Ramani et al.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "42fa4383",
   "metadata": {},
   "source": [
    "# Embedding and Imputation on Ramani et al. scHi-C"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ccb9310",
   "metadata": {},
   "source": [
    "## Notes\n",
    "\n",
    "This tutorial uses the new API of Higashi (wrapping all functions of Higashi into the Higashi() class).\n",
    "The old API of Higashi will still be supported and maintained).\n",
    "Please check changelog for the current status of migration from the old API to the new API."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "227b85a4",
   "metadata": {},
   "source": [
    "## Preparation\n",
    "\n",
    "### Download input files\n",
    "Download the demo data from the following link (Ramani et al.)\n",
    "https://drive.google.com/drive/folders/1S0KOMAj60MxQP6mgPV1OKjn_J-lVpzKM?usp=sharing\n",
    "\n",
    "The dataset contains 620 cells from the ML1/ML3 library of the Ramani et al. dataset.\n",
    "\n",
    "Change the file path in the corresponding JSON file according to the location of the downloaded files.\n",
    "\n",
    "### Install Higashi\n",
    "\n",
    "1. install pytorch>=1.8.0 with cuda support when available.\n",
    "2. `conda install -c ruochiz higashi`\n",
    "(Although higashi would install pytorch when needed, there is no guarantee that it will install the correct version with cuda support. It is recommended to install pytorch separately before higashi.)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ca128b3",
   "metadata": {},
   "source": [
    "## Start running Higashi¶"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a09e4d58",
   "metadata": {},
   "source": [
    "### 1. Import package, set the path to the configuration JSON file.¶"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0e5af46e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from higashi.Higashi_wrapper import *\n",
    "config = \"/work/magroup/ruochiz/Higashi/config_dir/config_ramani.JSON\"\n",
    "higashi_model = Higashi(config)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7d5cbf5",
   "metadata": {},
   "source": [
    "### 2. Process data for higashi model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4070da6f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "generating start/end dict for chromosome\n",
      "extracting from data.txt\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1e689dcb3df44d5bb05401ac46ec234",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/15891786 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "generating contact maps for baseline\n",
      "data loaded\n",
      "4110311 False\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b811bac07e834486bf248728b2dbd998",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "creating matrices tasks:   0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total_feats_size 403\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3111e5c064664f8084d926e815744a11",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "higashi_model.process_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23d2dab8",
   "metadata": {},
   "source": [
    "### 3. Prep the higashi model for training and imputation & Stage 1 training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "49960c26",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "setting to gpu:5\n",
      "training on data from: ['chr1', 'chr2', 'chr3', 'chr4', 'chr5', 'chr6', 'chr7', 'chr8', 'chr9', 'chr10', 'chr11', 'chr12', 'chr13', 'chr14', 'chr15', 'chr16', 'chr17', 'chr18', 'chr19', 'chr20', 'chr21', 'chr22', 'chrX']\n",
      "total_sparsity_cell 0.025731658257764963\n",
      "no contractive loss\n",
      "Node type num [620 250 244 199 192 181 172 160 147 142 136 136 134 116 108 103  91  82\n",
      "  79  60  64  49  52 156] [ 620  870 1114 1313 1505 1686 1858 2018 2165 2307 2443 2579 2713 2829\n",
      " 2937 3040 3131 3213 3292 3352 3416 3465 3517 3673]\n",
      "start making attribute\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3d38456a9ea343d6b48ac29cfc71cb8d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/300 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loss 0.5275911092758179 loss best 0.5245773196220398 epochs 124\n",
      "\n",
      "initializing data generator\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f5128d12bea404797794af1278719e0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "initializing data generator\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7643a3476387456f8b99d20b5dd82a80",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First stage training\n",
      "[ Epoch 0 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.6354, mse:  0.6771,  acc: 41.102 %, pearson: 0.272, spearman: 0.249, elapse: 36.865 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5800,  acc: 54.069 %,pearson: 0.412, spearman: 0.385,elapse: 0.076 s\n",
      "update_rate: 0.323048\t0.899341\tupdate_rate: 0.000000\t0.000000\tpair_ratio: 0.1\t\n",
      "[ Epoch 1 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5668, mse:  0.6508,  acc: 66.761 %, pearson: 0.436, spearman: 0.405, elapse: 28.179 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5451,  acc: 74.010 %,pearson: 0.474, spearman: 0.439,elapse: 0.070 s\n",
      "update_rate: 0.242997\t1.459130\tpair_ratio: 0.2\t\n",
      "[ Epoch 2 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5485, mse:  0.6509,  acc: 75.962 %, pearson: 0.489, spearman: 0.443, elapse: 30.677 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5356,  acc: 79.460 %,pearson: 0.512, spearman: 0.469,elapse: 0.068 s\n",
      "update_rate: 0.076554\t0.180740\tpair_ratio: 0.3\t\n",
      "[ Epoch 3 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5387, mse:  0.6616,  acc: 80.589 %, pearson: 0.516, spearman: 0.465, elapse: 30.222 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5426,  acc: 82.243 %,pearson: 0.516, spearman: 0.468,elapse: 0.136 s\n",
      "update_rate: 0.030458\t0.089562\tpair_ratio: 0.4\t\n",
      "[ Epoch 4 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5377, mse:  0.6695,  acc: 82.620 %, pearson: 0.525, spearman: 0.472, elapse: 30.097 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5350,  acc: 83.416 %,pearson: 0.535, spearman: 0.476,elapse: 0.072 s\n",
      "update_rate: 0.030237\t0.123520\tpair_ratio: 0.5\t\n",
      "[ Epoch 5 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5330, mse:  0.6686,  acc: 84.019 %, pearson: 0.536, spearman: 0.481, elapse: 30.259 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5490,  acc: 82.063 %,pearson: 0.521, spearman: 0.465,elapse: 0.063 s\n",
      "update_rate: 0.031520\t0.091921\t\n",
      "[ Epoch 6 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5306, mse:  0.6791,  acc: 84.792 %, pearson: 0.543, spearman: 0.487, elapse: 29.605 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5116,  acc: 85.459 %,pearson: 0.568, spearman: 0.515,elapse: 0.082 s\n",
      "update_rate: 0.029955\t0.089221\t\n",
