3358:  45%|████▍     | 492/1100 [04:28<05:21,  1.89it/s]\u001b[32m [repeated 90x across cluster]\u001b[0m\n",
      "Loss: -839.7262:  45%|████▌     | 500/1100 [04:33<05:34,  1.79it/s]\u001b[32m [repeated 75x across cluster]\u001b[0m\n",
      "Loss: -166.2049:   5%|▌         | 54/1000 [00:20<06:29,  2.43it/s]\u001b[32m [repeated 86x across cluster]\u001b[0m\n",
      "Loss: -944.1436:  91%|█████████▏| 639/700 [04:30<00:24,  2.53it/s]\n",
      "Loss: -948.1514:  91%|█████████▏| 640/700 [04:30<00:23,  2.51it/s]\n",
      "Loss: -949.1461:  92%|█████████▏| 641/700 [04:31<00:24,  2.44it/s]\n",
      "Loss: -950.5551:  92%|█████████▏| 642/700 [04:31<00:23,  2.46it/s]\n",
      "Loss: -204.0097:   7%|▋         | 67/1000 [00:25<05:53,  2.64it/s]\u001b[32m [repeated 84x across cluster]\u001b[0m\n",
      "Loss: -951.4547:  92%|█████████▏| 643/700 [04:32<00:23,  2.48it/s]\n",
      "Loss: -938.0853:  92%|█████████▏| 644/700 [04:32<00:22,  2.49it/s]\n",
      "Loss: -934.8396:  92%|█████████▏| 645/700 [04:33<00:21,  2.50it/s]\n",
      "Loss: -921.6472:  92%|█████████▏| 646/700 [04:33<00:21,  2.48it/s]\n",
      "Loss: -933.6413:  92%|█████████▏| 647/700 [04:33<00:20,  2.58it/s]\n",
      "Loss: -909.2312:  93%|█████████▎| 648/700 [04:34<00:20,  2.56it/s]\n",
      "Loss: -798.6295:  93%|█████████▎| 649/700 [04:34<00:20,  2.51it/s]\n",
      "Loss: -568.9406:  93%|█████████▎| 650/700 [04:34<00:19,  2.57it/s]\n",
      "Loss: -534.5858:  93%|█████████▎| 651/700 [04:35<00:19,  2.49it/s]\n",
      "Loss: -600.701:  93%|█████████▎| 652/700 [04:35<00:19,  2.50it/s] \n",
      "Loss: -846.728:  93%|█████████▎| 653/700 [04:36<00:18,  2.52it/s]\n",
      "Loss: -767.2161:  93%|█████████▎| 654/700 [04:36<00:18,  2.48it/s]\n",
      "Loss: -869.6505:  94%|█████████▎| 655/700 [04:36<00:17,  2.52it/s]\n",
      "Loss: -869.9405:  79%|███████▉  | 632/800 [04:40<01:09,  2.42it/s]\u001b[32m [repeated 75x across cluster]\u001b[0m\n",
      "Loss: -852.0239:  94%|█████████▎| 656/700 [04:37<00:17,  2.50it/s]\n",
      "Loss: -890.9329:  94%|█████████▍| 657/700 [04:37<00:16,  2.55it/s]\n",
      "Loss: -889.611:  94%|█████████▍| 658/700 [04:38<00:16,  2.61it/s] \n",
      "Loss: -906.1602:  94%|█████████▍| 659/700 [04:38<00:16,  2.55it/s]\n",
      "Loss: -895.3598:  94%|█████████▍| 660/700 [04:38<00:16,  2.48it/s]\n",
      "Loss: -917.1359:  94%|█████████▍| 661/700 [04:39<00:15,  2.54it/s]\n",
      "Loss: -924.7666:  95%|█████████▍| 662/700 [04:39<00:14,  2.56it/s]\n",
      "Loss: -920.5989:  95%|█████████▍| 663/700 [04:40<00:14,  2.50it/s]\n",
      "Loss: -916.8615:  95%|█████████▍| 664/700 [04:40<00:15,  2.36it/s]\n",
      "Loss: -916.735:  95%|█████████▌| 665/700 [04:41<00:16,  2.10it/s] \n",
      "Loss: -265.0231:   9%|▉         | 90/1000 [00:35<11:10,  1.36it/s]\u001b[32m [repeated 61x across cluster]\u001b[0m\n",
      "Loss: -899.1058:  95%|█████████▌| 666/700 [04:42<00:23,  1.45it/s]\n",
      "Loss: -851.3477:  95%|█████████▌| 667/700 [04:43<00:26,  1.26it/s]\n",
      "Loss: -879.9018:  95%|█████████▌| 668/700 [04:44<00:23,  1.36it/s]\n",
      "Loss: -881.54:  96%|█████████▌| 669/700 [04:44<00:19,  1.57it/s]  \n",
      "Loss: -864.049:  96%|█████████▌| 670/700 [04:45<00:19,  1.57it/s]\n",
      "Loss: -849.6136:  96%|█████████▌| 671/700 [04:45<00:17,  1.70it/s]\n",
      "Loss: -992.9401:  81%|████████  | 647/800 [04:49<01:44,  1.47it/s]\n",
      "Loss: -992.9401:  81%|████████  | 648/800 [04:49<01:39,  1.53it/s]\n",
      "Loss: -887.2234:  96%|█████████▌| 672/700 [04:45<00:14,  1.87it/s]\n",
      "Loss: -895.4613:  96%|█████████▌| 673/700 [04:46<00:14,  1.88it/s]\n",
      "Loss: -918.8047:  96%|█████████▋| 674/700 [04:46<00:13,  1.98it/s]\n",
      "Loss: -864.3073:  49%|████▉     | 542/1100 [04:58<06:32,  1.42it/s]\u001b[32m [repeated 49x across cluster]\u001b[0m\n",
      "Loss: -902.6119:  96%|█████████▋| 675/700 [04:47<00:12,  2.07it/s]\n",
      "Loss: -917.7122:  97%|█████████▋| 676/700 [04:47<00:11,  2.12it/s]\n",
      "Loss: -929.7492:  97%|█████████▋| 677/700 [04:48<00:10,  2.18it/s]\n",
      "Loss: -945.8818:  97%|█████████▋| 678/700 [04:48<00:09,  2.22it/s]\n",
      "Loss: -944.5731:  97%|█████████▋| 679/700 [04:49<00:09,  2.30it/s]\n",
      "Loss: -948.4271:  97%|█████████▋| 680/700 [04:49<00:08,  2.37it/s]\n",
      "Loss: -957.2618:  97%|█████████▋| 681/700 [04:49<00:08,  2.34it/s]\n",
      "Loss: -965.3005:  97%|█████████▋| 682/700 [04:50<00:07,  2.37it/s]\n",
      "Loss: -970.2897:  98%|█████████▊| 683/700 [04:50<00:06,  2.43it/s]\n",
      "Loss: -971.1611:  98%|█████████▊| 684/700 [04:51<00:06,  2.37it/s]\n",
      "Loss: -972.712:  98%|█████████▊| 685/700 [04:51<00:06,  2.36it/s] \n",
      "Loss: -977.8014:  98%|█████████▊| 686/700 [04:51<00:05,  2.40it/s]\n",
      "Loss: -1183.8494:  73%|███████▎  | 728/1000 [04:41<01:49,  2.49it/s]\u001b[32m [repeated 75x across cluster]\u001b[0m\n",
      "Loss: -975.7598:  98%|█████████▊| 687/700 [04:52<00:05,  2.36it/s]\n",
      "Loss: -983.7124:  98%|█████████▊| 688/700 [04:52<00:05,  2.36it/s]\n",
      "Loss: -982.6243:  98%|█████████▊| 689/700 [04:53<00:04,  2.25it/s]\n",
      "Loss: -961.2883:  99%|█████████▊| 690/700 [04:53<00:04,  2.42it/s]\n",
      "Loss: -900.9896:  99%|█████████▊| 691/700 [04:54<00:03,  2.47it/s]\n",
      "Loss: -904.4866:  99%|█████████▉| 692/700 [04:54<00:03,  2.48it/s]\n",
      "Loss: -914.4436:  99%|█████████▉| 693/700 [04:54<00:02,  2.45it/s]\n",
      "Loss: -884.2227:  99%|█████████▉| 694/700 [04:55<00:02,  2.54it/s]\n",
      "Loss: -799.4374:  99%|█████████▉| 695/700 [04:55<00:01,  2.52it/s]\n",
      "Loss: -714.8528:  99%|█████████▉| 696/700 [04:56<00:01,  2.56it/s]\n",
      "Loss: -726.6332: 100%|█████████▉| 697/700 [04:56<00:01,  2.54it/s]\n",
      "Loss: -850.0144: 100%|█████████▉| 698/700 [04:56<00:00,  2.61it/s]\n",
      "Loss: -925.6505:  51%|█████     | 561/1100 [05:09<04:38,  1.94it/s]\u001b[32m [repeated 77x across cluster]\u001b[0m\n",
      "Loss: -840.0666: 100%|█████████▉| 699/700 [04:57<00:00,  2.54it/s]\n",
      "Loss: -890.4391: 100%|██████████| 700/700 [04:57<00:00,  2.35it/s]\n",
      "2024-01-10 21:46:45,363\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '2944c5dd'), 'epochs': ('__ref_ph', '69f4b3a6'), 'time_steps': ('__ref_ph', '761202b5')}\n",
      "Loss: -161.8994:   9%|▉         | 104/1100 [01:08<10:30,  1.58it/s]\u001b[32m [repeated 77x across cluster]\u001b[0m\n",
      "  0%|          | 0/1000 [00:00<?, ?it/s]\n",
      "Loss: 7289.8892:   0%|          | 1/1000 [00:01<26:25,  1.59s/it]\n",
      "Loss: -380.5115:  15%|█▍        | 149/1000 [01:01<08:35,  1.65it/s]\u001b[32m [repeated 61x across cluster]\u001b[0m\n",
      "Loss: -910.9284:  53%|█████▎    | 584/1100 [05:24<06:19,  1.36it/s]\u001b[32m [repeated 57x across cluster]\u001b[0m\n",
      "Loss: -428.2836:  17%|█▋        | 170/1000 [01:11<05:58,  2.32it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "Loss: -467.1832:  18%|█▊        | 183/1000 [01:16<05:11,  2.63it/s]\u001b[32m [repeated 84x across cluster]\u001b[0m\n",
      "Loss: -1106.7034:  91%|█████████▏| 730/800 [05:26<00:30,  2.29it/s]\n",
      "Loss: -1106.7545:  91%|█████████▏| 731/800 [05:26<00:28,  2.39it/s]\n",
      "Loss: -1100.0377:  92%|█████████▏| 732/800 [05:27<00:29,  2.32it/s]\n",
      "Loss: -1095.9045:  92%|█████████▏| 733/800 [05:28<00:32,  2.04it/s]\n",
      "Loss: -1115.1992:  92%|█████████▏| 734/800 [05:28<00:35,  1.89it/s]\n",
      "Loss: -1103.2513:  92%|█████████▏| 735/800 [05:29<00:33,  1.92it/s]\n",
      "Loss: -1106.9934:  92%|█████████▏| 736/800 [05:29<00:32,  1.98it/s]\n",
      "Loss: -1102.6871:  92%|█████████▏| 737/800 [05:30<00:32,  1.94it/s]\n",
      "Loss: -1056.2111:  92%|█████████▏| 738/800 [05:30<00:29,  2.07it/s]\n",
      "Loss: -1071.1095:  78%|███████▊  | 699/900 [05:37<01:38,  2.05it/s]\u001b[32m [repeated 66x across cluster]\u001b[0m\n",
      "Loss: -990.1469:  92%|█████████▏| 739/800 [05:31<00:28,  2.14it/s] \n",
      "Loss: -849.4861:  92%|█████████▎| 740/800 [05:31<00:27,  2.19it/s]\n",
      "Loss: -337.8382:  93%|█████████▎| 741/800 [05:31<00:26,  2.19it/s]\n",
      "Loss: -123.4529:  93%|█████████▎| 742/800 [05:32<00:26,  2.16it/s]\n",
      "Loss: -494.3065:  93%|█████████▎| 743/800 [05:32<00:27,  2.08it/s]\n",
      "Loss: -834.7684:  93%|█████████▎| 744/800 [05:33<00:26,  2.13it/s]\n",
      "Loss: -652.7723:  93%|█████████▎| 745/800 [05:33<00:26,  2.08it/s]\n",
      "Loss: -919.7278:  93%|█████████▎| 746/800 [05:34<00:24,  2.16it/s]\n",
      "Loss: -865.1677:  93%|█████████▎| 747/800 [05:34<00:24,  2.14it/s]\n",
      "Loss: -920.9379:  94%|█████████▎| 748/800 [05:35<00:24,  2.12it/s]\n",
      "Loss: -950.7957:  94%|█████████▎| 749/800 [05:35<00:22,  2.23it/s]\n",
      "Loss: -1190.5637:  79%|███████▉  | 709/900 [05:42<01:38,  1.93it/s]\u001b[32m [repeated 65x across cluster]\u001b[0m\n",
      "Loss: -969.3805:  94%|█████████▍| 750/800 [05:36<00:23,  2.08it/s]\n",
      "Loss: -602.897:  83%|████████▎ | 827/1000 [05:22<01:16,  2.26it/s] \n",
      "Loss: -602.897:  83%|████████▎ | 828/1000 [05:22<01:34,  1.81it/s]\n",
      "Loss: -985.5894:  94%|█████████▍| 751/800 [05:37<00:31,  1.56it/s]\n",
      "Loss: -991.9973:  94%|█████████▍| 752/800 [05:37<00:32,  1.49it/s]\n",
      "Loss: -1000.3007:  94%|█████████▍| 753/800 [05:38<00:29,  1.57it/s]\n",
      "Loss: -1015.5933:  94%|█████████▍| 754/800 [05:39<00:28,  1.64it/s]\n",
      "Loss: -1012.4239:  94%|█████████▍| 755/800 [05:39<00:27,  1.67it/s]\n",
      "Loss: -1012.5552:  94%|█████████▍| 756/800 [05:40<00:24,  1.81it/s]\n",
      "Loss: -1031.8236:  95%|█████████▍| 757/800 [05:40<00:22,  1.88it/s]\n",
      "Loss: -1029.0747:  95%|█████████▍| 758/800 [05:41<00:21,  1.91it/s]\n",
      "Loss: -529.1183:  22%|██▏       | 215/1000 [01:31<06:51,  1.91it/s]\u001b[32m [repeated 47x across cluster]\u001b[0m\n",
      "Loss: -1029.1598:  95%|█████████▍| 759/800 [05:41<00:21,  1.95it/s]\n",
      "Loss: -418.7783:  22%|██▏       | 217/1000 [01:32<06:15,  2.08it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
      "Loss: -1026.0532:  95%|█████████▌| 760/800 [05:42<00:20,  2.00it/s]\n",
      "Loss: -1005.9974:  95%|█████████▌| 761/800 [05:42<00:18,  2.13it/s]\n",
      "Loss: -997.5983:  95%|█████████▌| 762/800 [05:42<00:17,  2.19it/s] \n",
      "Loss: -1030.6487:  95%|█████████▌| 763/800 [05:43<00:18,  2.02it/s]\n",
      "Loss: -987.4738:  96%|█████████▌| 764/800 [05:43<00:17,  2.09it/s] \n",
      "Loss: -987.8124:  96%|█████████▌| 765/800 [05:44<00:16,  2.15it/s]\n",
      "Loss: -950.3144:  96%|█████████▌| 766/800 [05:44<00:15,  2.13it/s]\n",
      "Loss: -923.4738:  96%|█████████▌| 767/800 [05:45<00:15,  2.20it/s]\n",
      "Loss: -927.1852:  96%|█████████▌| 768/800 [05:45<00:14,  2.28it/s]\n",
      "Loss: -667.9753:  85%|████████▍ | 848/1000 [05:31<01:08,  2.22it/s]\u001b[32m [repeated 64x across cluster]\u001b[0m\n",
      "Loss: -986.2284:  96%|█████████▌| 769/800 [05:46<00:15,  2.01it/s]\n",
      "Loss: -1026.6573:  96%|█████████▋| 770/800 [05:46<00:16,  1.85it/s]\n",
      "Loss: -1007.4108:  96%|█████████▋| 771/800 [05:47<00:15,  1.88it/s]\n",
      "Loss: -1044.8308:  96%|█████████▋| 772/800 [05:50<00:33,  1.19s/it]\n",
      "Loss: -1052.2115:  97%|█████████▋| 773/800 [05:50<00:27,  1.03s/it]\n",
      "Loss: -1050.0436:  97%|█████████▋| 773/800 [05:51<00:27,  1.03s/it]\n",
      "Loss: -1050.0436:  97%|█████████▋| 774/800 [05:51<00:22,  1.17it/s]\n",
      "Loss: -209.162:  15%|█▍        | 160/1100 [01:53<20:27,  1.31s/it]\u001b[32m [repeated 29x across cluster]\u001b[0m\n",
      "Loss: -1056.202:  97%|█████████▋| 774/800 [05:51<00:22,  1.17it/s] \n",
      "Loss: -1056.202:  97%|█████████▋| 775/800 [05:51<00:18,  1.36it/s]\n",
      "Loss: -1058.5948:  97%|█████████▋| 776/800 [05:52<00:15,  1.53it/s]\n",
      "Loss: -1071.038:  97%|█████████▋| 777/800 [05:52<00:13,  1.71it/s] \n",
      "Loss: -1072.9344:  97%|█████████▋| 778/800 [05:53<00:11,  1.83it/s]\n",
      "Loss: -1076.3749:  97%|█████████▋| 779/800 [05:53<00:10,  2.00it/s]\n",
      "Loss: -1081.5638:  98%|█████████▊| 780/800 [05:53<00:09,  2.11it/s]\n",
      "Loss: -1087.9811:  98%|█████████▊| 781/800 [05:54<00:08,  2.17it/s]\n",
      "Loss: -1092.3507:  98%|█████████▊| 782/800 [05:54<00:08,  2.17it/s]\n",
      "Loss: -1095.204:  98%|█████████▊| 783/800 [05:55<00:07,  2.31it/s] \n",
      "Loss: -1097.2383:  98%|█████████▊| 784/800 [05:55<00:06,  2.32it/s]\n",
      "Loss: -1097.4493:  98%|█████████▊| 785/800 [05:55<00:06,  2.29it/s]\n",
      "Loss: -979.02:  59%|█████▉    | 648/1100 [06:04<04:03,  1.85it/s]\u001b[32m [repeated 74x across cluster]\u001b[0m\n",
      "Loss: -1083.1237:  98%|█████████▊| 786/800 [05:56<00:06,  2.29it/s]\n",
      "Loss: -1045.2606:  98%|█████████▊| 787/800 [05:56<00:05,  2.34it/s]\n",
      "Loss: -991.1527:  98%|█████████▊| 788/800 [05:57<00:05,  2.26it/s] \n",
      "Loss: -920.5564:  99%|█████████▊| 789/800 [05:57<00:04,  2.37it/s]\n",
      "Loss: -836.0527:  99%|█████████▉| 790/800 [05:58<00:04,  2.30it/s]\n",
      "Loss: -999.9373:  99%|█████████▉| 791/800 [05:58<00:03,  2.28it/s]\n",
      "Loss: -1038.7174:  99%|█████████▉| 792/800 [05:58<00:03,  2.29it/s]\n",
      "Loss: -1039.1744:  99%|█████████▉| 793/800 [05:59<00:03,  2.26it/s]\n",
      "Loss: -1071.5028:  99%|█████████▉| 794/800 [05:59<00:02,  2.34it/s]\n",
      "Loss: -1081.033:  99%|█████████▉| 795/800 [06:00<00:02,  2.36it/s] \n",