      "[ Epoch 7 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5295, mse:  0.6706,  acc: 85.204 %, pearson: 0.549, spearman: 0.489, elapse: 29.703 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5324,  acc: 84.661 %,pearson: 0.542, spearman: 0.498,elapse: 0.082 s\n",
      "update_rate: 0.040738\t0.130041\t\n",
      "[ Epoch 8 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5252, mse:  0.6751,  acc: 85.504 %, pearson: 0.555, spearman: 0.495, elapse: 29.926 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5257,  acc: 84.447 %,pearson: 0.570, spearman: 0.509,elapse: 0.073 s\n",
      "update_rate: 0.019171\t0.082120\t\n",
      "[ Epoch 9 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5258, mse:  0.6754,  acc: 85.583 %, pearson: 0.554, spearman: 0.494, elapse: 29.725 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5377,  acc: 84.857 %,pearson: 0.554, spearman: 0.482,elapse: 0.074 s\n",
      "update_rate: 0.017176\t0.090444\t\n",
      "[ Epoch 10 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5241, mse:  0.6758,  acc: 85.746 %, pearson: 0.556, spearman: 0.495, elapse: 30.459 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5359,  acc: 85.111 %,pearson: 0.531, spearman: 0.484,elapse: 0.074 s\n",
      "update_rate: 0.015618\t0.085529\t\n",
      "[ Epoch 11 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5223, mse:  0.6764,  acc: 85.866 %, pearson: 0.558, spearman: 0.498, elapse: 30.343 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5434,  acc: 84.987 %,pearson: 0.537, spearman: 0.477,elapse: 0.083 s\n",
      "update_rate: 0.017996\t0.088646\t\n",
      "[ Epoch 12 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5245, mse:  0.7163,  acc: 85.781 %, pearson: 0.559, spearman: 0.495, elapse: 27.434 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5548,  acc: 81.801 %,pearson: 0.548, spearman: 0.479,elapse: 0.063 s\n",
      "update_rate: 0.012870\t0.082210\t\n",
      "[ Epoch 13 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5218, mse:  0.6702,  acc: 85.962 %, pearson: 0.563, spearman: 0.499, elapse: 29.050 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5306,  acc: 87.343 %,pearson: 0.557, spearman: 0.505,elapse: 0.095 s\n",
      "update_rate: 0.014402\t0.096428\t\n",
      "[ Epoch 14 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5191, mse:  0.6828,  acc: 86.109 %, pearson: 0.566, spearman: 0.501, elapse: 28.916 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5428,  acc: 84.473 %,pearson: 0.532, spearman: 0.471,elapse: 0.061 s\n",
      "update_rate: 0.014001\t0.088402\t\n",
      "[ Epoch 15 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5187, mse:  0.6898,  acc: 86.085 %, pearson: 0.565, spearman: 0.502, elapse: 29.989 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5473,  acc: 84.206 %,pearson: 0.516, spearman: 0.477,elapse: 0.064 s\n",
      "update_rate: 0.017288\t0.091383\t\n",
      "[ Epoch 16 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5197, mse:  0.6720,  acc: 86.039 %, pearson: 0.566, spearman: 0.500, elapse: 30.059 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5100,  acc: 87.258 %,pearson: 0.596, spearman: 0.523,elapse: 0.075 s\n",
      "update_rate: 0.011680\t0.081116\t\n",
      "[ Epoch 17 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5207, mse:  0.6863,  acc: 86.082 %, pearson: 0.566, spearman: 0.500, elapse: 28.747 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5333,  acc: 85.697 %,pearson: 0.554, spearman: 0.486,elapse: 0.086 s\n",
      "update_rate: 0.015077\t0.113325\t\n",
      "[ Epoch 18 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5177, mse:  0.6734,  acc: 86.159 %, pearson: 0.568, spearman: 0.502, elapse: 28.996 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5342,  acc: 85.494 %,pearson: 0.546, spearman: 0.487,elapse: 0.064 s\n",
      "update_rate: 0.023731\t0.115483\t\n",
      "[ Epoch 19 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5193, mse:  0.6860,  acc: 86.093 %, pearson: 0.567, spearman: 0.500, elapse: 30.232 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5277,  acc: 85.435 %,pearson: 0.571, spearman: 0.498,elapse: 0.065 s\n",
      "update_rate: 0.017020\t0.103286\t\n",
      "[ Epoch 20 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5163, mse:  0.6881,  acc: 86.197 %, pearson: 0.572, spearman: 0.502, elapse: 28.415 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5397,  acc: 85.156 %,pearson: 0.537, spearman: 0.485,elapse: 0.075 s\n",
      "update_rate: 0.011813\t0.082017\t\n",
      "[ Epoch 21 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5148, mse:  0.6759,  acc: 86.272 %, pearson: 0.572, spearman: 0.504, elapse: 30.497 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5235,  acc: 86.256 %,pearson: 0.559, spearman: 0.506,elapse: 0.064 s\n",
      "update_rate: 0.015305\t0.119051\t\n",
      "[ Epoch 22 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5154, mse:  0.6676,  acc: 86.264 %, pearson: 0.572, spearman: 0.503, elapse: 31.247 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5463,  acc: 85.318 %,pearson: 0.540, spearman: 0.470,elapse: 0.082 s\n",
      "update_rate: 0.013051\t0.103463\t\n",
      "[ Epoch 23 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5151, mse:  0.6825,  acc: 86.263 %, pearson: 0.572, spearman: 0.504, elapse: 31.479 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5334,  acc: 84.920 %,pearson: 0.544, spearman: 0.490,elapse: 0.065 s\n",
      "update_rate: 0.013545\t0.091171\t\n",
      "[ Epoch 24 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5146, mse:  0.7213,  acc: 86.291 %, pearson: 0.573, spearman: 0.504, elapse: 31.474 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5291,  acc: 84.733 %,pearson: 0.564, spearman: 0.504,elapse: 0.083 s\n",
      "update_rate: 0.014213\t0.117248\t\n",
      "[ Epoch 25 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5158, mse:  0.6743,  acc: 86.287 %, pearson: 0.574, spearman: 0.504, elapse: 30.877 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5350,  acc: 85.221 %,pearson: 0.552, spearman: 0.486,elapse: 0.066 s\n",
      "update_rate: 0.013600\t0.107671\t\n",
      "[ Epoch 26 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5154, mse:  0.6778,  acc: 86.232 %, pearson: 0.572, spearman: 0.504, elapse: 29.564 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5167,  acc: 85.690 %,pearson: 0.573, spearman: 0.502,elapse: 0.073 s\n",
      "update_rate: 0.011627\t0.079672\t\n",
      "[ Epoch 27 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5125, mse:  0.6816,  acc: 86.351 %, pearson: 0.576, spearman: 0.505, elapse: 30.822 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5342,  acc: 85.962 %,pearson: 0.549, spearman: 0.492,elapse: 0.066 s\n",
      "update_rate: 0.007172\t0.077339\t\n",
      "[ Epoch 28 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5132, mse:  0.6822,  acc: 86.235 %, pearson: 0.573, spearman: 0.504, elapse: 27.550 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5340,  acc: 86.191 %,pearson: 0.566, spearman: 0.502,elapse: 0.063 s\n",
      "update_rate: 0.008826\t0.076508\t\n",
      "[ Epoch 29 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5120, mse:  0.6827,  acc: 86.418 %, pearson: 0.579, spearman: 0.507, elapse: 29.418 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5362,  acc: 86.530 %,pearson: 0.542, spearman: 0.482,elapse: 0.064 s\n",
      "update_rate: 0.009570\t0.090333\t\n",
      "[ Epoch 30 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5115, mse:  0.6870,  acc: 86.338 %, pearson: 0.577, spearman: 0.506, elapse: 28.122 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5300,  acc: 84.464 %,pearson: 0.555, spearman: 0.488,elapse: 0.067 s\n",
      "update_rate: 0.009444\t0.089440\t\n",
      "[ Epoch 31 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5128, mse:  0.6957,  acc: 86.341 %, pearson: 0.579, spearman: 0.506, elapse: 27.391 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5369,  acc: 85.202 %,pearson: 0.546, spearman: 0.496,elapse: 0.065 s\n",
      "update_rate: 0.007712\t0.102009\t\n",
      "[ Epoch 32 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5139, mse:  0.7154,  acc: 86.322 %, pearson: 0.576, spearman: 0.505, elapse: 28.413 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5307,  acc: 84.707 %,pearson: 0.562, spearman: 0.493,elapse: 0.065 s\n",
      "update_rate: 0.011093\t0.142289\t\n",
      "[ Epoch 33 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5114, mse:  0.7193,  acc: 86.327 %, pearson: 0.578, spearman: 0.506, elapse: 26.391 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5396,  acc: 84.674 %,pearson: 0.543, spearman: 0.483,elapse: 0.076 s\n",
      "update_rate: 0.010437\t0.104828\t\n",
      "[ Epoch 34 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5110, mse:  0.6860,  acc: 86.439 %, pearson: 0.580, spearman: 0.508, elapse: 27.195 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5315,  acc: 84.295 %,pearson: 0.542, spearman: 0.488,elapse: 0.062 s\n",
      "update_rate: 0.006160\t0.083286\t\n",
      "[ Epoch 35 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5110, mse:  0.6762,  acc: 86.345 %, pearson: 0.580, spearman: 0.506, elapse: 28.335 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5275,  acc: 85.800 %,pearson: 0.561, spearman: 0.496,elapse: 0.072 s\n",
      "update_rate: 0.008604\t0.096679\t\n",
      "[ Epoch 36 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5107, mse:  0.7088,  acc: 86.375 %, pearson: 0.578, spearman: 0.507, elapse: 28.430 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5381,  acc: 85.801 %,pearson: 0.539, spearman: 0.489,elapse: 0.080 s\n",
      "update_rate: 0.011499\t0.117496\t\n",
      "[ Epoch 37 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5120, mse:  0.7678,  acc: 86.337 %, pearson: 0.578, spearman: 0.505, elapse: 28.609 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5361,  acc: 84.284 %,pearson: 0.541, spearman: 0.480,elapse: 0.066 s\n",
      "update_rate: 0.010309\t0.099568\t\n",
      "[ Epoch 38 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5118, mse:  0.6806,  acc: 86.335 %, pearson: 0.578, spearman: 0.506, elapse: 30.345 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5374,  acc: 84.019 %,pearson: 0.537, spearman: 0.487,elapse: 0.063 s\n",
      "update_rate: 0.012113\t0.103632\t\n",
      "[ Epoch 39 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5108, mse:  0.6876,  acc: 86.369 %, pearson: 0.578, spearman: 0.507, elapse: 28.340 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5393,  acc: 86.536 %,pearson: 0.539, spearman: 0.482,elapse: 0.062 s\n",
      "update_rate: 0.010063\t0.098517\t\n",
      "[ Epoch 40 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5104, mse:  0.6816,  acc: 86.384 %, pearson: 0.579, spearman: 0.508, elapse: 29.345 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5323,  acc: 86.183 %,pearson: 0.563, spearman: 0.504,elapse: 0.064 s\n",
      "update_rate: 0.008165\t0.105850\t\n",
      "[ Epoch 41 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5078, mse:  0.7021,  acc: 86.529 %, pearson: 0.584, spearman: 0.510, elapse: 28.196 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5207,  acc: 86.777 %,pearson: 0.591, spearman: 0.510,elapse: 0.066 s\n",
      "update_rate: 0.010921\t0.120252\t\n",
      "[ Epoch 42 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5083, mse:  0.7142,  acc: 86.387 %, pearson: 0.582, spearman: 0.509, elapse: 26.288 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5420,  acc: 86.417 %,pearson: 0.565, spearman: 0.496,elapse: 0.062 s\n",
      "update_rate: 0.008740\t0.092063\t\n",
      "[ Epoch 43 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5105, mse:  0.6738,  acc: 86.304 %, pearson: 0.581, spearman: 0.506, elapse: 28.824 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5380,  acc: 86.120 %,pearson: 0.554, spearman: 0.483,elapse: 0.065 s\n",
      "update_rate: 0.005830\t0.071042\t\n",
      "[ Epoch 44 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5074, mse:  0.6845,  acc: 86.411 %, pearson: 0.583, spearman: 0.510, elapse: 27.667 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5404,  acc: 83.586 %,pearson: 0.538, spearman: 0.490,elapse: 0.065 s\n",
      "update_rate: 0.008084\t0.087033\t\n",
      "[ Epoch 45 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5077, mse:  0.7073,  acc: 86.426 %, pearson: 0.582, spearman: 0.509, elapse: 27.865 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5225,  acc: 86.763 %,pearson: 0.572, spearman: 0.512,elapse: 0.065 s\n",
      "update_rate: 0.008579\t0.097202\t\n",
      "[ Epoch 46 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5085, mse:  0.6669,  acc: 86.416 %, pearson: 0.582, spearman: 0.509, elapse: 27.815 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5239,  acc: 85.475 %,pearson: 0.557, spearman: 0.493,elapse: 0.063 s\n",
      "update_rate: 0.007731\t0.104649\t\n",
      "[ Epoch 47 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5077, mse:  0.6724,  acc: 86.422 %, pearson: 0.583, spearman: 0.509, elapse: 28.139 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5386,  acc: 85.612 %,pearson: 0.549, spearman: 0.498,elapse: 0.065 s\n",
      "update_rate: 0.007805\t0.092815\t\n",
      "[ Epoch 48 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5098, mse:  0.6819,  acc: 86.430 %, pearson: 0.583, spearman: 0.509, elapse: 27.178 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5458,  acc: 86.717 %,pearson: 0.554, spearman: 0.480,elapse: 0.061 s\n",
      "update_rate: 0.008153\t0.070667\t\n",
      "[ Epoch 49 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5048, mse:  0.6844,  acc: 86.584 %, pearson: 0.587, spearman: 0.513, elapse: 26.218 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5388,  acc: 85.558 %,pearson: 0.547, spearman: 0.493,elapse: 0.065 s\n",
      "update_rate: 0.007766\t0.106476\t\n",