      "Loss: -1090.8685: 100%|█████████▉| 796/800 [06:00<00:01,  2.16it/s]\n",
      "Loss: -1093.8843: 100%|█████████▉| 797/800 [06:01<00:01,  2.25it/s]\n",
      "Loss: -584.2969:  26%|██▌       | 259/1000 [01:51<04:53,  2.52it/s]\u001b[32m [repeated 70x across cluster]\u001b[0m\n",
      "Loss: -1103.3992: 100%|█████████▉| 798/800 [06:01<00:00,  2.29it/s]\n",
      "Loss: -1109.4576: 100%|█████████▉| 799/800 [06:02<00:00,  2.31it/s]\n",
      "Loss: -1110.5848: 100%|██████████| 800/800 [06:02<00:00,  2.21it/s]\n",
      "2024-01-10 21:47:46,878\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '674f5ac8'), 'epochs': ('__ref_ph', 'cfeef121'), 'time_steps': ('__ref_ph', '7be63cef')}\n",
      "Loss: -604.0:  27%|██▋       | 271/1000 [01:56<04:44,  2.56it/s]\u001b[32m [repeated 70x across cluster]\u001b[0m\n",
      "  0%|          | 0/800 [00:00<?, ?it/s]\n",
      "Loss: 1540.7661:   0%|          | 1/800 [00:00<03:05,  4.31it/s]\n",
      "Loss: 760.4235:  86%|████████▌ | 774/900 [06:17<01:04,  1.95it/s]\u001b[32m [repeated 74x across cluster]\u001b[0m\n",
      "Loss: -820.6873:  91%|█████████▏| 913/1000 [05:59<00:31,  2.73it/s]\n",
      "Loss: -821.6628:  91%|█████████▏| 914/1000 [05:59<00:31,  2.70it/s]\n",
      "Loss: -821.8146:  92%|█████████▏| 915/1000 [05:59<00:33,  2.53it/s]\n",
      "Loss: -822.7372:  92%|█████████▏| 916/1000 [06:00<00:35,  2.36it/s]\n",
      "Loss: -822.6357:  92%|█████████▏| 917/1000 [06:00<00:34,  2.39it/s]\n",
      "Loss: -985.8288:  62%|██████▏   | 683/1100 [06:23<03:56,  1.76it/s]\n",
      "Loss: -985.8288:  62%|██████▏   | 684/1100 [06:23<03:50,  1.81it/s]\n",
      "Loss: -823.7349:  92%|█████████▏| 918/1000 [06:01<00:33,  2.43it/s]\n",
      "Loss: -825.0614:  92%|█████████▏| 919/1000 [06:01<00:34,  2.33it/s]\n",
      "Loss: -24.3756:   3%|▎         | 24/800 [00:06<03:20,  3.86it/s]\u001b[32m [repeated 79x across cluster]\u001b[0m\n",
      "Loss: -824.8899:  92%|█████████▏| 920/1000 [06:02<00:35,  2.22it/s]\n",
      "Loss: -826.3404:  92%|█████████▏| 921/1000 [06:02<00:36,  2.18it/s]\n",
      "Loss: -827.4328:  92%|█████████▏| 922/1000 [06:03<00:35,  2.22it/s]\n",
      "Loss: -827.8558:  92%|█████████▏| 923/1000 [06:03<00:32,  2.35it/s]\n",
      "Loss: -828.7617:  92%|█████████▏| 924/1000 [06:04<00:33,  2.28it/s]\n",
      "Loss: -829.2874:  92%|█████████▎| 925/1000 [06:04<00:32,  2.30it/s]\n",
      "Loss: -829.7929:  93%|█████████▎| 926/1000 [06:04<00:31,  2.34it/s]\n",
      "Loss: -830.3265:  93%|█████████▎| 927/1000 [06:05<00:29,  2.46it/s]\n",
      "Loss: -830.708:  93%|█████████▎| 928/1000 [06:05<00:28,  2.51it/s] \n",
      "Loss: -832.2344:  93%|█████████▎| 929/1000 [06:05<00:28,  2.51it/s]\n",
      "Loss: -907.8646:  63%|██████▎   | 692/1100 [06:28<03:54,  1.74it/s]\u001b[32m [repeated 4x across cluster]\u001b[0m\n",
      "Loss: -832.8113:  93%|█████████▎| 930/1000 [06:06<00:28,  2.47it/s]\n",
      "Loss: -833.4427:  93%|█████████▎| 931/1000 [06:06<00:27,  2.49it/s]\n",
      "Loss: -987.506:  63%|██████▎   | 694/1100 [06:30<03:55,  1.72it/s]\u001b[32m [repeated 73x across cluster]\u001b[0m\n",
      "Loss: -835.0615:  93%|█████████▎| 932/1000 [06:07<00:27,  2.50it/s]\n",
      "Loss: -835.3948:  93%|█████████▎| 933/1000 [06:07<00:26,  2.54it/s]\n",
      "Loss: -836.1642:  93%|█████████▎| 934/1000 [06:07<00:25,  2.61it/s]\n",
      "Loss: -837.2074:  94%|█████████▎| 935/1000 [06:08<00:25,  2.57it/s]\n",
      "Loss: -837.9713:  94%|█████████▎| 936/1000 [06:08<00:24,  2.65it/s]\n",
      "Loss: -838.6796:  94%|█████████▎| 937/1000 [06:09<00:23,  2.63it/s]\n",
      "Loss: -838.7583:  94%|█████████▍| 938/1000 [06:09<00:24,  2.56it/s]\n",
      "Loss: -840.3104:  94%|█████████▍| 939/1000 [06:09<00:24,  2.47it/s]\n",
      "Loss: -840.7042:  94%|█████████▍| 940/1000 [06:10<00:24,  2.49it/s]\n",
      "Loss: -839.5643:  94%|█████████▍| 941/1000 [06:10<00:23,  2.53it/s]\n",
      "Loss: -842.0989:  94%|█████████▍| 942/1000 [06:11<00:23,  2.51it/s]\n",
      "Loss: -842.6834:  94%|█████████▍| 943/1000 [06:11<00:22,  2.48it/s]\n",
      "Loss: -843.5247:  94%|█████████▍| 944/1000 [06:11<00:21,  2.55it/s]\n",
      "Loss: -211.7632:   8%|▊         | 67/800 [00:16<03:14,  3.77it/s]\u001b[32m [repeated 79x across cluster]\u001b[0m\n",
      "Loss: -844.5114:  94%|█████████▍| 945/1000 [06:12<00:26,  2.06it/s]\n",
      "Loss: -842.9993:  95%|█████████▍| 946/1000 [06:12<00:24,  2.19it/s]\n",
      "Loss: -845.4414:  95%|█████████▍| 947/1000 [06:13<00:23,  2.24it/s]\n",
      "Loss: -845.7116:  95%|█████████▍| 948/1000 [06:13<00:22,  2.30it/s]\n",
      "Loss: -847.3743:  95%|█████████▍| 949/1000 [06:14<00:22,  2.30it/s]\n",
      "Loss: -846.7965:  95%|█████████▌| 950/1000 [06:14<00:19,  2.52it/s]\n",
      "Loss: -847.7102:  95%|█████████▌| 951/1000 [06:15<00:20,  2.40it/s]\n",
      "Loss: -849.4777:  95%|█████████▌| 952/1000 [06:15<00:19,  2.40it/s]\n",
      "Loss: -850.0008:  95%|█████████▌| 953/1000 [06:15<00:19,  2.36it/s]\n",
      "Loss: -850.2791:  95%|█████████▌| 954/1000 [06:16<00:19,  2.38it/s]\n",
      "Loss: -850.6278:  96%|█████████▌| 955/1000 [06:16<00:19,  2.32it/s]\n",
      "Loss: -851.662:  96%|█████████▌| 956/1000 [06:17<00:19,  2.31it/s] \n",
      "Loss: -1006.0767:  65%|██████▍   | 711/1100 [06:40<04:02,  1.61it/s]\u001b[32m [repeated 73x across cluster]\u001b[0m\n",
      "Loss: -852.9038:  96%|█████████▌| 957/1000 [06:17<00:18,  2.34it/s]\n",
      "Loss: -853.4938:  96%|█████████▌| 958/1000 [06:17<00:17,  2.42it/s]\n",
      "Loss: -854.1745:  96%|█████████▌| 959/1000 [06:18<00:16,  2.56it/s]\n",
      "Loss: -855.626:  96%|█████████▌| 960/1000 [06:18<00:15,  2.59it/s] \n",
      "Loss: -855.9399:  96%|█████████▌| 961/1000 [06:19<00:16,  2.35it/s]\n",
      "Loss: -856.4637:  96%|█████████▌| 962/1000 [06:19<00:16,  2.36it/s]\n",
      "Loss: -856.7008:  96%|█████████▋| 963/1000 [06:19<00:14,  2.48it/s]\n",
      "Loss: -827.8618:  91%|█████████▏| 822/900 [06:40<00:37,  2.09it/s]\n",
      "Loss: -858.7175:  96%|█████████▋| 964/1000 [06:20<00:14,  2.54it/s]\n",
      "Loss: -858.5803:  96%|█████████▋| 965/1000 [06:20<00:13,  2.59it/s]\n",
      "Loss: -575.8174:  34%|███▍      | 345/1000 [02:27<04:07,  2.65it/s]\u001b[32m [repeated 74x across cluster]\u001b[0m\n",
      "Loss: -862.2153:  92%|█████████▏| 832/900 [06:45<00:40,  1.68it/s]\u001b[32m [repeated 20x across cluster]\u001b[0m\n",
      "Loss: -353.1564:  16%|█▌        | 125/800 [00:31<03:59,  2.82it/s]\n",
      "Loss: -353.1564:  16%|█▌        | 126/800 [00:31<03:44,  3.00it/s]\n",
      "Loss: -601.5409:  35%|███▌      | 354/1000 [02:32<06:14,  1.73it/s]\u001b[32m [repeated 51x across cluster]\u001b[0m\n",
      "Loss: -874.2639:  99%|█████████▊| 987/1000 [06:30<00:04,  2.64it/s]\u001b[32m [repeated 20x across cluster]\u001b[0m\n",
      "Loss: -226.0914:  14%|█▍        | 142/1000 [01:38<10:41,  1.34it/s]\u001b[32m [repeated 61x across cluster]\u001b[0m\n",
      "Loss: -904.5126:  94%|█████████▍| 844/900 [06:54<00:34,  1.63it/s]\n",
      "Loss: -904.5126:  94%|█████████▍| 845/900 [06:54<00:37,  1.48it/s]\n",
      "Loss: -910.7386:  94%|█████████▍| 847/900 [06:56<00:45,  1.16it/s]\u001b[32m [repeated 14x across cluster]\u001b[0m\n",
      "Loss: -427.4048:  20%|██        | 162/800 [00:41<04:02,  2.63it/s]\u001b[32m [repeated 40x across cluster]\u001b[0m\n",
      "Loss: -884.1058: 100%|██████████| 1000/1000 [06:39<00:00,  2.50it/s]\n",
      "2024-01-10 21:48:38,116\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', 'b96bf624'), 'time_steps': ('__ref_ph', '9af98f97')}\n",
      "Loss: -929.1417:  95%|█████████▍| 853/900 [07:01<00:34,  1.35it/s]\u001b[32m [repeated 10x across cluster]\u001b[0m\n",
      "Loss: -458.1929:  22%|██▏       | 177/800 [00:46<04:08,  2.51it/s]\u001b[32m [repeated 46x across cluster]\u001b[0m\n",
      "Loss: -949.7367:  96%|█████████▌| 860/900 [07:05<00:29,  1.38it/s]\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "Loss: -1036.8688:  84%|████████▎ | 669/800 [06:54<02:30,  1.15s/it]\u001b[32m [repeated 36x across cluster]\u001b[0m\n",
      "  0%|          | 0/700 [00:00<?, ?it/s]\n",
      "Loss: 2059.3782:   0%|          | 1/700 [00:00<04:44,  2.46it/s]\n",
      "Loss: -949.6706:  96%|█████████▋| 868/900 [07:11<00:17,  1.84it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "Loss: -527.6794:  26%|██▋       | 212/800 [00:56<02:17,  4.29it/s]\u001b[32m [repeated 73x across cluster]\u001b[0m\n",
      "Loss: -533.5776:  27%|██▋       | 213/800 [00:57<02:11,  4.48it/s]\n",
      "Loss: -533.5776:  27%|██▋       | 214/800 [00:57<02:08,  4.58it/s]\n",
      "Loss: -991.1896:  98%|█████████▊| 879/900 [07:16<00:09,  2.27it/s]\u001b[32m [repeated 11x across cluster]\u001b[0m\n",
      "Loss: -573.6989:  29%|██▉       | 235/800 [01:02<02:04,  4.55it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "Loss: -768.8644:  41%|████      | 407/1000 [03:01<03:42,  2.66it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
      "Loss: -956.5146:  99%|█████████▉| 891/900 [07:21<00:04,  2.23it/s]\u001b[32m [repeated 12x across cluster]\u001b[0m\n",
      "Loss: -278.206:  18%|█▊        | 181/1000 [02:08<08:38,  1.58it/s]\u001b[32m [repeated 77x across cluster]\u001b[0m\n",
      "Loss: -1002.6403: 100%|██████████| 900/900 [07:26<00:00,  2.02it/s]\n",
      "2024-01-10 21:49:04,303\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', 'ebf8d860'), 'epochs': ('__ref_ph', 'c733eb3d'), 'time_steps': ('__ref_ph', '201c1c09')}\n",
      "Loss: -996.4538: 100%|█████████▉| 899/900 [07:25<00:00,  1.96it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "Loss: -1033.7314:  72%|███████▏  | 789/1100 [07:31<02:46,  1.86it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "  0%|          | 0/1100 [00:00<?, ?it/s]\n",
      "Loss: 1643.2065:   0%|          | 1/1100 [00:00<06:24,  2.86it/s]\n",
      "Loss: 643.663:   0%|          | 3/1100 [00:01<07:10,  2.55it/s]\u001b[32m [repeated 90x across cluster]\u001b[0m\n",
      "Loss: -587.7017:  40%|████      | 323/800 [01:22<01:43,  4.63it/s]\u001b[32m [repeated 96x across cluster]\u001b[0m\n",
      "Loss: -950.0383:  91%|█████████▏| 730/800 [07:29<00:38,  1.82it/s]\n",
      "Loss: -1007.9589:  74%|███████▍  | 819/1100 [07:46<02:31,  1.85it/s]\u001b[32m [repeated 92x across cluster]\u001b[0m\n",
      "Loss: -986.2288:  91%|█████████▏| 731/800 [07:30<00:39,  1.74it/s]\n",
      "Loss: -999.897:  92%|█████████▏| 732/800 [07:30<00:38,  1.78it/s] \n",
      "Loss: -1010.3773:  92%|█████████▏| 733/800 [07:31<00:38,  1.75it/s]\n",
      "Loss: -1030.9554:  92%|█████████▏| 734/800 [07:31<00:35,  1.84it/s]\n",
      "Loss: -1031.1011:  92%|█████████▏| 735/800 [07:32<00:34,  1.91it/s]\n",
      "Loss: -1041.2976:  92%|█████████▏| 736/800 [07:32<00:33,  1.91it/s]\n",
      "Loss: -1043.3668:  92%|█████████▏| 737/800 [07:33<00:33,  1.86it/s]\n",
      "Loss: -1054.7216:  92%|█████████▏| 738/800 [07:33<00:33,  1.86it/s]\n",
      "Loss: -1054.5698:  92%|█████████▏| 739/800 [07:34<00:33,  1.83it/s]\n",
      "Loss: -1085.577:  75%|███████▌  | 829/1100 [07:51<02:45,  1.64it/s]\u001b[32m [repeated 91x across cluster]\u001b[0m\n",
      "Loss: -1062.2721:  92%|█████████▎| 740/800 [07:35<00:38,  1.54it/s]\n",
      "Loss: -1064.0938:  93%|█████████▎| 741/800 [07:36<00:47,  1.25it/s]\n",
      "Loss: -1070.4559:  93%|█████████▎| 742/800 [07:37<00:47,  1.21it/s]\n",
      "Loss: -939.8669:  49%|████▉     | 494/1000 [03:36<04:50,  1.74it/s]\n",
      "Loss: -939.8669:  50%|████▉     | 495/1000 [03:36<04:27,  1.89it/s]\n",
      "Loss: -1076.3047:  93%|█████████▎| 743/800 [07:37<00:43,  1.30it/s]\n",
      "Loss: -1084.561:  93%|█████████▎| 744/800 [07:38<00:38,  1.46it/s] \n",
      "Loss: -1088.4757:  93%|█████████▎| 745/800 [07:39<00:39,  1.38it/s]\n",
      "Loss: -257.7814:  14%|█▍        | 100/700 [00:43<06:23,  1.56it/s]\u001b[32m [repeated 61x across cluster]\u001b[0m\n",
      "Loss: -1092.1093:  93%|█████████▎| 746/800 [07:40<00:45,  1.18it/s]\n",
      "Loss: -1094.6321:  93%|█████████▎| 747/800 [07:40<00:40,  1.32it/s]\n",
      "Loss: -1096.0775:  94%|█████████▎| 748/800 [07:41<00:37,  1.38it/s]\n",
      "Loss: -1084.8403:  94%|█████████▎| 749/800 [07:42<00:34,  1.48it/s]\n",
      "Loss: -1041.4695:  94%|█████████▍| 750/800 [07:42<00:32,  1.54it/s]\n",
      "Loss: -980.6555:  94%|█████████▍| 751/800 [07:43<00:32,  1.51it/s] \n",
      "Loss: -893.176:  94%|█████████▍| 752/800 [07:44<00:31,  1.51it/s] \n",
      "Loss: -860.0776:  94%|█████████▍| 753/800 [07:44<00:33,  1.42it/s]\n",
      "Loss: -968.0447:  51%|█████     | 511/1000 [03:43<03:52,  2.11it/s]\u001b[32m [repeated 77x across cluster]\u001b[0m\n",
      "Loss: -833.4542:  94%|█████████▍| 754/800 [07:45<00:30,  1.51it/s]\n",
      "Loss: -1034.3357:  94%|█████████▍| 755/800 [07:45<00:28,  1.59it/s]\n",
      "Loss: -993.6752:  94%|█████████▍| 756/800 [07:46<00:25,  1.72it/s] \n",
      "Loss: -1035.4087:  95%|█████████▍| 757/800 [07:46<00:24,  1.75it/s]\n",
      "Loss: -1074.6171:  95%|█████████▍| 758/800 [07:47<00:23,  1.77it/s]\n",
      "Loss: -1048.7297:  95%|█████████▍| 759/800 [07:48<00:22,  1.82it/s]\n",
      "Loss: -1089.5463:  95%|█████████▌| 760/800 [07:48<00:21,  1.86it/s]\n",
      "Loss: -1053.605:  95%|█████████▌| 761/800 [07:49<00:21,  1.85it/s] \n",
      "Loss: -214.126:   7%|▋         | 73/1100 [00:30<06:39,  2.57it/s] \n",
      "Loss: -214.126:   7%|▋         | 74/1100 [00:30<07:01,  2.43it/s]\n",
      "Loss: -1050.2065:  95%|█████████▌| 762/800 [07:50<00:26,  1.42it/s]\n",
      "Loss: -218.9762:   7%|▋         | 76/1100 [00:32<08:25,  2.03it/s]\u001b[32m [repeated 86x across cluster]\u001b[0m\n",
      "Loss: -1067.3805:  95%|█████████▌| 763/800 [07:50<00:25,  1.46it/s]\n",
      "Loss: -1068.322:  96%|█████████▌| 764/800 [07:51<00:23,  1.50it/s] \n",
      "Loss: -1019.3599:  96%|█████████▌| 765/800 [07:52<00:22,  1.52it/s]\n",
      "Loss: -906.4251:  96%|█████████▌| 766/800 [07:52<00:20,  1.62it/s] \n",
      "Loss: -738.6844:  96%|█████████▌| 767/800 [07:53<00:19,  1.71it/s]\n",
      "Loss: -638.8271:  96%|█████████▌| 768/800 [07:53<00:18,  1.73it/s]\n",
      "Loss: -681.9751:  96%|█████████▌| 769/800 [07:54<00:17,  1.76it/s]\n",
      "Loss: -968.6689:  96%|█████████▋| 770/800 [07:54<00:17,  1.70it/s]\n",
      "Loss: -943.6644:  96%|█████████▋| 771/800 [07:55<00:16,  1.79it/s]\n",
      "Loss: -253.1693:   8%|▊         | 89/1100 [00:37<06:16,  2.68it/s]\u001b[32m [repeated 86x across cluster]\u001b[0m\n",
      "Loss: -965.1331:  96%|█████████▋| 772/800 [07:55<00:15,  1.80it/s]\n",
      "Loss: -1012.2309:  97%|█████████▋| 773/800 [07:56<00:14,  1.81it/s]\n",
      "Loss: -998.9889:  97%|█████████▋| 774/800 [07:57<00:14,  1.75it/s] \n",
      "Loss: -1050.0289:  97%|█████████▋| 775/800 [07:57<00:13,  1.82it/s]\n",
      "Loss: -1037.8429:  97%|█████████▋| 776/800 [07:58<00:12,  1.88it/s]\n",
      "Loss: -1029.9938:  97%|█████████▋| 777/800 [07:58<00:11,  1.93it/s]\n",
      "Loss: -996.3845:  97%|█████████▋| 778/800 [07:59<00:11,  1.97it/s] \n",
      "Loss: -1011.4579:  97%|█████████▋| 779/800 [07:59<00:10,  1.98it/s]\n",
      "Loss: -1012.1542:  98%|█████████▊| 780/800 [08:00<00:10,  1.96it/s]\n",
      "Loss: -286.8296:   9%|▉         | 103/1100 [00:42<06:08,  2.70it/s]\u001b[32m [repeated 92x across cluster]\u001b[0m\n",
      "Loss: -1009.2227:  98%|█████████▊| 781/800 [08:00<00:09,  1.93it/s]\n",
      "Loss: -1030.0178:  98%|█████████▊| 782/800 [08:01<00:09,  1.90it/s]\n",
      "Loss: -1019.3331:  98%|█████████▊| 783/800 [08:01<00:08,  1.89it/s]\n",
      "Loss: -1063.3923:  98%|█████████▊| 784/800 [08:02<00:08,  1.84it/s]\n",
      "Loss: -1070.8643:  98%|█████████▊| 785/800 [08:02<00:08,  1.77it/s]\n",
      "Loss: -1074.3135:  98%|█████████▊| 786/800 [08:03<00:08,  1.68it/s]\n",
      "Loss: -1077.6691:  98%|█████████▊| 787/800 [08:04<00:07,  1.74it/s]\n",
      "Loss: -1093.5441:  98%|█████████▊| 788/800 [08:04<00:06,  1.73it/s]\n",
      "Loss: -1103.6089:  99%|█████████▊| 789/800 [08:05<00:06,  1.81it/s]\n",
      "Loss: -1117.1309:  80%|████████  | 882/1100 [08:22<01:51,  1.96it/s]\u001b[32m [repeated 83x across cluster]\u001b[0m\n",
      "Loss: -1096.5494:  99%|█████████▉| 790/800 [08:05<00:05,  1.82it/s]\n",
      "Loss: -1086.8743:  99%|█████████▉| 791/800 [08:06<00:05,  1.79it/s]\n",
      "Loss: -1057.5027:  99%|█████████▉| 792/800 [08:06<00:04,  1.73it/s]\n",
      "Loss: -1077.2617:  99%|█████████▉| 793/800 [08:07<00:03,  1.76it/s]\n",
      "Loss: -1056.033:  99%|█████████▉| 794/800 [08:07<00:03,  1.82it/s] \n",
      "Loss: -1040.7639:  99%|█████████▉| 795/800 [08:08<00:02,  1.83it/s]\n",
      "Loss: -1007.515: 100%|█████████▉| 796/800 [08:09<00:02,  1.80it/s] \n",
      "Loss: -990.4926: 100%|█████████▉| 797/800 [08:09<00:01,  1.86it/s]\n",
      "Loss: -968.4987: 100%|█████████▉| 798/800 [08:10<00:01,  1.81it/s]\n",
      "Loss: -342.2597:  12%|█▏        | 129/1100 [00:52<05:50,  2.77it/s]\u001b[32m [repeated 89x across cluster]\u001b[0m\n",
      "Loss: -950.6291: 100%|█████████▉| 799/800 [08:10<00:00,  1.82it/s]\n",
      "Loss: -874.1348: 100%|██████████| 800/800 [08:11<00:00,  1.63it/s]\n",
      "2024-01-10 21:50:03,710\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '674f5ac8'), 'epochs': ('__ref_ph', 'cfeef121'), 'time_steps': ('__ref_ph', '56612a6d')}\n",
      "Loss: -982.0934:  68%|██████▊   | 544/800 [02:13<00:55,  4.57it/s]\u001b[32m [repeated 90x across cluster]\u001b[0m\n",
      "  0%|          | 0/1100 [00:00<?, ?it/s]\n",
      "Loss: 5551.5103:   0%|          | 1/1100 [00:00<12:39,  1.45it/s]\n",
      "Loss: -389.1947:  14%|█▍        | 156/1100 [01:02<05:57,  2.64it/s]\u001b[32m [repeated 94x across cluster]\u001b[0m\n",
      "Loss: -433.4412:  15%|█▌        | 170/1100 [01:07<05:44,  2.70it/s]\u001b[32m [repeated 100x across cluster]\u001b[0m\n",
      "Loss: -452.912:  17%|█▋        | 184/1100 [01:12<05:39,  2.70it/s]\u001b[32m [repeated 98x across cluster]\u001b[0m\n",
      "Loss: 41.4043:   2%|▏         | 23/1100 [00:16<15:15,  1.18it/s]\n",
      "Loss: 41.4043:   2%|▏         | 24/1100 [00:16<16:57,  1.06it/s]\n",
      "Loss: -1039.8481:  78%|███████▊  | 628/800 [02:33<01:00,  2.86it/s]\u001b[32m [repeated 57x across cluster]\u001b[0m\n",
      "Loss: -1109.2146:  81%|████████  | 648/800 [02:38<00:33,  4.59it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "Loss: -1113.7791:  66%|██████▋   | 665/1000 [04:44<01:59,  2.81it/s]\u001b[32m [repeated 100x across cluster]\u001b[0m\n",
      "Loss: -1351.4368:  88%|████████▊ | 966/1100 [09:07<01:04,  2.08it/s]\u001b[32m [repeated 100x across cluster]\u001b[0m\n",
      "Loss: -584.0527:  22%|██▏       | 243/1100 [01:38<06:48,  2.10it/s]\u001b[32m [repeated 92x across cluster]\u001b[0m\n",
      "Loss: -1167.7714:  91%|█████████▏| 730/800 [02:57<00:16,  4.34it/s]\n",
      "Loss: -1174.7596:  91%|█████████▏| 731/800 [02:57<00:16,  4.09it/s]\n",
      "Loss: -1178.3346:  92%|█████████▏| 732/800 [02:57<00:17,  3.98it/s]\n",
      "Loss: -1174.8809:  92%|█████████▏| 733/800 [02:57<00:16,  3.99it/s]\n",
      "Loss: -1186.5944:  92%|█████████▏| 734/800 [02:58<00:17,  3.79it/s]\n",
      "Loss: -1184.6615:  92%|█████████▏| 735/800 [02:58<00:16,  4.00it/s]\n",
      "Loss: -1190.016:  92%|█████████▏| 736/800 [02:58<00:15,  4.02it/s] \n",
      "Loss: -1194.8013:  92%|█████████▏| 737/800 [02:58<00:15,  3.97it/s]\n",
      "Loss: -640.1827:  41%|████      | 286/700 [02:04<03:12,  2.15it/s]\u001b[32m [repeated 75x across cluster]\u001b[0m\n",
      "Loss: -1195.5237:  92%|█████████▏| 738/800 [02:59<00:14,  4.22it/s]\n",
      "Loss: -1191.9299:  92%|█████████▏| 739/800 [02:59<00:14,  4.18it/s]\n",
      "Loss: -1185.4421:  92%|█████████▎| 740/800 [02:59<00:14,  4.27it/s]\n",
      "Loss: -1163.4215:  92%|█████████▎| 740/800 [02:59<00:14,  4.27it/s]\n",
      "Loss: -1163.4215:  93%|█████████▎| 741/800 [02:59<00:14,  4.09it/s]\n",
      "Loss: -1141.2346:  93%|█████████▎| 742/800 [03:00<00:13,  4.27it/s]\n",
      "Loss: -1104.5199:  93%|█████████▎| 743/800 [03:00<00:13,  4.13it/s]\n",
      "Loss: -1050.8533:  93%|█████████▎| 744/800 [03:00<00:13,  4.05it/s]\n",
      "Loss: -1095.3531:  93%|█████████▎| 745/800 [03:00<00:13,  4.12it/s]\n",
      "Loss: -1158.4766:  93%|█████████▎| 746/800 [03:01<00:12,  4.16it/s]\n",
      "Loss: -1158.5374:  93%|█████████▎| 747/800 [03:01<00:12,  4.19it/s]\n",
      "Loss: -1098.4539:  94%|█████████▎| 748/800 [03:01<00:12,  4.32it/s]\n",
      "Loss: -1090.3828:  94%|█████████▎| 749/800 [03:01<00:12,  4.03it/s]\n",
      "Loss: -1126.7347:  94%|█████████▍| 750/800 [03:02<00:12,  3.96it/s]\n",
      "Loss: -1146.1433:  94%|█████████▍| 751/800 [03:02<00:11,  4.18it/s]\n",
      "Loss: -1186.0289:  94%|█████████▍| 752/800 [03:02<00:12,  3.99it/s]\n",
      "Loss: -1204.5275:  94%|█████████▍| 753/800 [03:02<00:12,  3.77it/s]\n",
      "Loss: -1172.7396:  94%|█████████▍| 754/800 [03:03<00:11,  3.86it/s]\n",
      "Loss: -1178.3368:  94%|█████████▍| 755/800 [03:03<00:11,  3.96it/s]\n",
      "Loss: -1219.5287:  94%|█████████▍| 756/800 [03:03<00:10,  4.02it/s]\n",
      "Loss: -1207.4185:  95%|█████████▍| 757/800 [03:03<00:10,  4.14it/s]\n",
      "Loss: -1195.3032:  95%|█████████▍| 758/800 [03:04<00:09,  4.31it/s]\n",
      "Loss: -325.8117:  24%|██▍       | 266/1100 [01:48<06:09,  2.26it/s]\u001b[32m [repeated 66x across cluster]\u001b[0m\n",
      "Loss: -1214.0573:  95%|█████████▍| 759/800 [03:04<00:10,  3.99it/s]\n",
      "Loss: -1222.4098:  95%|█████████▌| 760/800 [03:04<00:09,  4.11it/s]\n",
      "Loss: -1221.3896:  95%|█████████▌| 761/800 [03:04<00:09,  4.31it/s]\n",
      "Loss: -1214.5798:  95%|█████████▌| 762/800 [03:04<00:08,  4.33it/s]\n",
      "Loss: -1221.8486:  95%|█████████▌| 763/800 [03:05<00:09,  4.05it/s]\n",
      "Loss: -1235.5543:  96%|█████████▌| 764/800 [03:05<00:08,  4.11it/s]\n",
      "Loss: -1240.3269:  96%|█████████▌| 765/800 [03:05<00:08,  4.01it/s]\n",
      "Loss: -1234.7821:  96%|█████████▌| 766/800 [03:05<00:08,  4.07it/s]\n",
      "Loss: -1231.886:  96%|█████████▌| 767/800 [03:06<00:08,  3.91it/s] \n",
      "Loss: -1239.2809:  96%|█████████▌| 768/800 [03:06<00:07,  4.19it/s]\n",
      "Loss: -1250.9528:  96%|█████████▌| 769/800 [03:06<00:07,  4.01it/s]\n",
      "Loss: -1256.8521:  96%|█████████▋| 770/800 [03:06<00:07,  4.21it/s]\n",
      "Loss: -1259.6874:  96%|█████████▋| 771/800 [03:07<00:06,  4.42it/s]\n",
      "Loss: -1259.7258:  96%|█████████▋| 772/800 [03:07<00:06,  4.46it/s]\n",
      "Loss: -1258.3126:  97%|█████████▋| 773/800 [03:07<00:06,  4.39it/s]\n",
      "Loss: -1250.0619:  97%|█████████▋| 774/800 [03:07<00:06,  4.24it/s]\n",
      "Loss: -1216.9983:  97%|█████████▋| 775/800 [03:08<00:05,  4.38it/s]\n",
      "Loss: -1121.6819:  97%|█████████▋| 776/800 [03:08<00:05,  4.45it/s]\n",
      "Loss: -888.4838:  97%|█████████▋| 777/800 [03:08<00:05,  4.42it/s] \n",
      "Loss: -540.1924:  97%|█████████▋| 778/800 [03:08<00:05,  4.34it/s]\n",
      "Loss: -267.1939:  97%|█████████▋| 779/800 [03:08<00:04,  4.36it/s]\n",
      "Loss: -615.8498:  98%|█████████▊| 780/800 [03:09<00:04,  4.20it/s]\n",
      "Loss: -573.7299:  25%|██▌       | 279/1100 [01:53<05:36,  2.44it/s]\u001b[32m [repeated 69x across cluster]\u001b[0m\n",
      "Loss: -962.5043:  98%|█████████▊| 781/800 [03:09<00:04,  3.99it/s]\n",
      "Loss: -912.0674:  98%|█████████▊| 782/800 [03:09<00:04,  3.67it/s]\n",
      "Loss: -1022.0427:  98%|█████████▊| 783/800 [03:10<00:04,  3.62it/s]\n",
      "Loss: -1063.9742:  98%|█████████▊| 784/800 [03:10<00:04,  3.53it/s]\n",
      "Loss: -1061.4783:  98%|█████████▊| 785/800 [03:10<00:04,  3.45it/s]\n",
      "Loss: -1106.4841:  98%|█████████▊| 786/800 [03:10<00:03,  3.72it/s]\n",
      "Loss: -1219.2471:  91%|█████████▏| 1004/1100 [09:30<00:57,  1.66it/s]\n",
      "Loss: -1098.8093:  98%|█████████▊| 787/800 [03:11<00:03,  3.90it/s]\n",
      "Loss: -1135.2012:  98%|█████████▊| 788/800 [03:11<00:03,  3.70it/s]\n",
      "Loss: -1373.3005:  92%|█████████▏| 1007/1100 [09:31<00:51,  1.82it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "2024-01-10 21:51:09,187\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', 'cfeef121'), 'time_steps': ('__ref_ph', 'd8e0d04d')}\n",
      "Loss: -640.5028:  42%|████▏     | 466/1100 [05:27<07:12,  1.47it/s]\u001b[32m [repeated 58x across cluster]\u001b[0m\n",
      "Loss: -1191.6733: 100%|██████████| 800/800 [03:14<00:00,  4.12it/s]\n",
      "Loss: -1412.9092:  92%|█████████▏| 1016/1100 [09:36<00:45,  1.85it/s]\u001b[32m [repeated 15x across cluster]\u001b[0m\n",
      "Loss: -673.8444:  47%|████▋     | 331/700 [02:25<02:35,  2.37it/s]\u001b[32m [repeated 64x across cluster]\u001b[0m\n",
      "  0%|          | 0/900 [00:00<?, ?it/s]\n",
      "Loss: 1384.6121:   0%|          | 1/900 [00:00<02:33,  5.84it/s]\n",
      "Loss: -1424.0659:  93%|█████████▎| 1026/1100 [09:41<00:40,  1.83it/s]\u001b[32m [repeated 10x across cluster]\u001b[0m\n",
      "Loss: -1180.9575:  77%|███████▋  | 766/1000 [05:25<01:27,  2.68it/s]\u001b[32m [repeated 69x across cluster]\u001b[0m\n",
      "Loss: -1450.98:  94%|█████████▍| 1036/1100 [09:47<00:37,  1.72it/s]\u001b[32m [repeated 10x across cluster]\u001b[0m\n",
      "Loss: -639.2578:  51%|█████     | 354/700 [02:35<02:40,  2.15it/s]\u001b[32m [repeated 83x across cluster]\u001b[0m\n",
      "Loss: -1302.7224:  95%|█████████▍| 1044/1100 [09:51<00:30,  1.86it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "Loss: -999.2111:  79%|███████▉  | 792/1000 [05:35<01:16,  2.71it/s]\u001b[32m [repeated 84x across cluster]\u001b[0m\n",
      "Loss: -1432.8997:  96%|█████████▌| 1054/1100 [09:56<00:22,  2.06it/s]\u001b[32m [repeated 10x across cluster]\u001b[0m\n",
      "Loss: -248.8745:   8%|▊         | 76/900 [00:17<03:19,  4.13it/s]\u001b[32m [repeated 87x across cluster]\u001b[0m\n",
      "Loss: -1466.3573:  97%|█████████▋| 1063/1100 [10:02<00:22,  1.67it/s]\u001b[32m [repeated 9x across cluster]\u001b[0m\n",
      "Loss: -652.8259:  47%|████▋     | 512/1100 [05:57<07:11,  1.36it/s]\u001b[32m [repeated 74x across cluster]\u001b[0m\n",
      "Loss: -1477.1633:  97%|█████████▋| 1072/1100 [10:07<00:16,  1.72it/s]\u001b[32m [repeated 9x across cluster]\u001b[0m\n",
      "Loss: -346.7863:  13%|█▎        | 118/900 [00:27<03:14,  4.03it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "Loss: -715.7372:  35%|███▍      | 382/1100 [02:35<05:13,  2.29it/s]\n",
      "Loss: -715.7372:  35%|███▍      | 383/1100 [02:35<05:09,  2.32it/s]\n",
      "Loss: -1493.507:  98%|█████████▊| 1080/1100 [10:12<00:14,  1.37it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "Loss: -738.4443:  35%|███▌      | 388/1100 [02:38<08:16,  1.43it/s]\u001b[32m [repeated 58x across cluster]\u001b[0m\n",