      "[ Epoch 50 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5069, mse:  0.6684,  acc: 86.529 %, pearson: 0.585, spearman: 0.511, elapse: 28.385 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5234,  acc: 86.380 %,pearson: 0.560, spearman: 0.508,elapse: 0.065 s\n",
      "update_rate: 0.006069\t0.093553\t\n",
      "[ Epoch 51 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5071, mse:  0.7125,  acc: 86.408 %, pearson: 0.584, spearman: 0.510, elapse: 28.550 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5321,  acc: 86.510 %,pearson: 0.573, spearman: 0.502,elapse: 0.064 s\n",
      "update_rate: 0.005266\t0.078633\t\n",
      "[ Epoch 52 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5045, mse:  0.6802,  acc: 86.552 %, pearson: 0.585, spearman: 0.512, elapse: 28.401 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5321,  acc: 85.977 %,pearson: 0.552, spearman: 0.490,elapse: 0.090 s\n",
      "update_rate: 0.006230\t0.076941\t\n",
      "[ Epoch 53 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5064, mse:  0.9336,  acc: 86.378 %, pearson: 0.584, spearman: 0.510, elapse: 27.517 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5376,  acc: 84.837 %,pearson: 0.555, spearman: 0.471,elapse: 0.064 s\n",
      "update_rate: 0.005448\t0.060127\t\n",
      "[ Epoch 54 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5062, mse:  0.6715,  acc: 86.520 %, pearson: 0.584, spearman: 0.511, elapse: 30.982 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5410,  acc: 86.042 %,pearson: 0.556, spearman: 0.492,elapse: 0.068 s\n",
      "update_rate: 0.005238\t0.070634\t\n",
      "[ Epoch 55 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5073, mse:  0.6661,  acc: 86.419 %, pearson: 0.583, spearman: 0.509, elapse: 30.222 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5203,  acc: 85.664 %,pearson: 0.573, spearman: 0.505,elapse: 0.065 s\n",
      "update_rate: 0.004959\t0.064685\t\n",
      "[ Epoch 56 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5093, mse:  0.6652,  acc: 86.358 %, pearson: 0.581, spearman: 0.508, elapse: 29.061 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5263,  acc: 86.170 %,pearson: 0.562, spearman: 0.507,elapse: 0.063 s\n",
      "update_rate: 0.005254\t0.063496\t\n",
      "[ Epoch 57 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5059, mse:  0.6651,  acc: 86.461 %, pearson: 0.585, spearman: 0.510, elapse: 28.346 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5333,  acc: 84.766 %,pearson: 0.554, spearman: 0.484,elapse: 0.065 s\n",
      "update_rate: 0.005926\t0.084598\t\n",
      "[ Epoch 58 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5050, mse:  0.6736,  acc: 86.459 %, pearson: 0.587, spearman: 0.511, elapse: 28.200 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5274,  acc: 86.458 %,pearson: 0.563, spearman: 0.501,elapse: 0.094 s\n",
      "update_rate: 0.007277\t0.066024\t\n",
      "[ Epoch 59 of 60 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5055, mse:  0.6645,  acc: 86.401 %, pearson: 0.585, spearman: 0.511, elapse: 29.767 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.5282,  acc: 84.935 %,pearson: 0.545, spearman: 0.492,elapse: 0.063 s\n",
      "update_rate: 0.006355\t0.066394\t\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  - (Validation-hyper) bce:  0.6040,  acc: 78.042 %, auc: 0.471, aupr: 0.407,elapse: 1.009 s\n"
     ]
    }
   ],
   "source": [
    "higashi_model.prep_model()\n",
    "# Stage 1 training\n",
    "higashi_model.train_for_embeddings()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "104f985a",
   "metadata": {},
   "source": [
    "### 4. Stage 2 training and imputation without neighbor information"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b415ec08",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96de564b04384476bdd563aeae5e47e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65999d09fced43b0b4af764bfd79f155",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pass_pseudo_id False\n",
      "pass_pseudo_id False\n",
      "Second stage training\n",
      "[ Epoch 0 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.5279, mse:  0.0000,  acc: 84.973 %, pearson: 0.529, spearman: 0.501, elapse: 64.212 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4633,  acc: 88.255 %,pearson: 0.647, spearman: 0.582,elapse: 0.105 s\n",
      "\n",
      "[ Epoch 1 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4973, mse:  0.0000,  acc: 88.044 %, pearson: 0.592, spearman: 0.539, elapse: 52.411 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4082,  acc: 91.732 %,pearson: 0.724, spearman: 0.610,elapse: 0.088 s\n",
      "\n",
      "[ Epoch 2 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4839, mse:  0.0000,  acc: 89.259 %, pearson: 0.621, spearman: 0.552, elapse: 51.112 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4068,  acc: 91.803 %,pearson: 0.725, spearman: 0.613,elapse: 0.099 s\n",
      "\n",
      "[ Epoch 3 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4730, mse:  0.0000,  acc: 90.224 %, pearson: 0.642, spearman: 0.562, elapse: 53.427 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4178,  acc: 91.107 %,pearson: 0.729, spearman: 0.617,elapse: 0.102 s\n",
      "\n",
      "[ Epoch 4 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4637, mse:  0.0000,  acc: 90.901 %, pearson: 0.657, spearman: 0.569, elapse: 54.640 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3648,  acc: 94.883 %,pearson: 0.766, spearman: 0.630,elapse: 0.130 s\n",
      "\n",
      "[ Epoch 5 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4569, mse:  0.0000,  acc: 91.414 %, pearson: 0.670, spearman: 0.574, elapse: 47.795 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3676,  acc: 94.681 %,pearson: 0.782, spearman: 0.627,elapse: 0.091 s\n",
      "\n",
      "[ Epoch 6 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4508, mse:  0.0000,  acc: 91.830 %, pearson: 0.679, spearman: 0.578, elapse: 47.757 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3708,  acc: 94.225 %,pearson: 0.786, spearman: 0.632,elapse: 0.104 s\n",
      "\n",
      "[ Epoch 7 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4477, mse:  0.0000,  acc: 92.050 %, pearson: 0.685, spearman: 0.579, elapse: 46.725 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3771,  acc: 94.134 %,pearson: 0.799, spearman: 0.632,elapse: 0.097 s\n",
      "\n",
      "[ Epoch 8 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4397, mse:  0.0000,  acc: 92.531 %, pearson: 0.694, spearman: 0.585, elapse: 44.193 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3538,  acc: 95.456 %,pearson: 0.786, spearman: 0.636,elapse: 0.101 s\n",
      "\n",
      "[ Epoch 9 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4381, mse:  0.0000,  acc: 92.695 %, pearson: 0.699, spearman: 0.586, elapse: 45.761 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3521,  acc: 95.286 %,pearson: 0.802, spearman: 0.636,elapse: 0.099 s\n",
      "\n",