      "Loss: -879.6404:  99%|█████████▉| 1087/1100 [10:17<00:08,  1.48it/s]\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "Loss: -782.4172:  36%|███▋      | 400/1100 [02:44<04:41,  2.48it/s]\u001b[32m [repeated 77x across cluster]\u001b[0m\n",
      "Loss: -1392.9968: 100%|█████████▉| 1095/1100 [10:22<00:03,  1.61it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "Loss: -1033.2467:  86%|████████▌ | 861/1000 [06:05<00:53,  2.59it/s]\u001b[32m [repeated 74x across cluster]\u001b[0m\n",
      "Loss: -1439.361: 100%|██████████| 1100/1100 [10:25<00:00,  1.76it/s] \n",
      "2024-01-10 21:52:01,076\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '674f5ac8'), 'epochs': ('__ref_ph', 'a95dbf50'), 'time_steps': ('__ref_ph', '201c1c09')}\n",
      "Loss: -1437.4087: 100%|█████████▉| 1099/1100 [10:24<00:00,  1.69it/s]\u001b[32m [repeated 4x across cluster]\u001b[0m\n",
      "Loss: -288.6313:  16%|█▌        | 173/1100 [01:54<09:07,  1.69it/s]\u001b[32m [repeated 84x across cluster]\u001b[0m\n",
      "  0%|          | 0/1000 [00:00<?, ?it/s]\n",
      "Loss: 1806.3527:   0%|          | 1/1000 [00:00<05:33,  2.99it/s]\n",
      "Loss: 450.9889:   1%|          | 6/1000 [00:01<04:51,  3.41it/s]\u001b[32m [repeated 86x across cluster]\u001b[0m\n",
      "Loss: 25.0529:   2%|▏         | 22/1000 [00:06<05:58,  2.73it/s]\u001b[32m [repeated 91x across cluster]\u001b[0m\n",
      "Loss: -882.5368:  68%|██████▊   | 476/700 [03:31<01:29,  2.51it/s]\u001b[32m [repeated 105x across cluster]\u001b[0m\n",
      "Loss: -1285.239:  91%|█████████▏| 913/1000 [06:26<00:33,  2.63it/s]\n",
      "Loss: -1286.3796:  91%|█████████▏| 914/1000 [06:27<00:31,  2.77it/s]\n",
      "Loss: -1288.3794:  92%|█████████▏| 915/1000 [06:27<00:32,  2.66it/s]\n",
      "Loss: -1288.4253:  92%|█████████▏| 916/1000 [06:28<00:32,  2.56it/s]\n",
      "Loss: -1290.4606:  92%|█████████▏| 917/1000 [06:28<00:31,  2.66it/s]\n",
      "Loss: -1292.1676:  92%|█████████▏| 918/1000 [06:28<00:30,  2.66it/s]\n",
      "Loss: -1292.2384:  92%|█████████▏| 919/1000 [06:29<00:32,  2.52it/s]\n",
      "Loss: -894.2318:  69%|██████▉   | 482/700 [03:34<01:34,  2.31it/s]\n",
      "Loss: -894.2318:  69%|██████▉   | 483/700 [03:34<01:47,  2.02it/s]\n",
      "Loss: -1294.0441:  92%|█████████▏| 920/1000 [06:29<00:31,  2.54it/s]\n",
      "Loss: -1295.4623:  92%|█████████▏| 921/1000 [06:29<00:30,  2.61it/s]\n",
      "Loss: -1295.9153:  92%|█████████▏| 922/1000 [06:30<00:29,  2.68it/s]\n",
      "Loss: -1297.6196:  92%|█████████▏| 923/1000 [06:30<00:29,  2.58it/s]\n",
      "Loss: -1298.5544:  92%|█████████▏| 924/1000 [06:31<00:31,  2.43it/s]\n",
      "Loss: -525.0624:  31%|███       | 280/900 [01:07<02:36,  3.95it/s]\u001b[32m [repeated 85x across cluster]\u001b[0m\n",
      "Loss: -1299.3513:  92%|█████████▎| 925/1000 [06:31<00:30,  2.50it/s]\n",
      "Loss: -1300.984:  93%|█████████▎| 926/1000 [06:31<00:28,  2.59it/s] \n",
      "Loss: -1301.809:  93%|█████████▎| 927/1000 [06:32<00:27,  2.69it/s]\n",
      "Loss: -1302.9391:  93%|█████████▎| 928/1000 [06:32<00:26,  2.74it/s]\n",
      "Loss: -1304.295:  93%|█████████▎| 929/1000 [06:33<00:28,  2.48it/s] \n",
      "Loss: -1304.9784:  93%|█████████▎| 930/1000 [06:33<00:28,  2.48it/s]\n",
      "Loss: -1306.2814:  93%|█████████▎| 931/1000 [06:33<00:26,  2.56it/s]\n",
      "Loss: -1307.4272:  93%|█████████▎| 932/1000 [06:34<00:25,  2.68it/s]\n",
      "Loss: -1308.2443:  93%|█████████▎| 933/1000 [06:34<00:25,  2.63it/s]\n",
      "Loss: -1309.5605:  93%|█████████▎| 934/1000 [06:34<00:23,  2.80it/s]\n",
      "Loss: -1310.5604:  94%|█████████▎| 935/1000 [06:35<00:23,  2.74it/s]\n",
      "Loss: -1311.6676:  94%|█████████▎| 936/1000 [06:35<00:23,  2.70it/s]\n",
      "Loss: -1312.7465:  94%|█████████▎| 937/1000 [06:36<00:23,  2.69it/s]\n",
      "Loss: -921.151:  71%|███████   | 498/700 [03:41<01:26,  2.33it/s] \n",
      "Loss: -654.9067:  34%|███▎      | 302/900 [01:12<02:26,  4.07it/s]\u001b[32m [repeated 91x across cluster]\u001b[0m\n",
      "Loss: -921.151:  71%|███████▏  | 499/700 [03:41<01:29,  2.25it/s]\n",
      "Loss: -1313.6238:  94%|█████████▍| 938/1000 [06:36<00:23,  2.67it/s]\n",
      "Loss: -1314.8708:  94%|█████████▍| 938/1000 [06:36<00:23,  2.67it/s]\n",
      "Loss: -1314.8708:  94%|█████████▍| 939/1000 [06:36<00:25,  2.41it/s]\n",
      "Loss: -1315.8768:  94%|█████████▍| 940/1000 [06:37<00:27,  2.20it/s]\n",
      "Loss: -1316.9551:  94%|█████████▍| 941/1000 [06:37<00:25,  2.29it/s]\n",
      "Loss: -1318.0312:  94%|█████████▍| 942/1000 [06:38<00:24,  2.39it/s]\n",
      "Loss: -1319.0605:  94%|█████████▍| 943/1000 [06:38<00:23,  2.43it/s]\n",
      "Loss: -1320.1445:  94%|█████████▍| 944/1000 [06:39<00:22,  2.51it/s]\n",
      "Loss: -1321.1299:  94%|█████████▍| 945/1000 [06:39<00:21,  2.60it/s]\n",
      "Loss: -1322.2412:  95%|█████████▍| 946/1000 [06:39<00:21,  2.54it/s]\n",
      "Loss: -1323.2809:  95%|█████████▍| 947/1000 [06:40<00:21,  2.49it/s]\n",
      "Loss: -1324.3291:  95%|█████████▍| 948/1000 [06:40<00:20,  2.54it/s]\n",
      "Loss: -1325.3743:  95%|█████████▍| 949/1000 [06:40<00:19,  2.58it/s]\n",
      "Loss: -270.0685:   9%|▉         | 93/1000 [00:27<04:10,  3.62it/s]\u001b[32m [repeated 82x across cluster]\u001b[0m\n",
      "Loss: -1326.4448:  95%|█████████▌| 950/1000 [06:41<00:19,  2.55it/s]\n",
      "Loss: -1327.4677:  95%|█████████▌| 951/1000 [06:41<00:19,  2.53it/s]\n",
      "Loss: -1328.5531:  95%|█████████▌| 952/1000 [06:42<00:17,  2.68it/s]\n",
      "Loss: -1329.6096:  95%|█████████▌| 953/1000 [06:42<00:17,  2.72it/s]\n",
      "Loss: -1330.6428:  95%|█████████▌| 954/1000 [06:42<00:17,  2.67it/s]\n",
      "Loss: -1331.6826:  96%|█████████▌| 955/1000 [06:43<00:17,  2.63it/s]\n",
      "Loss: -1332.7437:  96%|█████████▌| 956/1000 [06:43<00:17,  2.58it/s]\n",
      "Loss: -1333.7958:  96%|█████████▌| 957/1000 [06:43<00:16,  2.64it/s]\n",
      "Loss: -1334.8416:  96%|█████████▌| 958/1000 [06:44<00:16,  2.59it/s]\n",
      "Loss: -1335.925:  96%|█████████▌| 959/1000 [06:44<00:16,  2.54it/s] \n",
      "Loss: -1336.9442:  96%|█████████▌| 960/1000 [06:45<00:15,  2.56it/s]\n",
      "Loss: -1337.9973:  96%|█████████▌| 961/1000 [06:45<00:14,  2.64it/s]\n",
      "Loss: -1339.0603:  96%|█████████▌| 962/1000 [06:45<00:14,  2.54it/s]\n",
      "Loss: -1340.1316:  96%|█████████▋| 963/1000 [06:46<00:14,  2.55it/s]\n",
      "Loss: -951.9021:  75%|███████▍  | 522/700 [03:51<01:13,  2.42it/s]\u001b[32m [repeated 90x across cluster]\u001b[0m\n",
      "Loss: -1341.1689:  96%|█████████▋| 964/1000 [06:46<00:13,  2.59it/s]\n",
      "Loss: -1342.2336:  96%|█████████▋| 965/1000 [06:47<00:12,  2.78it/s]\n",
      "Loss: -1343.287:  97%|█████████▋| 966/1000 [06:47<00:12,  2.65it/s] \n",
      "Loss: -1344.3398:  97%|█████████▋| 967/1000 [06:47<00:12,  2.65it/s]\n",
      "Loss: -1345.4103:  97%|█████████▋| 968/1000 [06:48<00:12,  2.62it/s]\n",
      "Loss: -1346.4694:  97%|█████████▋| 969/1000 [06:48<00:12,  2.53it/s]\n",
      "Loss: -1347.498:  97%|█████████▋| 970/1000 [06:48<00:11,  2.64it/s] \n",
      "Loss: -1348.5902:  97%|█████████▋| 971/1000 [06:49<00:10,  2.66it/s]\n",
      "Loss: -1349.6755:  97%|█████████▋| 972/1000 [06:49<00:10,  2.74it/s]\n",
      "Loss: -1350.7412:  97%|█████████▋| 973/1000 [06:50<00:09,  2.77it/s]\n",
      "Loss: -1351.7898:  97%|█████████▋| 974/1000 [06:50<00:09,  2.81it/s]\n",
      "Loss: -1352.877:  98%|█████████▊| 975/1000 [06:50<00:09,  2.78it/s] \n",
      "Loss: -1353.9363:  98%|█████████▊| 976/1000 [06:51<00:08,  2.75it/s]\n",
      "Loss: -340.3882:  41%|████      | 366/900 [01:28<01:57,  4.53it/s]\u001b[32m [repeated 93x across cluster]\u001b[0m\n",
      "Loss: -1355.0129:  98%|█████████▊| 977/1000 [06:51<00:08,  2.82it/s]\n",
      "Loss: -1356.0947:  98%|█████████▊| 978/1000 [06:51<00:07,  2.76it/s]\n",
      "Loss: -1357.1603:  98%|█████████▊| 979/1000 [06:52<00:07,  2.77it/s]\n",
      "Loss: -1358.2592:  98%|█████████▊| 980/1000 [06:52<00:07,  2.76it/s]\n",
      "Loss: -1359.3395:  98%|█████████▊| 981/1000 [06:52<00:06,  2.74it/s]\n",
      "Loss: -1360.4264:  98%|█████████▊| 982/1000 [06:53<00:06,  2.67it/s]\n",
      "Loss: -1361.5031:  98%|█████████▊| 983/1000 [06:53<00:06,  2.67it/s]\n",
      "Loss: -1362.6046:  98%|█████████▊| 984/1000 [06:54<00:06,  2.59it/s]\n",
      "Loss: -1363.6736:  98%|█████████▊| 985/1000 [06:54<00:05,  2.60it/s]\n",
      "Loss: -1364.7996:  99%|█████████▊| 986/1000 [06:54<00:05,  2.62it/s]\n",
      "Loss: -1365.8813:  99%|█████████▊| 987/1000 [06:55<00:04,  2.69it/s]\n",
      "Loss: -1366.9879:  99%|█████████▉| 988/1000 [06:55<00:04,  2.63it/s]\n",
      "Loss: -1368.0989:  99%|█████████▉| 989/1000 [06:55<00:04,  2.63it/s]\n",
      "Loss: -1369.197:  99%|█████████▉| 990/1000 [06:56<00:03,  2.75it/s] \n",
      "Loss: -375.5147:  15%|█▍        | 148/1000 [00:42<03:55,  3.62it/s]\u001b[32m [repeated 90x across cluster]\u001b[0m\n",
      "Loss: -1370.2988:  99%|█████████▉| 991/1000 [06:56<00:03,  2.77it/s]\n",
      "Loss: -1371.4243:  99%|█████████▉| 992/1000 [06:57<00:03,  2.64it/s]\n",
      "Loss: -1372.5302:  99%|█████████▉| 993/1000 [06:57<00:02,  2.73it/s]\n",
      "Loss: -379.5287:  15%|█▌        | 151/1000 [00:43<03:49,  3.69it/s]\n",
      "Loss: -379.5287:  15%|█▌        | 152/1000 [00:43<03:51,  3.67it/s]\n",
      "Loss: -1373.6591:  99%|█████████▉| 994/1000 [06:57<00:02,  2.69it/s]\n",
      "Loss: -1374.7738: 100%|█████████▉| 995/1000 [06:58<00:01,  2.77it/s]\n",
      "Loss: -1375.9016: 100%|█████████▉| 996/1000 [06:58<00:01,  2.85it/s]\n",
      "Loss: -1377.025: 100%|█████████▉| 997/1000 [06:58<00:01,  2.74it/s] \n",
      "Loss: -1378.1669: 100%|█████████▉| 998/1000 [06:59<00:00,  2.80it/s]\n",
      "Loss: -1379.3063: 100%|█████████▉| 999/1000 [06:59<00:00,  2.70it/s]\n",
      "Loss: -1380.4409: 100%|██████████| 1000/1000 [06:59<00:00,  2.38it/s]\n",
      "2024-01-10 21:52:53,901\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', 'b96bf624'), 'time_steps': ('__ref_ph', '201c1c09')}\n",
      "Loss: -547.7902:  46%|████▌     | 410/900 [01:38<01:48,  4.52it/s]\u001b[32m [repeated 88x across cluster]\u001b[0m\n",
      "Loss: -591.6639:  48%|████▊     | 429/900 [01:43<01:56,  4.03it/s]\u001b[32m [repeated 75x across cluster]\u001b[0m\n",
      "  0%|          | 0/700 [00:00<?, ?it/s]\n",
      "Loss: 2031.9069:   0%|          | 0/700 [00:00<?, ?it/s]\n",
      "Loss: 2031.9069:   0%|          | 1/700 [00:00<03:14,  3.59it/s]\n",
      "Loss: -630.7079:  50%|████▉     | 448/900 [01:48<01:45,  4.29it/s]\u001b[32m [repeated 88x across cluster]\u001b[0m\n",
      "Loss: -512.7585:  22%|██▏       | 215/1000 [01:02<03:23,  3.85it/s]\u001b[32m [repeated 108x across cluster]\u001b[0m\n",
      "Loss: -158.7805:   8%|▊         | 55/700 [00:13<02:32,  4.24it/s]\u001b[32m [repeated 114x across cluster]\u001b[0m\n",
      "Loss: -589.688:  25%|██▌       | 252/1000 [01:12<03:31,  3.53it/s]\u001b[32m [repeated 112x across cluster]\u001b[0m\n",
      "Loss: -274.7718:  14%|█▎        | 96/700 [00:23<02:50,  3.55it/s]\u001b[32m [repeated 106x across cluster]\u001b[0m\n",
      "Loss: -316.9274:  17%|█▋        | 117/700 [00:29<02:22,  4.09it/s]\u001b[32m [repeated 102x across cluster]\u001b[0m\n",
      "Loss: -1034.8159:  91%|█████████▏| 639/700 [04:43<00:27,  2.25it/s]\n",
      "Loss: -1028.8336:  91%|█████████▏| 640/700 [04:44<00:26,  2.30it/s]\n",
      "Loss: -1042.4874:  92%|█████████▏| 641/700 [04:44<00:25,  2.27it/s]\n",
      "Loss: -1043.925:  92%|█████████▏| 642/700 [04:44<00:25,  2.29it/s] \n",
      "Loss: -1043.176:  92%|█████████▏| 643/700 [04:45<00:25,  2.26it/s]\n",
      "Loss: -1061.5126:  92%|█████████▏| 644/700 [04:45<00:24,  2.31it/s]\n",
      "Loss: -1053.6886:  92%|█████████▏| 645/700 [04:46<00:24,  2.26it/s]\n",
      "Loss: -1060.7717:  92%|█████████▏| 646/700 [04:46<00:23,  2.32it/s]\n",
      "Loss: -373.3811:  30%|███       | 304/1000 [01:27<03:18,  3.51it/s]\u001b[32m [repeated 101x across cluster]\u001b[0m\n",
      "Loss: -1064.7698:  92%|█████████▏| 647/700 [04:47<00:22,  2.35it/s]\n",
      "Loss: -1066.0072:  93%|█████████▎| 648/700 [04:47<00:21,  2.39it/s]\n",
      "Loss: -1070.5294:  93%|█████████▎| 649/700 [04:47<00:22,  2.26it/s]\n",
      "Loss: -1071.9213:  93%|█████████▎| 650/700 [04:48<00:21,  2.31it/s]\n",
      "Loss: -1074.896:  93%|█████████▎| 651/700 [04:48<00:21,  2.33it/s] \n",
      "Loss: -1068.74:  93%|█████████▎| 652/700 [04:49<00:22,  2.14it/s] \n",
      "Loss: -1055.924:  93%|█████████▎| 653/700 [04:49<00:21,  2.14it/s]\n",
      "Loss: -1010.5035:  93%|█████████▎| 654/700 [04:50<00:22,  2.06it/s]\n",
      "Loss: -929.927:  94%|█████████▎| 655/700 [04:50<00:20,  2.15it/s]  \n",
      "Loss: -917.1513:  94%|█████████▎| 656/700 [04:51<00:19,  2.24it/s]\n",
      "Loss: -1007.1317:  94%|█████████▍| 657/700 [04:51<00:18,  2.33it/s]\n",
      "Loss: -381.9855:  22%|██▏       | 157/700 [00:39<02:09,  4.21it/s]\u001b[32m [repeated 91x across cluster]\u001b[0m\n",
      "Loss: -1054.5813:  94%|█████████▍| 658/700 [04:51<00:18,  2.32it/s]\n",
      "Loss: -997.7354:  94%|█████████▍| 659/700 [04:52<00:18,  2.27it/s] \n",
      "Loss: -1016.4161:  94%|█████████▍| 660/700 [04:52<00:17,  2.28it/s]\n",
      "Loss: -1041.323:  94%|█████████▍| 661/700 [04:53<00:18,  2.12it/s] \n",
      "Loss: -1015.4967:  95%|█████████▍| 662/700 [04:53<00:17,  2.16it/s]\n",
      "Loss: -546.4476:  33%|███▎      | 327/1000 [01:35<03:32,  3.17it/s]\n",