      "[ Epoch 10 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4362, mse:  0.0000,  acc: 92.827 %, pearson: 0.701, spearman: 0.587, elapse: 46.425 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3595,  acc: 95.638 %,pearson: 0.792, spearman: 0.638,elapse: 0.149 s\n",
      "\n",
      "[ Epoch 11 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4338, mse:  0.0000,  acc: 93.041 %, pearson: 0.705, spearman: 0.588, elapse: 45.394 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3533,  acc: 95.566 %,pearson: 0.809, spearman: 0.637,elapse: 0.105 s\n",
      "\n",
      "[ Epoch 12 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4289, mse:  0.0000,  acc: 93.222 %, pearson: 0.709, spearman: 0.590, elapse: 45.188 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3455,  acc: 96.133 %,pearson: 0.799, spearman: 0.637,elapse: 0.106 s\n",
      "\n",
      "[ Epoch 13 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4284, mse:  0.0000,  acc: 93.341 %, pearson: 0.711, spearman: 0.591, elapse: 48.022 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3453,  acc: 96.094 %,pearson: 0.804, spearman: 0.638,elapse: 0.091 s\n",
      "no improve: 1\t\n",
      "[ Epoch 14 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4259, mse:  0.0000,  acc: 93.424 %, pearson: 0.715, spearman: 0.592, elapse: 48.913 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3315,  acc: 96.543 %,pearson: 0.818, spearman: 0.639,elapse: 0.108 s\n",
      "\n",
      "[ Epoch 15 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4212, mse:  0.0000,  acc: 93.664 %, pearson: 0.719, spearman: 0.594, elapse: 49.288 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3320,  acc: 96.530 %,pearson: 0.818, spearman: 0.640,elapse: 0.106 s\n",
      "\n",
      "[ Epoch 16 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4235, mse:  0.0000,  acc: 93.602 %, pearson: 0.719, spearman: 0.594, elapse: 47.218 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3372,  acc: 96.615 %,pearson: 0.807, spearman: 0.639,elapse: 0.103 s\n",
      "no improve: 1\t\n",
      "[ Epoch 17 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4235, mse:  0.0000,  acc: 93.707 %, pearson: 0.719, spearman: 0.594, elapse: 45.967 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3448,  acc: 95.729 %,pearson: 0.808, spearman: 0.636,elapse: 0.103 s\n",
      "no improve: 2\t\n",
      "[ Epoch 18 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4168, mse:  0.0000,  acc: 93.905 %, pearson: 0.724, spearman: 0.596, elapse: 50.365 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3358,  acc: 96.497 %,pearson: 0.804, spearman: 0.639,elapse: 0.103 s\n",
      "\n",
      "[ Epoch 19 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4181, mse:  0.0000,  acc: 93.901 %, pearson: 0.724, spearman: 0.596, elapse: 51.993 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3228,  acc: 97.116 %,pearson: 0.823, spearman: 0.641,elapse: 0.112 s\n",
      "no improve: 1\t\n",
      "[ Epoch 20 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4183, mse:  0.0000,  acc: 93.901 %, pearson: 0.724, spearman: 0.596, elapse: 49.173 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3434,  acc: 96.055 %,pearson: 0.814, spearman: 0.640,elapse: 0.366 s\n",
      "no improve: 2\t\n",
      "[ Epoch 21 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4165, mse:  0.0000,  acc: 94.053 %, pearson: 0.727, spearman: 0.597, elapse: 62.958 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3430,  acc: 96.243 %,pearson: 0.816, spearman: 0.640,elapse: 0.165 s\n",
      "no improve: 3\t\n",
      "[ Epoch 22 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4137, mse:  0.0000,  acc: 94.153 %, pearson: 0.728, spearman: 0.598, elapse: 54.888 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3173,  acc: 97.546 %,pearson: 0.823, spearman: 0.643,elapse: 0.180 s\n",
      "\n",
      "[ Epoch 23 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4130, mse:  0.0000,  acc: 94.216 %, pearson: 0.730, spearman: 0.599, elapse: 33.639 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3337,  acc: 96.777 %,pearson: 0.812, spearman: 0.643,elapse: 0.094 s\n",
      "no improve: 1\t\n",
      "[ Epoch 24 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4121, mse:  0.0000,  acc: 94.246 %, pearson: 0.730, spearman: 0.599, elapse: 32.730 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3519,  acc: 95.645 %,pearson: 0.813, spearman: 0.638,elapse: 0.104 s\n",
      "\n",
      "[ Epoch 25 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4118, mse:  0.0000,  acc: 94.274 %, pearson: 0.732, spearman: 0.599, elapse: 32.502 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3309,  acc: 97.070 %,pearson: 0.815, spearman: 0.641,elapse: 0.079 s\n",
      "no improve: 1\t\n",
      "[ Epoch 26 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4105, mse:  0.0000,  acc: 94.332 %, pearson: 0.732, spearman: 0.600, elapse: 32.861 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3252,  acc: 97.168 %,pearson: 0.817, spearman: 0.642,elapse: 0.100 s\n",
      "\n",
      "[ Epoch 27 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4080, mse:  0.0000,  acc: 94.438 %, pearson: 0.735, spearman: 0.601, elapse: 33.355 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3339,  acc: 96.641 %,pearson: 0.810, spearman: 0.641,elapse: 0.093 s\n",
      "\n",
      "[ Epoch 28 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4095, mse:  0.0000,  acc: 94.413 %, pearson: 0.733, spearman: 0.600, elapse: 32.508 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3409,  acc: 96.367 %,pearson: 0.810, spearman: 0.640,elapse: 0.079 s\n",
      "no improve: 1\t\n",
      "[ Epoch 29 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4086, mse:  0.0000,  acc: 94.444 %, pearson: 0.736, spearman: 0.600, elapse: 32.378 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3475,  acc: 96.283 %,pearson: 0.818, spearman: 0.641,elapse: 0.090 s\n",
      "no improve: 2\t\n",
      "[ Epoch 30 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4090, mse:  0.0000,  acc: 94.447 %, pearson: 0.734, spearman: 0.600, elapse: 33.282 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3261,  acc: 97.507 %,pearson: 0.832, spearman: 0.643,elapse: 0.093 s\n",
      "no improve: 3\t\n",
      "[ Epoch 31 of 45 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4077, mse:  0.0000,  acc: 94.533 %, pearson: 0.736, spearman: 0.601, elapse: 33.187 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3414,  acc: 96.771 %,pearson: 0.821, spearman: 0.642,elapse: 0.096 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  - (Validation-hyper) bce:  0.3443,  acc: 96.361 %, auc: 0.818, aupr: 0.642,elapse: 1.559 s\n",
      "total number of triplets to predict: (223102, 3)\n",