      "Loss: -546.4476:  33%|███▎      | 328/1000 [01:35<03:38,  3.07it/s]\n",
      "Loss: -1062.0265:  95%|█████████▍| 663/700 [04:54<00:20,  1.79it/s]\n",
      "Loss: -1054.7388:  95%|█████████▍| 664/700 [04:55<00:20,  1.76it/s]\n",
      "Loss: -1056.4557:  95%|█████████▌| 665/700 [04:55<00:18,  1.88it/s]\n",
      "Loss: -1066.9725:  95%|█████████▌| 666/700 [04:56<00:16,  2.04it/s]\n",
      "Loss: -1074.0312:  95%|█████████▌| 667/700 [04:56<00:15,  2.14it/s]\n",
      "Loss: -1073.374:  95%|█████████▌| 668/700 [04:56<00:14,  2.23it/s] \n",
      "Loss: -807.1151:  64%|██████▎   | 699/1100 [08:04<04:39,  1.43it/s]\u001b[32m [repeated 89x across cluster]\u001b[0m\n",
      "Loss: -1086.0333:  96%|█████████▌| 669/700 [04:57<00:14,  2.16it/s]\n",
      "Loss: -1086.0273:  96%|█████████▌| 670/700 [04:57<00:13,  2.23it/s]\n",
      "Loss: -1089.5604:  96%|█████████▌| 671/700 [04:58<00:13,  2.07it/s]\n",
      "Loss: -1085.1129:  96%|█████████▌| 672/700 [04:59<00:15,  1.82it/s]\n",
      "Loss: -1081.8156:  96%|█████████▌| 673/700 [04:59<00:14,  1.82it/s]\n",
      "Loss: -1094.6554:  96%|█████████▋| 674/700 [05:00<00:13,  1.96it/s]\n",
      "Loss: -449.4192:  27%|██▋       | 187/700 [00:47<02:21,  3.63it/s]\n",
      "Loss: -449.4192:  27%|██▋       | 188/700 [00:47<02:18,  3.69it/s]\n",
      "Loss: -1072.3264:  96%|█████████▋| 675/700 [05:00<00:12,  2.08it/s]\n",
      "Loss: -1051.5435:  97%|█████████▋| 676/700 [05:00<00:11,  2.11it/s]\n",
      "Loss: -959.6433:  97%|█████████▋| 677/700 [05:01<00:10,  2.24it/s] \n",
      "Loss: -780.2009:  97%|█████████▋| 678/700 [05:01<00:09,  2.28it/s]\n",
      "Loss: -657.9573:  35%|███▌      | 352/1000 [01:42<03:12,  3.37it/s]\u001b[32m [repeated 91x across cluster]\u001b[0m\n",
      "Loss: -759.4304:  97%|█████████▋| 679/700 [05:02<00:09,  2.25it/s]\n",
      "Loss: -919.3723:  97%|█████████▋| 680/700 [05:02<00:08,  2.31it/s]\n",
      "Loss: -1011.3218:  97%|█████████▋| 681/700 [05:02<00:08,  2.34it/s]\n",
      "Loss: -987.3997:  97%|█████████▋| 682/700 [05:03<00:07,  2.42it/s] \n",
      "Loss: -1031.3253:  98%|█████████▊| 683/700 [05:03<00:06,  2.54it/s]\n",
      "Loss: -1032.1799:  98%|█████████▊| 684/700 [05:04<00:07,  2.10it/s]\n",
      "Loss: -1054.3462:  98%|█████████▊| 685/700 [05:05<00:08,  1.67it/s]\n",
      "Loss: -1058.2605:  98%|█████████▊| 686/700 [05:05<00:07,  1.81it/s]\n",
      "Loss: -1060.2268:  98%|█████████▊| 687/700 [05:06<00:06,  1.92it/s]\n",
      "Loss: -1073.8159:  98%|█████████▊| 688/700 [05:06<00:05,  2.13it/s]\n",
      "Loss: -1079.9598:  98%|█████████▊| 689/700 [05:06<00:04,  2.22it/s]\n",
      "Loss: -1138.8827:  63%|██████▎   | 698/1100 [04:45<02:38,  2.53it/s]\u001b[32m [repeated 90x across cluster]\u001b[0m\n",
      "Loss: -1075.4672:  99%|█████████▊| 690/700 [05:07<00:04,  2.30it/s]\n",
      "Loss: -1084.6427:  99%|█████████▊| 691/700 [05:07<00:03,  2.28it/s]\n",
      "Loss: -1088.7848:  99%|█████████▉| 692/700 [05:08<00:03,  2.20it/s]\n",
      "Loss: -1091.8354:  99%|█████████▉| 693/700 [05:08<00:03,  2.21it/s]\n",
      "Loss: -1093.0969:  99%|█████████▉| 694/700 [05:09<00:02,  2.28it/s]\n",
      "Loss: -1092.9888:  99%|█████████▉| 695/700 [05:09<00:02,  2.35it/s]\n",
      "Loss: -1095.0315:  99%|█████████▉| 696/700 [05:09<00:01,  2.32it/s]\n",
      "Loss: -1095.3492: 100%|█████████▉| 697/700 [05:10<00:01,  2.42it/s]\n",
      "Loss: -1103.1371: 100%|█████████▉| 698/700 [05:10<00:00,  2.49it/s]\n",
      "Loss: -1101.5582: 100%|█████████▉| 699/700 [05:11<00:00,  2.22it/s]\n",
      "Loss: -1102.4719: 100%|██████████| 700/700 [05:11<00:00,  2.25it/s]\n",
      "2024-01-10 21:54:00,656\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', '69f4b3a6'), 'time_steps': ('__ref_ph', '56612a6d')}\n",
      "Loss: -973.9708:  76%|███████▌  | 683/900 [02:43<00:58,  3.68it/s]\u001b[32m [repeated 95x across cluster]\u001b[0m\n",
      "Loss: -1194.8414:  66%|██████▌   | 723/1100 [04:55<02:24,  2.61it/s]\u001b[32m [repeated 95x across cluster]\u001b[0m\n",
      "  0%|          | 0/1100 [00:00<?, ?it/s]\n",
      "Loss: 7419.501:   0%|          | 1/1100 [00:00<15:54,  1.15it/s]\n",
      "Loss: -1142.2008:  67%|██████▋   | 735/1100 [05:00<02:36,  2.34it/s]\u001b[32m [repeated 93x across cluster]\u001b[0m\n",
      "Loss: -890.4011:  68%|██████▊   | 676/1000 [07:22<03:36,  1.50it/s]\n",
      "Loss: -890.4011:  68%|██████▊   | 677/1000 [07:22<03:26,  1.56it/s]\n",
      "Loss: -1180.5653:  68%|██████▊   | 747/1100 [05:05<02:26,  2.41it/s]\u001b[32m [repeated 101x across cluster]\u001b[0m\n",
      "Loss: -794.7762:  46%|████▌     | 457/1000 [02:13<02:26,  3.72it/s]\u001b[32m [repeated 110x across cluster]\u001b[0m\n",
      "Loss: -528.5548:  48%|████▊     | 339/700 [01:24<01:18,  4.63it/s]\u001b[32m [repeated 109x across cluster]\u001b[0m\n",
      "Loss: -1120.7754:  71%|███████▏  | 786/1100 [05:20<01:54,  2.74it/s]\u001b[32m [repeated 101x across cluster]\u001b[0m\n",
      "Loss: -1060.1722:  91%|█████████▏| 822/900 [03:16<00:17,  4.34it/s]\n",
      "Loss: -1089.9474:  91%|█████████▏| 823/900 [03:17<00:18,  4.19it/s]\n",
      "Loss: -1073.3008:  92%|█████████▏| 824/900 [03:17<00:17,  4.46it/s]\n",
      "Loss: -1076.1984:  92%|█████████▏| 825/900 [03:17<00:17,  4.34it/s]\n",
      "Loss: -1104.6829:  92%|█████████▏| 826/900 [03:17<00:16,  4.49it/s]\n",
      "Loss: -1091.6233:  92%|█████████▏| 827/900 [03:18<00:17,  4.27it/s]\n",
      "Loss: -1088.3528:  92%|█████████▏| 828/900 [03:18<00:18,  3.85it/s]\n",
      "Loss: -1102.5013:  92%|█████████▏| 829/900 [03:18<00:17,  3.95it/s]\n",
      "Loss: -1105.4037:  92%|█████████▏| 830/900 [03:18<00:15,  4.38it/s]\n",
      "Loss: -1105.6531:  92%|█████████▏| 831/900 [03:19<00:15,  4.48it/s]\n",
      "Loss: -1082.9196:  92%|█████████▏| 832/900 [03:19<00:15,  4.51it/s]\n",
      "Loss: -1175.5719:  73%|███████▎  | 799/1100 [05:25<01:54,  2.64it/s]\u001b[32m [repeated 96x across cluster]\u001b[0m\n",
      "Loss: -1093.3486:  93%|█████████▎| 833/900 [03:19<00:14,  4.52it/s]\n",
      "Loss: -1116.354:  93%|█████████▎| 834/900 [03:19<00:14,  4.55it/s] \n",
      "Loss: -1099.985:  93%|█████████▎| 835/900 [03:19<00:14,  4.48it/s]\n",
      "Loss: -1084.7052:  93%|█████████▎| 836/900 [03:20<00:14,  4.39it/s]\n",
      "Loss: -1085.2821:  93%|█████████▎| 837/900 [03:20<00:14,  4.22it/s]\n",
      "Loss: -1055.1873:  93%|█████████▎| 838/900 [03:20<00:17,  3.63it/s]\n",
      "Loss: -996.376:  93%|█████████▎| 839/900 [03:21<00:16,  3.66it/s]  \n",
      "Loss: -953.1869:  93%|█████████▎| 840/900 [03:21<00:16,  3.70it/s]\n",
      "Loss: -956.6487:  93%|█████████▎| 841/900 [03:21<00:15,  3.79it/s]\n",
      "Loss: -1078.7556:  94%|█████████▎| 842/900 [03:21<00:14,  3.97it/s]\n",
      "Loss: -1123.0907:  94%|█████████▎| 843/900 [03:22<00:14,  3.90it/s]\n",
      "Loss: -1062.2246:  94%|█████████▍| 844/900 [03:22<00:13,  4.07it/s]\n",
      "Loss: -1079.3396:  94%|█████████▍| 845/900 [03:22<00:13,  4.05it/s]\n",
      "Loss: -1105.1123:  94%|█████████▍| 846/900 [03:22<00:12,  4.26it/s]\n",
      "Loss: -1086.3098:  94%|█████████▍| 847/900 [03:23<00:13,  4.04it/s]\n",
      "Loss: -1103.4502:  94%|█████████▍| 848/900 [03:23<00:12,  4.20it/s]\n",
      "Loss: -1123.3599:  94%|█████████▍| 849/900 [03:23<00:11,  4.36it/s]\n",
      "Loss: -1118.9703:  94%|█████████▍| 850/900 [03:23<00:11,  4.17it/s]\n",
      "Loss: -1121.5721:  95%|█████████▍| 851/900 [03:23<00:11,  4.27it/s]\n",
      "Loss: -1125.2007:  95%|█████████▍| 852/900 [03:24<00:11,  4.15it/s]\n",
      "Loss: -1130.4421:  95%|█████████▍| 853/900 [03:24<00:10,  4.39it/s]\n",
      "Loss: -794.075:  72%|███████▏  | 719/1000 [07:48<03:04,  1.52it/s]\u001b[32m [repeated 81x across cluster]\u001b[0m\n",
      "Loss: -1129.6755:  95%|█████████▍| 854/900 [03:24<00:10,  4.19it/s]\n",
      "Loss: -1131.1907:  95%|█████████▌| 855/900 [03:24<00:10,  4.15it/s]\n",
      "Loss: -1125.4977:  95%|█████████▌| 856/900 [03:25<00:10,  4.19it/s]\n",
      "Loss: -1122.281:  95%|█████████▌| 857/900 [03:25<00:09,  4.33it/s] \n",
      "Loss: -1124.5961:  95%|█████████▌| 858/900 [03:25<00:09,  4.28it/s]\n",
      "Loss: -1119.9336:  95%|█████████▌| 859/900 [03:25<00:09,  4.55it/s]\n",
      "Loss: -1121.2032:  96%|█████████▌| 860/900 [03:25<00:08,  4.53it/s]\n",
      "Loss: -1119.5862:  96%|█████████▌| 861/900 [03:26<00:10,  3.88it/s]\n",
      "Loss: -1117.0206:  96%|█████████▌| 862/900 [03:26<00:09,  4.01it/s]\n",
      "Loss: -1137.6075:  96%|█████████▌| 863/900 [03:26<00:09,  3.94it/s]\n",
      "Loss: -1138.4409:  96%|█████████▌| 864/900 [03:27<00:08,  4.04it/s]\n",
      "Loss: -1139.7725:  96%|█████████▌| 865/900 [03:27<00:08,  4.15it/s]\n",
      "Loss: -1134.8123:  96%|█████████▌| 866/900 [03:27<00:07,  4.27it/s]\n",
      "Loss: -1105.5178:  96%|█████████▋| 867/900 [03:27<00:07,  4.55it/s]\n",
      "Loss: -1052.2585:  96%|█████████▋| 868/900 [03:27<00:07,  4.55it/s]\n",
      "Loss: -979.6086:  97%|█████████▋| 869/900 [03:28<00:07,  4.23it/s] \n",
      "Loss: -901.3831:  97%|█████████▋| 870/900 [03:28<00:07,  3.91it/s]\n",
      "Loss: -871.2549:  97%|█████████▋| 871/900 [03:28<00:07,  3.98it/s]\n",
      "Loss: -978.5294:  97%|█████████▋| 872/900 [03:28<00:06,  4.16it/s]\n",
      "Loss: -1066.0065:  97%|█████████▋| 873/900 [03:29<00:06,  4.20it/s]\n",
      "Loss: -1067.4685:  97%|█████████▋| 874/900 [03:29<00:06,  4.16it/s]\n",
      "Loss: -791.0132:  60%|██████    | 420/700 [01:45<01:20,  3.49it/s]\u001b[32m [repeated 77x across cluster]\u001b[0m\n",
      "Loss: -1073.1116:  97%|█████████▋| 875/900 [03:29<00:05,  4.22it/s]\n",
      "Loss: -1101.7303:  97%|█████████▋| 876/900 [03:29<00:05,  4.03it/s]\n",
      "Loss: -1076.6311:  97%|█████████▋| 877/900 [03:30<00:06,  3.80it/s]\n",
      "Loss: -1127.3118:  98%|█████████▊| 878/900 [03:30<00:05,  3.99it/s]\n",
      "Loss: -1115.7917:  98%|█████████▊| 879/900 [03:30<00:05,  3.78it/s]\n",
      "Loss: -1110.7731:  98%|█████████▊| 880/900 [03:30<00:05,  3.90it/s]\n",
      "Loss: -1146.4188:  98%|█████████▊| 881/900 [03:31<00:05,  3.50it/s]\n",
      "Loss: -1122.801:  98%|█████████▊| 882/900 [03:31<00:05,  3.43it/s] \n",
      "Loss: -1138.3135:  98%|█████████▊| 883/900 [03:31<00:04,  3.61it/s]\n",
      "Loss: -1150.0543:  98%|█████████▊| 884/900 [03:32<00:04,  3.79it/s]\n",
      "Loss: -1142.928:  98%|█████████▊| 885/900 [03:32<00:03,  3.79it/s] \n",
      "Loss: -1151.2419:  98%|█████████▊| 886/900 [03:32<00:03,  3.92it/s]\n",
      "Loss: -1161.064:  99%|█████████▊| 887/900 [03:32<00:03,  3.96it/s] \n",
      "Loss: -1158.5455:  99%|█████████▊| 888/900 [03:33<00:02,  4.08it/s]\n",
      "Loss: -1160.609:  99%|█████████▉| 889/900 [03:33<00:02,  4.02it/s] \n",
      "Loss: -1168.9357:  99%|█████████▉| 890/900 [03:33<00:02,  4.08it/s]\n",
      "Loss: -1170.3195:  99%|█████████▉| 891/900 [03:33<00:02,  4.04it/s]\n",
      "Loss: -1175.7139:  99%|█████████▉| 892/900 [03:34<00:02,  3.93it/s]\n",
      "Loss: -1175.5012:  99%|█████████▉| 893/900 [03:34<00:01,  3.53it/s]\n",
      "Loss: -910.0589:  56%|█████▌    | 561/1000 [02:43<02:22,  3.09it/s]\u001b[32m [repeated 76x across cluster]\u001b[0m\n",
      "Loss: -1181.3937:  99%|█████████▉| 894/900 [03:34<00:01,  3.75it/s]\n",
      "Loss: -1185.744:  99%|█████████▉| 895/900 [03:34<00:01,  3.96it/s] \n",
      "Loss: -1182.7476: 100%|█████████▉| 896/900 [03:35<00:00,  4.16it/s]\n",
      "Loss: -1181.5768: 100%|█████████▉| 897/900 [03:35<00:00,  3.76it/s]\n",
      "Loss: -1166.4657: 100%|█████████▉| 898/900 [03:35<00:00,  4.03it/s]\n",
      "Loss: -1064.7235: 100%|█████████▉| 899/900 [03:35<00:00,  4.02it/s]\n",
      "Loss: -719.7444: 100%|██████████| 900/900 [03:36<00:00,  4.16it/s] \n",
      "2024-01-10 21:54:53,233\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', 'c733eb3d'), 'time_steps': ('__ref_ph', 'd8e0d04d')}\n",
      "Loss: -928.9756:  58%|█████▊    | 579/1000 [02:48<01:50,  3.80it/s]\u001b[32m [repeated 87x across cluster]\u001b[0m\n",
      "  0%|          | 0/1100 [00:00<?, ?it/s]\n",
      "Loss: 2230.6892:   0%|          | 1/1100 [00:00<08:47,  2.08it/s]\n",
      "Loss: -1170.5657:  78%|███████▊  | 860/1100 [05:51<01:50,  2.17it/s]\u001b[32m [repeated 82x across cluster]\u001b[0m\n",
      "Loss: -937.7681:  72%|███████▏  | 502/700 [02:05<00:44,  4.50it/s]\u001b[32m [repeated 96x across cluster]\u001b[0m\n",
      "Loss: -858.2332:  63%|██████▎   | 630/1000 [03:03<01:39,  3.73it/s]\u001b[32m [repeated 83x across cluster]\u001b[0m\n",
      "Loss: -896.2795:  75%|███████▌  | 826/1100 [09:33<03:44,  1.22it/s]\n",
      "Loss: -896.2795:  75%|███████▌  | 827/1100 [09:33<03:44,  1.21it/s]\n",
      "Loss: -1007.951:  77%|███████▋  | 536/700 [02:15<00:50,  3.27it/s]\u001b[32m [repeated 73x across cluster]\u001b[0m\n",
      "Loss: -963.7856:  79%|███████▉  | 552/700 [02:20<00:36,  4.01it/s]\u001b[32m [repeated 72x across cluster]\u001b[0m\n",
      "Loss: -951.4922:  68%|██████▊   | 676/1000 [03:19<01:23,  3.87it/s]\u001b[32m [repeated 96x across cluster]\u001b[0m\n",
      "Loss: -213.3959:  12%|█▏        | 137/1100 [01:23<07:59,  2.01it/s]\u001b[32m [repeated 104x across cluster]\u001b[0m\n",
      "Loss: -1023.0959:  88%|████████▊ | 615/700 [02:35<00:20,  4.21it/s]\u001b[32m [repeated 98x across cluster]\u001b[0m\n",
      "Loss: -220.2249:   8%|▊         | 84/1100 [00:37<07:56,  2.13it/s]\n",
      "Loss: -220.2249:   8%|▊         | 85/1100 [00:37<07:31,  2.25it/s]\n",
      "Loss: -1091.2272:  91%|█████████ | 634/700 [02:40<00:15,  4.18it/s]\u001b[32m [repeated 88x across cluster]\u001b[0m\n",
      "Loss: -1105.9163:  91%|█████████▏| 639/700 [02:41<00:14,  4.24it/s]\n",
      "Loss: -998.8828:  72%|███████▎  | 725/1000 [03:32<01:20,  3.42it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
      "Loss: -1114.1705:  91%|█████████▏| 640/700 [02:42<00:15,  3.78it/s]\n",
      "Loss: -1118.1598:  92%|█████████▏| 641/700 [02:42<00:17,  3.33it/s]\n",
      "Loss: -1118.1659:  92%|█████████▏| 642/700 [02:42<00:17,  3.30it/s]\n",
      "Loss: -1122.0333:  92%|█████████▏| 643/700 [02:43<00:16,  3.44it/s]\n",
      "Loss: -1129.913:  92%|█████████▏| 644/700 [02:43<00:15,  3.57it/s] \n",
      "Loss: -1132.0679:  92%|█████████▏| 645/700 [02:43<00:15,  3.60it/s]\n",
      "Loss: -1131.0027:  92%|█████████▏| 646/700 [02:43<00:14,  3.71it/s]\n",
      "Loss: -1130.6019:  92%|█████████▏| 647/700 [02:44<00:14,  3.65it/s]\n",
      "Loss: -1128.1851:  93%|█████████▎| 648/700 [02:44<00:13,  3.72it/s]\n",
      "Loss: -1123.4956:  93%|█████████▎| 649/700 [02:44<00:13,  3.81it/s]\n",
      "Loss: -1117.682:  93%|█████████▎| 650/700 [02:44<00:13,  3.85it/s] \n",
      "Loss: -1091.1555:  93%|█████████▎| 651/700 [02:45<00:11,  4.10it/s]\n",
      "Loss: -1048.1605:  93%|█████████▎| 652/700 [02:45<00:12,  3.82it/s]\n",
      "Loss: -954.5124:  93%|█████████▎| 653/700 [02:45<00:12,  3.81it/s] \n",
      "Loss: -977.0108:  75%|███████▍  | 746/1000 [03:39<01:17,  3.26it/s]\u001b[32m [repeated 72x across cluster]\u001b[0m\n",
      "Loss: -931.1792:  93%|█████████▎| 654/700 [02:45<00:11,  3.88it/s]\n",
      "Loss: -931.8812:  94%|█████████▎| 655/700 [02:46<00:11,  3.82it/s]\n",
      "Loss: -986.7021:  94%|█████████▎| 656/700 [02:46<00:12,  3.49it/s]\n",
      "Loss: -860.7828:  94%|█████████▍| 657/700 [02:46<00:12,  3.42it/s]\n",
      "Loss: -760.8436:  94%|█████████▍| 658/700 [02:47<00:13,  3.19it/s]\n",
      "Loss: -940.2763:  94%|█████████▍| 659/700 [02:47<00:13,  3.14it/s]\n",
      "Loss: -1068.4972:  94%|█████████▍| 660/700 [02:47<00:11,  3.34it/s]\n",
      "Loss: -985.4308:  94%|█████████▍| 661/700 [02:48<00:11,  3.49it/s] \n",
      "Loss: -1041.9667:  95%|█████████▍| 662/700 [02:48<00:11,  3.43it/s]\n",
      "Loss: -1049.3666:  95%|█████████▍| 663/700 [02:48<00:10,  3.66it/s]\n",
      "Loss: -1059.8107:  95%|█████████▍| 664/700 [02:48<00:10,  3.51it/s]\n",
      "Loss: -1072.4215:  95%|█████████▌| 665/700 [02:49<00:09,  3.64it/s]\n",
      "Loss: -1067.1442:  95%|█████████▌| 666/700 [02:49<00:08,  3.85it/s]\n",
      "Loss: -1078.0634:  95%|█████████▌| 667/700 [02:49<00:08,  4.02it/s]\n",
      "Loss: -1085.1429:  95%|█████████▌| 668/700 [02:49<00:07,  4.06it/s]\n",
      "Loss: -1090.816:  96%|█████████▌| 669/700 [02:50<00:07,  3.93it/s] \n",
      "Loss: -1102.3124:  96%|█████████▌| 670/700 [02:50<00:07,  4.02it/s]\n",
      "Loss: -1099.8661:  96%|█████████▌| 671/700 [02:50<00:07,  4.10it/s]\n",
      "Loss: -933.054:  80%|████████  | 880/1100 [10:10<02:47,  1.32it/s]\u001b[32m [repeated 67x across cluster]\u001b[0m\n",
      "Loss: -1113.2697:  96%|█████████▌| 672/700 [02:50<00:08,  3.50it/s]\n",
      "Loss: -1111.5292:  96%|█████████▌| 673/700 [02:51<00:07,  3.46it/s]\n",
      "Loss: -1119.4696:  96%|█████████▋| 674/700 [02:51<00:07,  3.53it/s]\n",
      "Loss: -1123.1786:  96%|█████████▋| 675/700 [02:51<00:08,  2.89it/s]\n",
      "Loss: -1124.9194:  97%|█████████▋| 676/700 [02:52<00:07,  3.12it/s]\n",
      "Loss: -1133.6075:  97%|█████████▋| 677/700 [02:52<00:08,  2.71it/s]\n",
      "Loss: -1137.5394:  97%|█████████▋| 678/700 [02:53<00:07,  2.91it/s]\n",
      "Loss: -1136.7173:  97%|█████████▋| 679/700 [02:53<00:06,  3.06it/s]\n",
      "Loss: -1147.7162:  97%|█████████▋| 680/700 [02:53<00:06,  3.27it/s]\n",
      "Loss: -1144.1532:  97%|█████████▋| 681/700 [02:53<00:05,  3.42it/s]\n",
      "Loss: -1150.8446:  97%|█████████▋| 682/700 [02:54<00:04,  3.68it/s]\n",
      "Loss: -1155.9797:  98%|█████████▊| 683/700 [02:54<00:04,  3.87it/s]\n",
      "Loss: -1158.7745:  98%|█████████▊| 684/700 [02:54<00:04,  3.95it/s]\n",
      "Loss: -1162.6462:  98%|█████████▊| 685/700 [02:54<00:03,  4.20it/s]\n",
      "Loss: -1165.3647:  98%|█████████▊| 686/700 [02:54<00:03,  4.17it/s]\n",
      "Loss: -1170.6819:  98%|█████████▊| 687/700 [02:55<00:03,  4.13it/s]\n",
      "Loss: -1174.9404:  98%|█████████▊| 688/700 [02:55<00:03,  3.45it/s]\n",
      "Loss: -1176.4347:  98%|█████████▊| 689/700 [02:55<00:02,  3.73it/s]\n",
      "Loss: -1062.4669:  78%|███████▊  | 778/1000 [03:49<01:04,  3.42it/s]\u001b[32m [repeated 64x across cluster]\u001b[0m\n",
      "Loss: -1178.5959:  99%|█████████▊| 690/700 [02:56<00:02,  3.72it/s]\n",
      "Loss: -1179.0475:  99%|█████████▊| 691/700 [02:56<00:02,  3.87it/s]\n",
      "Loss: -1169.2869:  99%|█████████▉| 692/700 [02:56<00:02,  3.90it/s]\n",
      "Loss: -1131.4399:  99%|█████████▉| 693/700 [02:56<00:01,  4.24it/s]\n",
      "Loss: -1040.5698:  99%|█████████▉| 694/700 [02:57<00:01,  3.79it/s]\n",
      "Loss: -879.1193:  99%|█████████▉| 695/700 [02:57<00:01,  3.71it/s] \n",
      "Loss: -809.7675:  99%|█████████▉| 696/700 [02:57<00:01,  3.66it/s]\n",
      "Loss: -994.4756: 100%|█████████▉| 697/700 [02:57<00:00,  3.83it/s]\n",
      "Loss: -990.1422: 100%|█████████▉| 698/700 [02:58<00:00,  3.89it/s]\n",
      "Loss: -1023.9626: 100%|█████████▉| 699/700 [02:58<00:00,  3.81it/s]\n",
      "Loss: -1023.1342: 100%|██████████| 700/700 [02:58<00:00,  3.92it/s]\n",
      "2024-01-10 21:56:00,393\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', '69f4b3a6'), 'time_steps': ('__ref_ph', '7be63cef')}\n",
      "\u001b[36m(func pid=58670)\u001b[0m /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown\n",
      "\u001b[36m(func pid=58670)\u001b[0m   warnings.warn('resource_tracker: There appear to be %d '\n",
      "Loss: -1069.2256:  80%|███████▉  | 796/1000 [03:54<00:48,  4.17it/s]\u001b[32m [repeated 78x across cluster]\u001b[0m\n",
      "Loss: -1439.9225:  91%|█████████▏| 1004/1100 [06:53<00:29,  3.21it/s]\n",
      "Loss: -1441.0896:  91%|█████████▏| 1005/1100 [06:53<00:31,  3.04it/s]\n",
      "Loss: -1442.2004:  91%|█████████▏| 1006/1100 [06:53<00:31,  3.02it/s]\n",
      "Loss: -1443.3097:  92%|█████████▏| 1007/1100 [06:54<00:31,  2.99it/s]\n",
      "Loss: -1444.2845:  92%|█████████▏| 1008/1100 [06:54<00:29,  3.08it/s]\n",
      "Loss: -1444.8495:  92%|█████████▏| 1009/1100 [06:55<00:32,  2.82it/s]\n",
      "Loss: -1444.5243:  92%|█████████▏| 1010/1100 [06:55<00:30,  2.91it/s]\n",
      "Loss: -1444.0419:  92%|█████████▏| 1011/1100 [06:55<00:30,  2.93it/s]\n",
      "Loss: -1443.3379:  92%|█████████▏| 1012/1100 [06:55<00:29,  3.03it/s]\n",
      "Loss: -1439.6395:  92%|█████████▏| 1013/1100 [06:56<00:28,  3.04it/s]\n",
      "Loss: -1437.9711:  92%|█████████▏| 1014/1100 [06:56<00:27,  3.12it/s]\n",
      "Loss: -1439.4191:  92%|█████████▏| 1015/1100 [06:56<00:26,  3.16it/s]\n",
      "Loss: -1110.3732:  82%|████████▏ | 817/1000 [03:59<00:41,  4.36it/s]\u001b[32m [repeated 79x across cluster]\u001b[0m\n",
      "Loss: -1430.8275:  92%|█████████▏| 1016/1100 [06:57<00:26,  3.22it/s]\n",
      "Loss: -1426.5811:  92%|█████████▏| 1017/1100 [06:57<00:27,  3.02it/s]\n",
      "Loss: -1439.9758:  93%|█████████▎| 1018/1100 [06:57<00:26,  3.08it/s]\n",
      "Loss: -1437.4089:  93%|█████████▎| 1019/1100 [06:58<00:26,  3.07it/s]\n",
      "Loss: -1445.8275:  93%|█████████▎| 1020/1100 [06:58<00:25,  3.11it/s]\n",
      "Loss: -1448.9962:  93%|█████████▎| 1021/1100 [06:58<00:26,  2.99it/s]\n",
      "Loss: -1442.9409:  93%|█████████▎| 1022/1100 [06:59<00:25,  3.01it/s]\n",
      "Loss: -1448.6768:  93%|█████████▎| 1023/1100 [06:59<00:25,  3.07it/s]\n",
      "Loss: -1449.6309:  93%|█████████▎| 1024/1100 [06:59<00:25,  3.02it/s]\n",
      "Loss: -1455.3519:  93%|█████████▎| 1025/1100 [07:00<00:24,  3.08it/s]\n",
      "Loss: -1454.0424:  93%|█████████▎| 1026/1100 [07:00<00:24,  3.05it/s]\n",
      "Loss: -1458.673:  93%|█████████▎| 1027/1100 [07:00<00:23,  3.13it/s] \n",
      "Loss: -1460.1553:  93%|█████████▎| 1028/1100 [07:01<00:23,  3.09it/s]\n",
      "Loss: -1460.8446:  94%|█████████▎| 1029/1100 [07:01<00:22,  3.10it/s]\n",
      "Loss: -1462.998:  94%|█████████▎| 1030/1100 [07:01<00:22,  3.09it/s] \n",
      "Loss: -1024.6234:  86%|████████▌ | 859/1000 [09:20<01:13,  1.93it/s]\u001b[32m [repeated 73x across cluster]\u001b[0m\n",
      "Loss: -1456.9871:  94%|█████████▎| 1031/1100 [07:02<00:23,  2.91it/s]\n",
      "Loss: -1461.4243:  94%|█████████▍| 1032/1100 [07:02<00:23,  2.94it/s]\n",
      "Loss: -1465.4631:  94%|█████████▍| 1033/1100 [07:02<00:22,  3.00it/s]\n",
      "Loss: -1465.8888:  94%|█████████▍| 1034/1100 [07:03<00:25,  2.62it/s]\n",
      "Loss: -1462.6337:  94%|█████████▍| 1035/1100 [07:03<00:23,  2.72it/s]\n",
      "Loss: -1461.4248:  94%|█████████▍| 1036/1100 [07:04<00:23,  2.78it/s]\n",
      "Loss: -1449.3972:  94%|█████████▍| 1037/1100 [07:04<00:22,  2.82it/s]\n",
      "Loss: -1431.8185:  94%|█████████▍| 1038/1100 [07:04<00:21,  2.93it/s]\n",
      "Loss: -1389.3083:  94%|█████████▍| 1039/1100 [07:04<00:20,  2.98it/s]\n",
      "Loss: -1332.2529:  95%|█████████▍| 1040/1100 [07:05<00:19,  3.02it/s]\n",
      "Loss: -1315.3998:  95%|█████████▍| 1041/1100 [07:05<00:20,  2.93it/s]\n",
      "Loss: -1379.6226:  95%|█████████▍| 1042/1100 [07:05<00:19,  3.00it/s]\n",
      "Loss: -1461.7651:  95%|█████████▍| 1043/1100 [07:06<00:19,  2.96it/s]\n",
      "Loss: -1442.931:  95%|█████████▍| 1044/1100 [07:06<00:19,  2.92it/s] \n",
      "Loss: -1411.0543:  95%|█████████▌| 1045/1100 [07:07<00:18,  2.98it/s]\n",
      "Loss: -759.2092:  86%|████████▌ | 858/1000 [04:09<00:34,  4.13it/s]\u001b[32m [repeated 75x across cluster]\u001b[0m\n",
      "Loss: -1432.1156:  95%|█████████▌| 1046/1100 [07:07<00:18,  2.94it/s]\n",
      "Loss: -1454.4071:  95%|█████████▌| 1047/1100 [07:07<00:18,  2.80it/s]\n",
      "Loss: -1448.8191:  95%|█████████▌| 1048/1100 [07:08<00:17,  2.93it/s]\n",
      "Loss: -1455.6063:  95%|█████████▌| 1049/1100 [07:08<00:17,  2.94it/s]\n",
      "Loss: -1462.2079:  95%|█████████▌| 1050/1100 [07:08<00:17,  2.93it/s]\n",
      "Loss: -1454.8037:  96%|█████████▌| 1051/1100 [07:09<00:16,  2.97it/s]\n",
      "Loss: -1467.3193:  96%|█████████▌| 1052/1100 [07:09<00:15,  3.08it/s]\n",
      "Loss: -1475.8188:  96%|█████████▌| 1053/1100 [07:09<00:14,  3.15it/s]\n",
      "Loss: -1469.2542:  96%|█████████▌| 1054/1100 [07:09<00:14,  3.23it/s]\n",
      "Loss: -1471.0566:  96%|█████████▌| 1055/1100 [07:10<00:14,  3.13it/s]\n",
      "Loss: -1478.944:  96%|█████████▌| 1056/1100 [07:10<00:14,  3.13it/s] \n",
      "Loss: -1481.5278:  96%|█████████▌| 1057/1100 [07:10<00:14,  3.01it/s]\n",
      "Loss: -1481.6045:  96%|█████████▌| 1058/1100 [07:11<00:13,  3.03it/s]\n",
      "Loss: -1482.0912:  96%|█████████▋| 1059/1100 [07:11<00:13,  2.96it/s]\n",
      "Loss: -1485.048:  96%|█████████▋| 1060/1100 [07:12<00:13,  2.97it/s] \n",
      "Loss: -836.9877:  88%|████████▊ | 878/1000 [09:30<01:04,  1.88it/s]\u001b[32m [repeated 73x across cluster]\u001b[0m\n",
      "Loss: -1487.5957:  96%|█████████▋| 1061/1100 [07:12<00:13,  2.94it/s]\n",
      "Loss: -1490.8422:  97%|█████████▋| 1062/1100 [07:12<00:13,  2.86it/s]\n",
      "Loss: -1492.2461:  97%|█████████▋| 1063/1100 [07:13<00:12,  2.96it/s]\n",
      "Loss: -1491.6115:  97%|█████████▋| 1064/1100 [07:13<00:12,  2.86it/s]\n",
      "Loss: -1492.0013:  97%|█████████▋| 1065/1100 [07:13<00:12,  2.87it/s]\n",
      "Loss: -1497.0052:  97%|█████████▋| 1066/1100 [07:14<00:11,  2.90it/s]\n",
      "Loss: -1499.3838:  97%|█████████▋| 1067/1100 [07:14<00:11,  2.91it/s]\n",
      "Loss: -1499.7114:  97%|█████████▋| 1068/1100 [07:14<00:10,  3.01it/s]\n",
      "Loss: -1500.9281:  97%|█████████▋| 1069/1100 [07:15<00:10,  3.04it/s]\n",
      "Loss: -1502.8451:  97%|█████████▋| 1070/1100 [07:15<00:12,  2.50it/s]\n",
      "Loss: -1502.8652:  97%|█████████▋| 1071/1100 [07:15<00:11,  2.61it/s]\n",
      "Loss: -1502.7679:  97%|█████████▋| 1072/1100 [07:16<00:11,  2.44it/s]\n",
      "Loss: -1503.267:  98%|█████████▊| 1073/1100 [07:16<00:10,  2.54it/s] \n",
      "Loss: -1501.7122:  98%|█████████▊| 1074/1100 [07:17<00:09,  2.65it/s]\n",
      "Loss: -1028.3478:  89%|████████▊ | 887/1000 [09:35<01:02,  1.81it/s]\u001b[32m [repeated 69x across cluster]\u001b[0m\n",
      "Loss: -1496.9247:  98%|█████████▊| 1075/1100 [07:17<00:09,  2.63it/s]\n",
      "Loss: -1485.5952:  98%|█████████▊| 1076/1100 [07:17<00:08,  2.74it/s]\n",
      "Loss: -1460.6555:  98%|█████████▊| 1077/1100 [07:18<00:08,  2.74it/s]\n",
      "Loss: -1408.3806:  98%|█████████▊| 1078/1100 [07:18<00:08,  2.70it/s]\n",
      "Loss: -1333.0248:  98%|█████████▊| 1079/1100 [07:18<00:07,  2.72it/s]\n",
      "Loss: -1293.1787:  98%|█████████▊| 1080/1100 [07:19<00:07,  2.82it/s]\n",
      "Loss: -1359.0942:  98%|█████████▊| 1081/1100 [07:19<00:06,  2.76it/s]\n",
      "Loss: -1421.668:  98%|█████████▊| 1082/1100 [07:20<00:06,  2.81it/s] \n",
      "Loss: -1400.5701:  98%|█████████▊| 1083/1100 [07:20<00:05,  2.84it/s]\n",
      "Loss: -1098.135:  91%|█████████▏| 913/1000 [04:23<00:21,  4.01it/s] \n",
      "Loss: -1405.1132:  99%|█████████▊| 1084/1100 [07:20<00:05,  2.82it/s]\n",
      "Loss: -1026.1914:  90%|████████▉ | 896/1000 [09:40<00:55,  1.87it/s]\u001b[32m [repeated 64x across cluster]\u001b[0m\n",