      "Imputing exp_zinb3_nbr_0_impute: 1 of 620, takes 0.49 s estimate 302.04 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 11 of 620, takes 3.86 s estimate 213.64 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 21 of 620, takes 7.22 s estimate 205.91 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 31 of 620, takes 10.53 s estimate 200.07 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 41 of 620, takes 14.31 s estimate 202.07 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 51 of 620, takes 17.72 s estimate 197.65 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 61 of 620, takes 20.98 s estimate 192.28 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 71 of 620, takes 24.43 s estimate 188.89 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 81 of 620, takes 27.88 s estimate 185.53 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 91 of 620, takes 31.23 s estimate 181.56 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 101 of 620, takes 34.57 s estimate 177.67 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 111 of 620, takes 37.89 s estimate 173.76 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 121 of 620, takes 41.19 s estimate 169.85 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 131 of 620, takes 44.49 s estimate 166.08 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 141 of 620, takes 47.78 s estimate 162.31 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 151 of 620, takes 51.28 s estimate 159.29 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 161 of 620, takes 54.55 s estimate 155.52 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 171 of 620, takes 57.80 s estimate 151.76 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 181 of 620, takes 61.04 s estimate 148.05 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 191 of 620, takes 64.45 s estimate 144.75 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 201 of 620, takes 67.68 s estimate 141.08 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 211 of 620, takes 70.89 s estimate 137.41 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 221 of 620, takes 74.17 s estimate 133.90 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 231 of 620, takes 77.39 s estimate 130.32 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 241 of 620, takes 80.63 s estimate 126.80 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 251 of 620, takes 84.20 s estimate 123.79 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 261 of 620, takes 87.60 s estimate 120.49 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 271 of 620, takes 90.82 s estimate 116.96 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 281 of 620, takes 93.75 s estimate 113.10 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 291 of 620, takes 96.97 s estimate 109.63 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 301 of 620, takes 100.34 s estimate 106.34 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 311 of 620, takes 103.60 s estimate 102.93 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 321 of 620, takes 106.92 s estimate 99.59 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 331 of 620, takes 110.24 s estimate 96.25 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 341 of 620, takes 113.44 s estimate 92.82 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 351 of 620, takes 116.73 s estimate 89.46 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 361 of 620, takes 120.27 s estimate 86.29 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 371 of 620, takes 123.60 s estimate 82.96 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 381 of 620, takes 126.94 s estimate 79.63 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 391 of 620, takes 130.20 s estimate 76.25 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 401 of 620, takes 133.44 s estimate 72.87 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 411 of 620, takes 136.74 s estimate 69.53 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 421 of 620, takes 140.14 s estimate 66.24 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 431 of 620, takes 143.47 s estimate 62.91 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 441 of 620, takes 147.03 s estimate 59.68 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 451 of 620, takes 150.24 s estimate 56.30 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 461 of 620, takes 153.60 s estimate 52.98 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 471 of 620, takes 156.91 s estimate 49.64 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 481 of 620, takes 160.18 s estimate 46.29 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 491 of 620, takes 163.51 s estimate 42.96 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 501 of 620, takes 166.82 s estimate 39.62 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 511 of 620, takes 170.23 s estimate 36.31 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 521 of 620, takes 173.54 s estimate 32.98 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 531 of 620, takes 177.01 s estimate 29.67 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 541 of 620, takes 180.29 s estimate 26.33 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 551 of 620, takes 183.60 s estimate 22.99 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 561 of 620, takes 186.80 s estimate 19.65 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 571 of 620, takes 189.87 s estimate 16.29 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 581 of 620, takes 193.13 s estimate 12.96 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 591 of 620, takes 196.33 s estimate 9.63 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 601 of 620, takes 199.61 s estimate 6.31 s to finish\n",
      "Imputing exp_zinb3_nbr_0_impute: 611 of 620, takes 202.77 s estimate 2.99 s to finish\n",
      "finish imputing, used 205.69 s\n"
     ]
    }
   ],
   "source": [
    "higashi_model.train_for_imputation_nbr_0()\n",
    "higashi_model.impute_no_nbr()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca78a204",
   "metadata": {},
   "source": [
    "### 5. Stage 3 training and imputation with neighbor information"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3b211f9e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "768a6c22d4d14ac9bd1204bf98cc4ca9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f311d54d68f04902aed428015f64ca67",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "getting cell nbr's nbr list\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4bcfd0a745324c5f8d95d2b2f89eccd0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/620 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[list([]) array([  1, 323, 128,  13, 229, 270])\n",
      " array([  2, 287,  30, 461, 442, 243])\n",
      " array([  3,  52, 222,  82, 314, 161])\n",
      " array([  4, 304, 269, 270,  37,   1])\n",
      " array([  5, 161, 173, 163,  82, 182])\n",
      " array([  6,  26, 211, 128,  21, 184])\n",
      " array([  7, 111, 158,  25, 188,  77])\n",
      " array([  8, 163,  52, 257,   5, 421])\n",
      " array([  9,  31, 119,  93, 111,  95])] [list([])\n",
      " array([0.3312287 , 0.13527231, 0.13490379, 0.13316663, 0.13272686,\n",
      "        0.1327017 ], dtype=float32)\n",
      " array([0.3429045 , 0.14798632, 0.12863994, 0.12776986, 0.12732387,\n",