      "Loss: -1017.8267:  86%|████████▌ | 948/1100 [10:52<01:27,  1.73it/s]\n",
      "Loss: -1017.8267:  86%|████████▋ | 949/1100 [10:52<01:32,  1.64it/s]\n",
      "Loss: -1119.012:  93%|█████████▎| 932/1000 [04:28<00:17,  3.89it/s]\u001b[32m [repeated 32x across cluster]\u001b[0m\n",
      "Loss: -1519.9113: 100%|██████████| 1100/1100 [07:26<00:00,  2.46it/s]\n",
      "2024-01-10 21:56:37,114\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', 'a95dbf50'), 'time_steps': ('__ref_ph', '201c1c09')}\n",
      "Loss: -554.5134:  22%|██▏       | 239/1100 [01:37<05:02,  2.84it/s]\u001b[32m [repeated 49x across cluster]\u001b[0m\n",
      "Loss: -1141.7882:  95%|█████████▌| 954/1000 [04:33<00:10,  4.56it/s]\u001b[32m [repeated 24x across cluster]\u001b[0m\n",
      "Loss: -1101.9716:  91%|█████████▏| 913/1000 [09:49<00:41,  2.12it/s]\n",
      "Loss: -1046.3303:  88%|████████▊ | 965/1100 [11:01<01:11,  1.88it/s]\u001b[32m [repeated 51x across cluster]\u001b[0m\n",
      "Loss: -1164.1815:  98%|█████████▊| 976/1000 [04:37<00:05,  4.46it/s]\u001b[32m [repeated 30x across cluster]\u001b[0m\n",
      "Loss: -570.8958:  25%|██▍       | 270/1100 [01:47<04:19,  3.20it/s]\u001b[32m [repeated 49x across cluster]\u001b[0m\n",
      "Loss: -1185.0769: 100%|██████████| 1000/1000 [04:43<00:00,  3.53it/s]\n",
      "Loss: -1178.6099: 100%|█████████▉| 999/1000 [04:42<00:00,  4.56it/s]\u001b[32m [repeated 34x across cluster]\u001b[0m\n",
      "2024-01-10 21:56:51,379\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '449a5b42'), 'epochs': ('__ref_ph', 'b96bf624'), 'time_steps': ('__ref_ph', '7b44693e')}\n",
      "\u001b[36m(func pid=58661)\u001b[0m /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown\n",
      "\u001b[36m(func pid=58661)\u001b[0m   warnings.warn('resource_tracker: There appear to be %d '\n",
      "Loss: -410.2927:  27%|██▋       | 292/1100 [02:44<05:16,  2.55it/s]\u001b[32m [repeated 50x across cluster]\u001b[0m\n",
      "Loss: -1176.3253:  94%|█████████▍| 944/1000 [10:03<00:23,  2.35it/s]\u001b[32m [repeated 12x across cluster]\u001b[0m\n",
      "Loss: -631.3756:  28%|██▊       | 306/1100 [01:57<03:35,  3.68it/s]\u001b[32m [repeated 57x across cluster]\u001b[0m\n",
      "Loss: -1054.7158:  91%|█████████▏| 1004/1100 [11:19<00:44,  2.18it/s]\n",
      "Loss: -1204.4711:  96%|█████████▌| 957/1000 [10:08<00:17,  2.45it/s]\u001b[32m [repeated 13x across cluster]\u001b[0m\n",
      "Loss: -508.628:  29%|██▉       | 324/1100 [02:02<03:52,  3.33it/s]\u001b[32m [repeated 51x across cluster]\u001b[0m\n",
      "Loss: -1098.884:  92%|█████████▏| 1016/1100 [11:24<00:37,  2.25it/s]\u001b[32m [repeated 24x across cluster]\u001b[0m\n",
      "Loss: -609.79:  31%|███       | 342/1100 [02:07<03:32,  3.58it/s]\u001b[32m [repeated 41x across cluster]\u001b[0m\n",
      "Loss: -1126.871:  93%|█████████▎| 1027/1100 [11:29<00:32,  2.22it/s]\u001b[32m [repeated 23x across cluster]\u001b[0m\n",
      "Loss: -637.4246:  33%|███▎      | 360/1100 [02:12<03:30,  3.51it/s]\u001b[32m [repeated 43x across cluster]\u001b[0m\n",
      "Loss: -1151.0073:  94%|█████████▍| 1038/1100 [11:34<00:28,  2.21it/s]\u001b[32m [repeated 23x across cluster]\u001b[0m\n",
      "Loss: -983.4208:  63%|██████▎   | 691/1100 [07:07<03:04,  2.21it/s]\u001b[32m [repeated 42x across cluster]\u001b[0m\n",
      "Loss: -1226.7726: 100%|██████████| 1000/1000 [10:26<00:00,  1.60it/s]\n",
      "2024-01-10 21:57:19,710\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '2944c5dd'), 'epochs': ('__ref_ph', 'b96bf624'), 'time_steps': ('__ref_ph', '9af98f97')}\n",
      "\u001b[36m(func pid=58678)\u001b[0m /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown\n",
      "\u001b[36m(func pid=58678)\u001b[0m   warnings.warn('resource_tracker: There appear to be %d '\n",
      "Loss: -1094.7067:  95%|█████████▌| 1049/1100 [11:39<00:21,  2.36it/s]\u001b[32m [repeated 17x across cluster]\u001b[0m\n",
      "Loss: -334.2709:  36%|███▌      | 398/1100 [02:23<02:48,  4.16it/s]\u001b[32m [repeated 47x across cluster]\u001b[0m\n",
      "Loss: -1158.2318:  97%|█████████▋| 1062/1100 [11:44<00:14,  2.57it/s]\u001b[32m [repeated 13x across cluster]\u001b[0m\n",
      "Loss: -719.294:  38%|███▊      | 418/1100 [02:28<02:52,  3.95it/s]\u001b[32m [repeated 48x across cluster]\u001b[0m\n",
      "Loss: -1180.084:  98%|█████████▊| 1075/1100 [11:50<00:09,  2.54it/s]\u001b[32m [repeated 13x across cluster]\u001b[0m\n",
      "Loss: -797.0446:  40%|███▉      | 437/1100 [02:33<03:01,  3.65it/s]\u001b[32m [repeated 45x across cluster]\u001b[0m\n",
      "Loss: -1190.8163:  99%|█████████▊| 1086/1100 [11:54<00:05,  2.43it/s]\u001b[32m [repeated 11x across cluster]\u001b[0m\n",
      "Loss: -573.9714:  38%|███▊      | 415/1100 [03:30<03:54,  2.92it/s]\u001b[32m [repeated 48x across cluster]\u001b[0m\n",
      "Loss: -1114.8981: 100%|█████████▉| 1099/1100 [12:00<00:00,  2.52it/s]\u001b[32m [repeated 13x across cluster]\u001b[0m\n",
      "Loss: -1098.6088: 100%|██████████| 1100/1100 [12:00<00:00,  1.53it/s]\n",
      "2024-01-10 21:57:42,350\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '2944c5dd'), 'epochs': ('__ref_ph', 'a95dbf50'), 'time_steps': ('__ref_ph', '56612a6d')}\n",
      "Loss: -911.9957:  43%|████▎     | 478/1100 [02:43<02:26,  4.24it/s]\u001b[32m [repeated 48x across cluster]\u001b[0m\n",
      "Loss: -888.102:  70%|██████▉   | 769/1100 [07:38<01:58,  2.79it/s]\u001b[32m [repeated 54x across cluster]\u001b[0m\n",
      "Loss: -950.3128:  71%|███████   | 783/1100 [07:43<01:48,  2.93it/s]\u001b[32m [repeated 53x across cluster]\u001b[0m\n",
      "Loss: -1039.9966:  50%|████▉     | 546/1100 [02:58<01:59,  4.62it/s]\u001b[32m [repeated 54x across cluster]\u001b[0m\n",
      "Loss: -1023.5513:  52%|█████▏    | 570/1100 [03:03<01:54,  4.62it/s]\u001b[32m [repeated 56x across cluster]\u001b[0m\n",
      "Loss: -1090.1774:  54%|█████▍    | 593/1100 [03:08<01:47,  4.69it/s]\u001b[32m [repeated 54x across cluster]\u001b[0m\n",
      "Loss: -1022.905:  77%|███████▋  | 842/1100 [08:03<01:26,  2.98it/s]\u001b[32m [repeated 56x across cluster]\u001b[0m\n",
      "Loss: -843.1127:  58%|█████▊    | 640/1100 [03:18<01:38,  4.66it/s]\u001b[32m [repeated 55x across cluster]\u001b[0m\n",
      "Loss: -891.6769:  60%|██████    | 663/1100 [03:24<01:43,  4.23it/s]\u001b[32m [repeated 55x across cluster]\u001b[0m\n",
      "Loss: -1085.1376:  80%|████████  | 885/1100 [08:18<01:14,  2.90it/s]\u001b[32m [repeated 53x across cluster]\u001b[0m\n",
      "Loss: -999.0845:  64%|██████▍   | 707/1100 [03:34<01:31,  4.31it/s]\u001b[32m [repeated 50x across cluster]\u001b[0m\n",
      "Loss: -960.6323:  66%|██████▋   | 730/1100 [03:39<01:19,  4.67it/s]\u001b[32m [repeated 55x across cluster]\u001b[0m\n",
      "Loss: -1026.9862:  68%|██████▊   | 752/1100 [03:44<01:17,  4.49it/s]\u001b[32m [repeated 51x across cluster]\u001b[0m\n",
      "Loss: -984.6339:  70%|███████   | 774/1100 [03:49<01:15,  4.29it/s]\u001b[32m [repeated 52x across cluster]\u001b[0m\n",
      "Loss: -803.8255:  72%|███████▏  | 797/1100 [03:54<01:13,  4.11it/s]\u001b[32m [repeated 55x across cluster]\u001b[0m\n",
      "Loss: -866.2517:  61%|██████▏   | 674/1100 [04:50<02:09,  3.28it/s]\n",
      "Loss: -866.2517:  61%|██████▏   | 675/1100 [04:50<02:09,  3.29it/s]\n",
      "Loss: -1280.6606:  88%|████████▊ | 969/1100 [08:49<00:45,  2.90it/s]\u001b[32m [repeated 50x across cluster]\u001b[0m\n",
      "Loss: -1132.2628:  77%|███████▋  | 842/1100 [04:04<00:55,  4.66it/s]\u001b[32m [repeated 55x across cluster]\u001b[0m\n",
      "Loss: -945.0236:  65%|██████▍   | 712/1100 [05:01<02:15,  2.87it/s]\u001b[32m [repeated 54x across cluster]\u001b[0m\n",
      "Loss: -1331.8918:  91%|█████████▏| 1004/1100 [09:01<00:35,  2.73it/s]\n",
      "Loss: -1333.7178:  91%|█████████▏| 1005/1100 [09:02<00:34,  2.76it/s]\n",
      "Loss: -1331.2751:  91%|█████████▏| 1006/1100 [09:02<00:33,  2.79it/s]\n",
      "Loss: -1329.5763:  92%|█████████▏| 1007/1100 [09:02<00:33,  2.81it/s]\n",
      "Loss: -1324.631:  92%|█████████▏| 1008/1100 [09:03<00:32,  2.82it/s] \n",
      "Loss: -1312.4745:  92%|█████████▏| 1009/1100 [09:03<00:32,  2.84it/s]\n",
      "Loss: -1296.9006:  92%|█████████▏| 1010/1100 [09:03<00:31,  2.85it/s]\n",
      "Loss: -1300.407:  92%|█████████▏| 1011/1100 [09:04<00:31,  2.87it/s] \n",
      "Loss: -1307.2268:  92%|█████████▏| 1012/1100 [09:04<00:31,  2.83it/s]\n",
      "Loss: -1016.5146:  81%|████████  | 887/1100 [04:14<00:48,  4.43it/s]\u001b[32m [repeated 44x across cluster]\u001b[0m\n",
      "Loss: -1320.3989:  92%|█████████▏| 1013/1100 [09:05<00:30,  2.86it/s]\n",
      "Loss: -1337.9569:  92%|█████████▏| 1014/1100 [09:05<00:30,  2.81it/s]\n",
      "Loss: -1347.2753:  92%|█████████▏| 1015/1100 [09:05<00:31,  2.73it/s]\n",
      "Loss: -1338.1138:  92%|█████████▏| 1016/1100 [09:06<00:30,  2.78it/s]\n",
      "Loss: -1340.4875:  92%|█████████▏| 1017/1100 [09:06<00:29,  2.80it/s]\n",
      "Loss: -1344.7788:  93%|█████████▎| 1018/1100 [09:06<00:29,  2.80it/s]\n",
      "Loss: -1356.0072:  93%|█████████▎| 1019/1100 [09:07<00:30,  2.68it/s]\n",
      "Loss: -1353.8776:  93%|█████████▎| 1020/1100 [09:07<00:29,  2.76it/s]\n",
      "Loss: -1351.8955:  93%|█████████▎| 1021/1100 [09:07<00:28,  2.81it/s]\n",
      "Loss: -1350.4701:  93%|█████████▎| 1022/1100 [09:08<00:27,  2.85it/s]\n",
      "Loss: -1355.2812:  93%|█████████▎| 1023/1100 [09:08<00:27,  2.85it/s]\n",
      "Loss: -1363.8467:  93%|█████████▎| 1024/1100 [09:08<00:26,  2.86it/s]\n",
      "Loss: -623.028:  67%|██████▋   | 742/1100 [05:11<01:46,  3.36it/s] \n",
      "Loss: -623.028:  68%|██████▊   | 743/1100 [05:11<01:44,  3.40it/s]\n",
      "Loss: -1365.2758:  93%|█████████▎| 1025/1100 [09:09<00:25,  2.91it/s]\n",
      "Loss: -1365.7465:  93%|█████████▎| 1026/1100 [09:09<00:28,  2.63it/s]\n",
      "Loss: -1210.4937:  83%|████████▎ | 909/1100 [04:19<00:42,  4.47it/s]\u001b[32m [repeated 37x across cluster]\u001b[0m\n",
      "Loss: -1369.2297:  93%|█████████▎| 1027/1100 [09:10<00:27,  2.69it/s]\n",
      "Loss: -1369.6343:  93%|█████████▎| 1028/1100 [09:10<00:26,  2.67it/s]\n",
      "Loss: -1369.3334:  94%|█████████▎| 1029/1100 [09:10<00:27,  2.59it/s]\n",
      "Loss: -1369.1312:  94%|█████████▎| 1030/1100 [09:11<00:26,  2.66it/s]\n",
      "Loss: -1362.5262:  94%|█████████▎| 1031/1100 [09:11<00:25,  2.71it/s]\n",
      "Loss: -1341.3988:  94%|█████████▍| 1032/1100 [09:11<00:24,  2.77it/s]\n",
      "Loss: -1312.2975:  94%|█████████▍| 1033/1100 [09:12<00:23,  2.81it/s]\n",
      "Loss: -1285.8071:  94%|█████████▍| 1034/1100 [09:12<00:23,  2.84it/s]\n",
      "Loss: -1296.7968:  94%|█████████▍| 1035/1100 [09:13<00:23,  2.76it/s]\n",
      "Loss: -1331.2576:  94%|█████████▍| 1036/1100 [09:13<00:23,  2.75it/s]\n",
      "Loss: -1334.4048:  94%|█████████▍| 1037/1100 [09:13<00:23,  2.68it/s]\n",
      "Loss: -1324.7087:  94%|█████████▍| 1038/1100 [09:14<00:22,  2.73it/s]\n",
      "Loss: -1315.1248:  94%|█████████▍| 1039/1100 [09:14<00:22,  2.77it/s]\n",
      "Loss: -1302.8779:  95%|█████████▍| 1040/1100 [09:14<00:21,  2.79it/s]\n",
      "Loss: -1235.9241:  85%|████████▍ | 930/1100 [04:24<00:38,  4.41it/s]\u001b[32m [repeated 37x across cluster]\u001b[0m\n",
      "Loss: -1304.833:  95%|█████████▍| 1041/1100 [09:15<00:20,  2.82it/s] \n",
      "Loss: -1303.2411:  95%|█████████▍| 1042/1100 [09:15<00:20,  2.82it/s]\n",
      "Loss: -1336.2738:  95%|█████████▍| 1043/1100 [09:15<00:19,  2.85it/s]\n",
      "Loss: -1362.2839:  95%|█████████▍| 1044/1100 [09:16<00:19,  2.87it/s]\n",
      "Loss: -1368.2743:  95%|█████████▌| 1045/1100 [09:16<00:19,  2.85it/s]\n",
      "Loss: -1373.3209:  95%|█████████▌| 1046/1100 [09:16<00:18,  2.86it/s]\n",
      "Loss: -1363.838:  95%|█████████▌| 1047/1100 [09:17<00:18,  2.87it/s] \n",
      "Loss: -1363.106:  95%|█████████▌| 1048/1100 [09:17<00:18,  2.87it/s]\n",
      "Loss: -1377.6259:  95%|█████████▌| 1049/1100 [09:17<00:17,  2.86it/s]\n",
      "Loss: -1386.6775:  95%|█████████▌| 1050/1100 [09:18<00:17,  2.87it/s]\n",
      "Loss: -1394.7837:  96%|█████████▌| 1051/1100 [09:18<00:17,  2.87it/s]\n",
      "Loss: -1393.0231:  96%|█████████▌| 1052/1100 [09:19<00:16,  2.83it/s]\n",
      "Loss: -1391.3854:  96%|█████████▌| 1053/1100 [09:19<00:16,  2.84it/s]\n",
      "Loss: -1396.3193:  96%|█████████▌| 1054/1100 [09:19<00:16,  2.85it/s]\n",
      "Loss: -1262.1794:  87%|████████▋ | 954/1100 [04:30<00:31,  4.58it/s]\u001b[32m [repeated 41x across cluster]\u001b[0m\n",
      "Loss: -1401.2455:  96%|█████████▌| 1055/1100 [09:20<00:15,  2.89it/s]\n",
      "Loss: -1403.5846:  96%|█████████▌| 1056/1100 [09:20<00:15,  2.89it/s]\n",
      "Loss: -1401.7371:  96%|█████████▌| 1057/1100 [09:20<00:14,  2.91it/s]\n",
      "Loss: -1402.4086:  96%|█████████▌| 1058/1100 [09:21<00:14,  2.92it/s]\n",
      "Loss: -1398.3676:  96%|█████████▋| 1059/1100 [09:21<00:13,  2.93it/s]\n",
      "Loss: -1396.2782:  96%|█████████▋| 1060/1100 [09:21<00:13,  2.92it/s]\n",
      "Loss: -1394.7216:  96%|█████████▋| 1061/1100 [09:22<00:13,  2.92it/s]\n",
      "Loss: -1388.4025:  97%|█████████▋| 1062/1100 [09:22<00:12,  2.93it/s]\n",
      "Loss: -1370.2711:  97%|█████████▋| 1063/1100 [09:22<00:12,  2.91it/s]\n",
      "Loss: -1339.6348:  97%|█████████▋| 1064/1100 [09:23<00:12,  2.84it/s]\n",
      "Loss: -1300.0536:  97%|█████████▋| 1065/1100 [09:23<00:12,  2.86it/s]\n",
      "Loss: -1278.5934:  97%|█████████▋| 1066/1100 [09:23<00:12,  2.81it/s]\n",