      "        0.12537548], dtype=float32)\n",
      " array([0.3519162 , 0.14112534, 0.12917556, 0.1282021 , 0.12641005,\n",
      "        0.12317079], dtype=float32)\n",
      " array([0.36637768, 0.13916051, 0.13213073, 0.12381034, 0.11935434,\n",
      "        0.11916636], dtype=float32)\n",
      " array([0.3251313 , 0.14442989, 0.13803163, 0.1361649 , 0.12902556,\n",
      "        0.12721674], dtype=float32)\n",
      " array([0.42240748, 0.11887022, 0.11845106, 0.1168405 , 0.11433152,\n",
      "        0.10909923], dtype=float32)\n",
      " array([0.38637385, 0.12964743, 0.12601624, 0.12070298, 0.11866991,\n",
      "        0.11858955], dtype=float32)\n",
      " array([0.33793476, 0.14705443, 0.134056  , 0.12882584, 0.12655142,\n",
      "        0.12557748], dtype=float32)\n",
      " array([0.34192237, 0.14655867, 0.12958705, 0.12869608, 0.12681097,\n",
      "        0.12642482], dtype=float32)                                ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4be87d0910d0412ea3fdc8a9a160feed",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/621 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final stage training\n",
      "[ Epoch 0 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4714, mse:  0.0000,  acc: 89.718 %, pearson: 0.664, spearman: 0.555, elapse: 58.329 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4008,  acc: 91.745 %,pearson: 0.765, spearman: 0.613,elapse: 0.080 s\n",
      "\n",
      "[ Epoch 1 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4671, mse:  0.0000,  acc: 90.034 %, pearson: 0.672, spearman: 0.558, elapse: 32.670 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4045,  acc: 91.536 %,pearson: 0.765, spearman: 0.614,elapse: 0.091 s\n",
      "\n",
      "[ Epoch 2 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4633, mse:  0.0000,  acc: 90.239 %, pearson: 0.677, spearman: 0.560, elapse: 33.988 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4094,  acc: 91.139 %,pearson: 0.755, spearman: 0.606,elapse: 0.092 s\n",
      "\n",
      "[ Epoch 3 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4632, mse:  0.0000,  acc: 90.259 %, pearson: 0.679, spearman: 0.561, elapse: 33.509 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4000,  acc: 91.712 %,pearson: 0.789, spearman: 0.616,elapse: 0.083 s\n",
      "no improve: 1\t\n",
      "[ Epoch 4 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4625, mse:  0.0000,  acc: 90.295 %, pearson: 0.679, spearman: 0.561, elapse: 33.665 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3910,  acc: 91.895 %,pearson: 0.773, spearman: 0.614,elapse: 0.095 s\n",
      "no improve: 2\t\n",
      "[ Epoch 5 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4626, mse:  0.0000,  acc: 90.354 %, pearson: 0.680, spearman: 0.561, elapse: 34.291 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3896,  acc: 92.617 %,pearson: 0.785, spearman: 0.617,elapse: 0.098 s\n",
      "no improve: 3\t\n",
      "[ Epoch 6 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4598, mse:  0.0000,  acc: 90.414 %, pearson: 0.683, spearman: 0.563, elapse: 33.604 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3855,  acc: 92.350 %,pearson: 0.796, spearman: 0.619,elapse: 0.084 s\n",
      "\n",
      "[ Epoch 7 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4603, mse:  0.0000,  acc: 90.420 %, pearson: 0.682, spearman: 0.562, elapse: 33.951 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3910,  acc: 92.220 %,pearson: 0.777, spearman: 0.615,elapse: 0.118 s\n",
      "no improve: 1\t\n",
      "[ Epoch 8 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4610, mse:  0.0000,  acc: 90.430 %, pearson: 0.683, spearman: 0.563, elapse: 34.217 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.4058,  acc: 91.628 %,pearson: 0.775, spearman: 0.610,elapse: 0.121 s\n",
      "no improve: 2\t\n",
      "[ Epoch 9 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4592, mse:  0.0000,  acc: 90.528 %, pearson: 0.685, spearman: 0.564, elapse: 32.916 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3914,  acc: 92.201 %,pearson: 0.789, spearman: 0.620,elapse: 0.138 s\n",
      "no improve: 3\t\n",
      "[ Epoch 10 of 30 ]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       " - (Training) :   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Train)   bce:  0.4597, mse:  0.0000,  acc: 90.557 %, pearson: 0.686, spearman: 0.563, elapse: 34.285 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "- (Valid) bce:  0.3842,  acc: 92.962 %,pearson: 0.799, spearman: 0.621,elapse: 0.111 s\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  - (Validation)   :   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  - (Validation-hyper) bce:  0.3861,  acc: 92.832 %, auc: 0.790, aupr: 0.615,elapse: 1.955 s\n",
      "total number of triplets to predict: (223102, 3)\n",
      "Imputing exp_zinb3_nbr_5_impute: 1 of 620, takes 0.42 s estimate 257.59 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 11 of 620, takes 4.36 s estimate 241.51 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 21 of 620, takes 8.25 s estimate 235.45 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 31 of 620, takes 12.28 s estimate 233.40 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 41 of 620, takes 16.20 s estimate 228.73 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 51 of 620, takes 20.01 s estimate 223.27 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 61 of 620, takes 23.85 s estimate 218.56 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 71 of 620, takes 27.68 s estimate 214.03 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 81 of 620, takes 31.71 s estimate 211.03 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 91 of 620, takes 35.63 s estimate 207.12 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 101 of 620, takes 39.49 s estimate 202.91 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 111 of 620, takes 43.41 s estimate 199.06 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 121 of 620, takes 47.57 s estimate 196.18 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 131 of 620, takes 51.40 s estimate 191.86 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 141 of 620, takes 54.82 s estimate 186.22 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 151 of 620, takes 58.44 s estimate 181.52 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 161 of 620, takes 62.29 s estimate 177.59 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 171 of 620, takes 66.12 s estimate 173.60 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 181 of 620, takes 69.98 s estimate 169.73 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 191 of 620, takes 74.36 s estimate 167.02 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 201 of 620, takes 78.21 s estimate 163.04 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 211 of 620, takes 82.02 s estimate 158.99 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 221 of 620, takes 