      "Loss: -1292.1273:  97%|█████████▋| 1067/1100 [09:24<00:11,  2.81it/s]\n",
      "Loss: -1326.0067:  97%|█████████▋| 1068/1100 [09:24<00:11,  2.80it/s]\n",
      "Loss: -1324.879:  97%|█████████▋| 1069/1100 [09:24<00:11,  2.78it/s] \n",
      "Loss: -1287.469:  89%|████████▉ | 977/1100 [04:35<00:28,  4.33it/s]\u001b[32m [repeated 40x across cluster]\u001b[0m\n",
      "Loss: -1307.3408:  97%|█████████▋| 1070/1100 [09:25<00:10,  2.81it/s]\n",
      "Loss: -1319.7753:  97%|█████████▋| 1071/1100 [09:25<00:10,  2.85it/s]\n",
      "Loss: -1362.1644:  97%|█████████▋| 1072/1100 [09:25<00:09,  2.87it/s]\n",
      "Loss: -1351.6388:  98%|█████████▊| 1073/1100 [09:26<00:09,  2.79it/s]\n",
      "Loss: -1358.7544:  98%|█████████▊| 1074/1100 [09:26<00:09,  2.81it/s]\n",
      "Loss: -1397.9728:  98%|█████████▊| 1075/1100 [09:27<00:08,  2.84it/s]\n",
      "Loss: -1404.6006:  98%|█████████▊| 1076/1100 [09:27<00:08,  2.69it/s]\n",
      "Loss: -1398.9609:  98%|█████████▊| 1077/1100 [09:27<00:08,  2.75it/s]\n",
      "Loss: -1396.6805:  98%|█████████▊| 1078/1100 [09:28<00:08,  2.74it/s]\n",
      "Loss: -1412.5109:  98%|█████████▊| 1079/1100 [09:28<00:07,  2.75it/s]\n",
      "Loss: -1424.4868:  98%|█████████▊| 1080/1100 [09:28<00:07,  2.81it/s]\n",
      "Loss: -1418.2864:  98%|█████████▊| 1081/1100 [09:29<00:06,  2.85it/s]\n",
      "Loss: -1425.1841:  98%|█████████▊| 1082/1100 [09:29<00:06,  2.83it/s]\n",
      "Loss: -1429.0094:  98%|█████████▊| 1083/1100 [09:29<00:05,  2.84it/s]\n",
      "Loss: -1430.3301:  99%|█████████▊| 1084/1100 [09:30<00:05,  2.83it/s]\n",
      "Loss: -1313.6385:  91%|█████████ | 1000/1100 [04:40<00:22,  4.53it/s]\u001b[32m [repeated 40x across cluster]\u001b[0m\n",
      "Loss: -1434.8137:  99%|█████████▊| 1085/1100 [09:30<00:05,  2.85it/s]\n",
      "Loss: -1440.0516:  99%|█████████▊| 1086/1100 [09:30<00:04,  2.89it/s]\n",
      "Loss: -1318.2241:  91%|█████████▏| 1004/1100 [04:41<00:20,  4.59it/s]\n",
      "Loss: -1320.5311:  92%|█████████▏| 1015/1100 [04:43<00:18,  4.57it/s]\u001b[32m [repeated 18x across cluster]\u001b[0m\n",
      "Loss: -761.9321:  75%|███████▌  | 828/1100 [05:37<01:19,  3.41it/s]\u001b[32m [repeated 20x across cluster]\u001b[0m\n",
      "Loss: -1186.8586: 100%|██████████| 1100/1100 [09:35<00:00,  1.91it/s]\n",
      "2024-01-10 21:59:46,568\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '674f5ac8'), 'epochs': ('__ref_ph', 'a95dbf50'), 'time_steps': ('__ref_ph', '201c1c09')}\n",
      "\u001b[36m(func pid=58681)\u001b[0m /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown\n",
      "\u001b[36m(func pid=58681)\u001b[0m   warnings.warn('resource_tracker: There appear to be %d '\n",
      "Loss: -1346.0065:  94%|█████████▍| 1039/1100 [04:48<00:11,  5.17it/s]\u001b[32m [repeated 30x across cluster]\u001b[0m\n",
      "Loss: -681.0369:  77%|███████▋  | 847/1100 [05:42<01:06,  3.79it/s]\u001b[32m [repeated 19x across cluster]\u001b[0m\n",
      "Loss: -1303.6957:  97%|█████████▋| 1065/1100 [04:53<00:07,  4.93it/s]\u001b[32m [repeated 26x across cluster]\u001b[0m\n",
      "Loss: -1011.0742:  79%|███████▉  | 867/1100 [05:47<01:00,  3.88it/s]\u001b[32m [repeated 20x across cluster]\u001b[0m\n",
      "Loss: -1387.3842:  99%|█████████▉| 1092/1100 [04:58<00:01,  5.08it/s]\u001b[32m [repeated 27x across cluster]\u001b[0m\n",
      "Loss: -1397.3243: 100%|██████████| 1100/1100 [05:00<00:00,  3.66it/s]\n",
      "2024-01-10 22:00:01,001\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', 'ebf8d860'), 'epochs': ('__ref_ph', 'a95dbf50'), 'time_steps': ('__ref_ph', '7b44693e')}\n",
      "\u001b[36m(func pid=58673)\u001b[0m /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown\n",
      "\u001b[36m(func pid=58673)\u001b[0m   warnings.warn('resource_tracker: There appear to be %d '\n",
      "Loss: -757.1154:  81%|████████  | 887/1100 [05:53<00:52,  4.08it/s]\u001b[32m [repeated 20x across cluster]\u001b[0m\n",
      "Loss: -1396.0111: 100%|█████████▉| 1099/1100 [05:00<00:00,  5.21it/s]\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "Loss: -1081.3752:  83%|████████▎ | 909/1100 [05:58<00:45,  4.20it/s]\u001b[32m [repeated 22x across cluster]\u001b[0m\n",
      "Loss: -1128.92:  85%|████████▍ | 931/1100 [06:03<00:39,  4.27it/s]\u001b[32m [repeated 22x across cluster]\u001b[0m\n",
      "Loss: -1176.4904:  87%|████████▋ | 953/1100 [06:08<00:34,  4.21it/s]\u001b[32m [repeated 22x across cluster]\u001b[0m\n",
      "Loss: -1217.6581:  89%|████████▊ | 974/1100 [06:13<00:30,  4.20it/s]\u001b[32m [repeated 21x across cluster]\u001b[0m\n",
      "Loss: -1241.4584:  90%|█████████ | 995/1100 [06:18<00:25,  4.16it/s]\u001b[32m [repeated 21x across cluster]\u001b[0m\n",
      "Loss: -1198.2997:  91%|█████████▏| 1004/1100 [06:20<00:22,  4.20it/s]\n",
      "Loss: -1179.8972:  91%|█████████▏| 1005/1100 [06:21<00:22,  4.24it/s]\n",
      "Loss: -1168.1039:  91%|█████████▏| 1006/1100 [06:21<00:21,  4.27it/s]\n",
      "Loss: -1161.9214:  92%|█████████▏| 1007/1100 [06:21<00:21,  4.28it/s]\n",
      "Loss: -1189.6438:  92%|█████████▏| 1008/1100 [06:21<00:21,  4.33it/s]\n",
      "Loss: -1240.8453:  92%|█████████▏| 1009/1100 [06:22<00:21,  4.29it/s]\n",
      "Loss: -1258.8634:  92%|█████████▏| 1010/1100 [06:22<00:21,  4.27it/s]\n",
      "Loss: -1241.9041:  92%|█████████▏| 1011/1100 [06:22<00:20,  4.30it/s]\n",
      "Loss: -1234.0435:  92%|█████████▏| 1012/1100 [06:22<00:20,  4.33it/s]\n",
      "Loss: -1241.1239:  92%|█████████▏| 1013/1100 [06:22<00:20,  4.29it/s]\n",
      "Loss: -1270.7513:  92%|█████████▏| 1014/1100 [06:23<00:20,  4.29it/s]\n",
      "Loss: -1272.7854:  92%|█████████▏| 1015/1100 [06:23<00:19,  4.31it/s]\n",
      "Loss: -1264.5005:  92%|█████████▏| 1016/1100 [06:23<00:19,  4.33it/s]\n",
      "Loss: -1264.1459:  92%|█████████▏| 1017/1100 [06:23<00:19,  4.34it/s]\n",
      "Loss: -1225.7048:  91%|█████████ | 1003/1100 [06:20<00:23,  4.22it/s]\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "Loss: -1276.7639:  93%|█████████▎| 1018/1100 [06:24<00:19,  4.29it/s]\n",
      "Loss: -1285.4095:  93%|█████████▎| 1019/1100 [06:24<00:18,  4.28it/s]\n",
      "Loss: -1288.0325:  93%|█████████▎| 1020/1100 [06:24<00:18,  4.29it/s]\n",
      "Loss: -1286.7599:  93%|█████████▎| 1021/1100 [06:24<00:18,  4.31it/s]\n",
      "Loss: -1280.8433:  93%|█████████▎| 1022/1100 [06:25<00:18,  4.33it/s]\n",
      "Loss: -1292.2271:  93%|█████████▎| 1023/1100 [06:25<00:17,  4.35it/s]\n",
      "Loss: -1297.2794:  93%|█████████▎| 1024/1100 [06:25<00:17,  4.34it/s]\n",
      "Loss: -1300.3768:  93%|█████████▎| 1025/1100 [06:25<00:17,  4.35it/s]\n",
      "Loss: -1301.6803:  93%|█████████▎| 1026/1100 [06:25<00:17,  4.32it/s]\n",
      "Loss: -1302.042:  93%|█████████▎| 1027/1100 [06:26<00:16,  4.36it/s] \n",
      "Loss: -1302.4912:  93%|█████████▎| 1028/1100 [06:26<00:16,  4.37it/s]\n",
      "Loss: -1303.7205:  94%|█████████▎| 1029/1100 [06:26<00:16,  4.21it/s]\n",
      "Loss: -1302.8798:  94%|█████████▎| 1030/1100 [06:26<00:16,  4.27it/s]\n",
      "Loss: -1300.5958:  94%|█████████▎| 1031/1100 [06:27<00:17,  3.96it/s]\n",
      "Loss: -1295.4752:  94%|█████████▍| 1032/1100 [06:27<00:16,  4.08it/s]\n",
      "Loss: -1292.007:  94%|█████████▍| 1033/1100 [06:27<00:16,  4.17it/s] \n",
      "Loss: -1277.2377:  94%|█████████▍| 1034/1100 [06:27<00:16,  4.07it/s]\n",
      "Loss: -1261.7512:  94%|█████████▍| 1035/1100 [06:28<00:16,  4.04it/s]\n",
      "Loss: -1257.4717:  94%|█████████▍| 1036/1100 [06:28<00:15,  4.11it/s]\n",
      "Loss: -1241.8732:  94%|█████████▍| 1037/1100 [06:28<00:15,  4.13it/s]\n",
      "Loss: -1225.3047:  94%|█████████▍| 1038/1100 [06:28<00:14,  4.20it/s]\n",
      "Loss: -1157.2706:  94%|█████████▍| 1039/1100 [06:29<00:14,  4.14it/s]\n",
      "Loss: -958.1986:  95%|█████████▍| 1040/1100 [06:29<00:14,  4.21it/s] \n",
      "Loss: -874.3219:  95%|█████████▍| 1041/1100 [06:29<00:14,  4.18it/s]\n",
      "Loss: -1049.0958:  95%|█████████▍| 1042/1100 [06:29<00:13,  4.21it/s]\n",
      "Loss: -1282.0947:  95%|█████████▍| 1043/1100 [06:30<00:13,  4.25it/s]\n",
      "Loss: -1146.1793:  95%|█████████▍| 1044/1100 [06:30<00:13,  4.30it/s]\n",
      "Loss: -1160.368:  95%|█████████▌| 1045/1100 [06:30<00:12,  4.32it/s] \n",
      "Loss: -1274.5555:  95%|█████████▌| 1046/1100 [06:30<00:12,  4.25it/s]\n",
      "Loss: -1198.076:  95%|█████████▌| 1047/1100 [06:31<00:12,  4.18it/s] \n",
      "Loss: -1254.9106:  95%|█████████▌| 1048/1100 [06:31<00:12,  4.17it/s]\n",
      "Loss: -1272.0365:  95%|█████████▌| 1049/1100 [06:31<00:12,  4.18it/s]\n",
      "Loss: -1245.6779:  95%|█████████▌| 1050/1100 [06:31<00:11,  4.18it/s]\n",
      "Loss: -1295.5443:  96%|█████████▌| 1051/1100 [06:31<00:11,  4.17it/s]\n",
      "Loss: -1267.2079:  96%|█████████▌| 1052/1100 [06:32<00:11,  4.20it/s]\n",
      "Loss: -1299.6389:  96%|█████████▌| 1053/1100 [06:32<00:11,  4.26it/s]\n",
      "Loss: -1290.2784:  96%|█████████▌| 1054/1100 [06:32<00:10,  4.20it/s]\n",
      "Loss: -1300.3557:  96%|█████████▌| 1055/1100 [06:32<00:10,  4.20it/s]\n",
      "Loss: -1305.8081:  96%|█████████▌| 1056/1100 [06:33<00:10,  4.19it/s]\n",
      "Loss: -1306.238:  96%|█████████▌| 1057/1100 [06:33<00:10,  4.19it/s] \n",
      "Loss: -1315.9764:  96%|█████████▌| 1058/1100 [06:33<00:10,  4.12it/s]\n",
      "Loss: -1312.46:  96%|█████████▋| 1059/1100 [06:33<00:09,  4.16it/s]  \n",
      "Loss: -1322.9906:  96%|█████████▋| 1060/1100 [06:34<00:09,  4.22it/s]\n",
      "Loss: -1320.1614:  96%|█████████▋| 1061/1100 [06:34<00:09,  4.23it/s]\n",
      "Loss: -1327.1158:  97%|█████████▋| 1062/1100 [06:34<00:09,  4.19it/s]\n",
      "Loss: -1328.4683:  97%|█████████▋| 1063/1100 [06:34<00:08,  4.23it/s]\n",
      "Loss: -1333.0204:  97%|█████████▋| 1064/1100 [06:35<00:08,  4.29it/s]\n",
      "Loss: -1334.2161:  97%|█████████▋| 1065/1100 [06:35<00:08,  4.31it/s]\n",
      "Loss: -1333.8064:  97%|█████████▋| 1066/1100 [06:35<00:07,  4.31it/s]\n",
      "Loss: -1330.7147:  97%|█████████▋| 1067/1100 [06:35<00:07,  4.32it/s]\n",
      "Loss: -1308.782:  97%|█████████▋| 1068/1100 [06:35<00:07,  4.29it/s] \n",
      "Loss: -1274.1582:  97%|█████████▋| 1069/1100 [06:36<00:07,  4.29it/s]\n",
      "Loss: -1227.219:  97%|█████████▋| 1070/1100 [06:36<00:06,  4.31it/s] \n",
      "Loss: -1161.4663:  97%|█████████▋| 1071/1100 [06:36<00:06,  4.27it/s]\n",
      "Loss: -1115.3973:  97%|█████████▋| 1072/1100 [06:36<00:06,  4.27it/s]\n",
      "Loss: -1251.7111:  98%|█████████▊| 1073/1100 [06:37<00:06,  4.28it/s]\n",
      "Loss: -1280.5:  98%|█████████▊| 1074/1100 [06:37<00:06,  4.25it/s]   \n",
      "Loss: -1283.8066:  98%|█████████▊| 1075/1100 [06:37<00:05,  4.30it/s]\n",
      "Loss: -1270.7845:  98%|█████████▊| 1076/1100 [06:37<00:05,  4.23it/s]\n",
      "Loss: -1315.957:  98%|█████████▊| 1077/1100 [06:38<00:05,  4.24it/s] \n",
      "Loss: -1311.2448:  98%|█████████▊| 1078/1100 [06:38<00:05,  4.27it/s]\n",
      "Loss: -1313.3441:  98%|█████████▊| 1079/1100 [06:38<00:04,  4.32it/s]\n",
      "Loss: -1327.9142:  98%|█████████▊| 1080/1100 [06:38<00:04,  4.30it/s]\n",
      "Loss: -1333.0645:  98%|█████████▊| 1081/1100 [06:39<00:04,  4.27it/s]\n",
      "Loss: -1331.204:  98%|█████████▊| 1082/1100 [06:39<00:04,  4.29it/s] \n",
      "Loss: -1343.939:  98%|█████████▊| 1083/1100 [06:39<00:03,  4.29it/s]\n",
      "Loss: -1346.1842:  99%|█████████▊| 1084/1100 [06:39<00:03,  4.30it/s]\n",
      "Loss: -1344.6288:  99%|█████████▊| 1085/1100 [06:39<00:03,  4.32it/s]\n",
      "Loss: -1355.9946:  99%|█████████▊| 1086/1100 [06:40<00:03,  4.32it/s]\n",
      "Loss: -1354.5732:  99%|█████████▉| 1087/1100 [06:40<00:02,  4.34it/s]\n",
      "Loss: -1359.5649:  99%|█████████▉| 1088/1100 [06:40<00:02,  4.29it/s]\n",
      "Loss: -1362.4618:  99%|█████████▉| 1089/1100 [06:40<00:02,  4.26it/s]\n",
      "Loss: -1364.2865:  99%|█████████▉| 1090/1100 [06:41<00:02,  4.22it/s]\n",
      "Loss: -1370.2869:  99%|█████████▉| 1091/1100 [06:41<00:02,  4.22it/s]\n",
      "Loss: -1369.8402:  99%|█████████▉| 1092/1100 [06:41<00:01,  4.26it/s]\n",
      "Loss: -1370.1293:  99%|█████████▉| 1093/1100 [06:41<00:01,  4.26it/s]\n",
      "Loss: -1376.9597:  99%|█████████▉| 1094/1100 [06:42<00:01,  4.23it/s]\n",
      "Loss: -1378.588: 100%|█████████▉| 1095/1100 [06:42<00:01,  4.24it/s] \n",
      "Loss: -1377.8291: 100%|█████████▉| 1096/1100 [06:42<00:00,  4.28it/s]\n",
      "Loss: -1377.1149: 100%|█████████▉| 1097/1100 [06:42<00:00,  4.29it/s]\n",
      "Loss: -1378.9128: 100%|█████████▉| 1098/1100 [06:43<00:00,  4.26it/s]\n",
      "Loss: -1384.7375: 100%|█████████▉| 1099/1100 [06:43<00:00,  4.24it/s]\n",
      "Loss: -1383.7296: 100%|██████████| 1100/1100 [06:43<00:00,  2.73it/s]\n",
      "2024-01-10 22:00:51,958\tINFO tensorboardx.py:275 -- Removed the following hyperparameter values when logging to tensorboard: {'n_flows': ('__ref_ph', '2944c5dd'), 'epochs': ('__ref_ph', 'a95dbf50'), 'time_steps': ('__ref_ph', '7b44693e')}\n",
      "2024-01-10 22:00:51,965\tINFO tune.py:1042 -- Total run time: 1162.26 seconds (1162.24 seconds for the tuning loop).\n"
     ]
    }
   ],
   "source": [
    "tune_result = tuner.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_flows': 8, 'epochs': 1100, 'time_steps': 125}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tune_result.get_best_result(metric='loss', mode='min').config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}