85.83 s estimate 154.96 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 231 of 620, takes 89.72 s estimate 151.09 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 241 of 620, takes 93.56 s estimate 147.13 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 251 of 620, takes 97.36 s estimate 143.13 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 261 of 620, takes 101.57 s estimate 139.71 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 271 of 620, takes 105.44 s estimate 135.79 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 281 of 620, takes 109.19 s estimate 131.73 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 291 of 620, takes 113.08 s estimate 127.85 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 301 of 620, takes 116.92 s estimate 123.91 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 311 of 620, takes 120.75 s estimate 119.98 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 321 of 620, takes 124.54 s estimate 116.00 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 331 of 620, takes 128.33 s estimate 112.04 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 341 of 620, takes 132.33 s estimate 108.27 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 351 of 620, takes 136.13 s estimate 104.33 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 361 of 620, takes 139.91 s estimate 100.38 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 371 of 620, takes 143.73 s estimate 96.46 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 381 of 620, takes 147.53 s estimate 92.55 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 391 of 620, takes 150.94 s estimate 88.40 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 401 of 620, takes 154.45 s estimate 84.35 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 411 of 620, takes 158.21 s estimate 80.45 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 421 of 620, takes 162.00 s estimate 76.58 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 431 of 620, takes 166.01 s estimate 72.80 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 441 of 620, takes 169.74 s estimate 68.90 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 451 of 620, takes 173.44 s estimate 64.99 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 461 of 620, takes 177.17 s estimate 61.11 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 471 of 620, takes 180.92 s estimate 57.23 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 481 of 620, takes 184.71 s estimate 53.38 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 491 of 620, takes 188.44 s estimate 49.51 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 501 of 620, takes 192.26 s estimate 45.67 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 511 of 620, takes 195.97 s estimate 41.80 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 521 of 620, takes 199.70 s estimate 37.95 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 531 of 620, takes 203.47 s estimate 34.10 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 541 of 620, takes 207.27 s estimate 30.27 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 551 of 620, takes 211.04 s estimate 26.43 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 561 of 620, takes 214.78 s estimate 22.59 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 571 of 620, takes 218.52 s estimate 18.75 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 581 of 620, takes 222.30 s estimate 14.92 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 591 of 620, takes 226.06 s estimate 11.09 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 601 of 620, takes 229.81 s estimate 7.27 s to finish\n",
      "Imputing exp_zinb3_nbr_5_impute: 611 of 620, takes 233.55 s estimate 3.44 s to finish\n",
      "finish imputing, used 236.95 s\n"
     ]
    }
   ],
   "source": [
    "higashi_model.train_for_imputation_with_nbr()\n",
    "higashi_model.impute_with_nbr()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32faa687",
   "metadata": {},
   "source": [
    "### 5. Visulizing embedding results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d22cac0a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(620, 64)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize embedding results\n",
    "cell_embeddings = higashi_model.fetch_cell_embeddings()\n",
    "print (cell_embeddings.shape)\n",
    "\n",
    "from umap import UMAP\n",
    "from sklearn.decomposition import PCA\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "cell_type = higashi_model.label_info['cell type']\n",
    "fig = plt.figure(figsize=(14, 5))\n",
    "ax = plt.subplot(1, 2, 1)\n",
    "vec = PCA(n_components=2).fit_transform(cell_embeddings)\n",
    "sns.scatterplot(x=vec[:, 0], y=vec[:, 1], hue=cell_type, ax=ax, s=6, linewidth=0)\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "labels, handles = zip(*sorted(zip(labels, handles), key=lambda t: t[0]))\n",
    "ax.legend(handles=handles, labels=labels, bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0., ncol=1)\n",
    "ax = plt.subplot(1, 2, 2)\n",
    "vec = UMAP(n_components=2).fit_transform(cell_embeddings)\n",
    "sns.scatterplot(x=vec[:, 0], y=vec[:, 1], hue=cell_type, ax=ax, s=6, linewidth=0)\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "labels, handles = zip(*sorted(zip(labels, handles), key=lambda t: t[0]))\n",
    "ax.legend(handles=handles, labels=labels, bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0., ncol=1)\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f116199",
   "metadata": {},
   "source": [
    "### 6. Visualizing imputation results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "1320110b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x720 with 15 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "count = 0\n",
    "fig = plt.figure(figsize=(6, 2*5))\n",
    "for id_ in np.random.randint(0, 620, 5):\n",
    "    ori, nbr0, nbr5 = higashi_model.fetch_map(\"chr3\", id_)\n",
    "    count += 1\n",
    "    ax = plt.subplot(5, 3, count * 3 - 2)\n",
    "    ax.imshow(ori.toarray(), cmap='Reds', vmin=0.0, vmax=np.quantile(ori.data, 0.6))\n",
    "    ax.set_xticks([], [])\n",
    "    ax.set_yticks([], [])\n",
    "    if count == 1:\n",
    "        ax.set_title(\"raw\")\n",
    "    \n",
    "    ax = plt.subplot(5, 3, count * 3 - 1)\n",
    "    ax.imshow(nbr0.toarray(), cmap='Reds', vmin=0.0, vmax=np.quantile(nbr0.data, 0.95))\n",
    "    ax.set_xticks([], [])\n",
    "    ax.set_yticks([], [])\n",
    "    if count == 1:\n",
    "        ax.set_title(\"higashi, k=0\")\n",
    "    \n",
    "    ax = plt.subplot(5, 3, count * 3)\n",
    "    ax.imshow(nbr5.toarray(), cmap='Reds', vmin=0.0, vmax=np.quantile(nbr5.data, 0.95))\n",
    "    ax.set_xticks([], [])\n",
    "    ax.set_yticks([], [])\n",
    "    if count == 1:\n",
    "        ax.set_title(\"higashi, k=5\")\n",
    "    \n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "286a1767",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}