Comment_Toxicity.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8ee9ee93-509d-448f-9a12-d531d53d211c",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import tensorflow as tf\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8ae52396-5b96-4da0-973b-cbc4c6735ca7",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv('train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c3d66bb3-6ba0-4ffb-914b-592e62cf4757",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>comment_text</th>\n",
       "      <th>toxic</th>\n",
       "      <th>severe_toxic</th>\n",
       "      <th>obscene</th>\n",
       "      <th>threat</th>\n",
       "      <th>insult</th>\n",
       "      <th>identity_hate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0000997932d777bf</td>\n",
       "      <td>Explanation\\nWhy the edits made under my usern...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>000103f0d9cfb60f</td>\n",
       "      <td>D'aww! He matches this background colour I'm s...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>000113f07ec002fd</td>\n",
       "      <td>Hey man, I'm really not trying to edit war. It...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0001b41b1c6bb37e</td>\n",
       "      <td>\"\\nMore\\nI can't make any real suggestions on ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0001d958c54c6e35</td>\n",
       "      <td>You, sir, are my hero. Any chance you remember...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 id                                       comment_text  toxic  \\\n",
       "0  0000997932d777bf  Explanation\\nWhy the edits made under my usern...      0   \n",
       "1  000103f0d9cfb60f  D'aww! He matches this background colour I'm s...      0   \n",
       "2  000113f07ec002fd  Hey man, I'm really not trying to edit war. It...      0   \n",
       "3  0001b41b1c6bb37e  \"\\nMore\\nI can't make any real suggestions on ...      0   \n",
       "4  0001d958c54c6e35  You, sir, are my hero. Any chance you remember...      0   \n",
       "\n",
       "   severe_toxic  obscene  threat  insult  identity_hate  \n",
       "0             0        0       0       0              0  \n",
       "1             0        0       0       0              0  \n",
       "2             0        0       0       0              0  \n",
       "3             0        0       0       0              0  \n",
       "4             0        0       0       0              0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7638e282-bb1f-437b-8180-939013a4d59f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"Hey man, I'm really not trying to edit war. It's just that this guy is constantly removing relevant information and talking to me through edits instead of my talk page. He seems to care more about the formatting than the actual info.\""
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[2]['comment_text']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d22e4f70-22a2-4e40-b3d4-03f8d18b0352",
   "metadata": {},
   "source": [
    "## Preprocess"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ee58a603-451f-4668-9715-0c2f985e2eea",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Package                           Version\n",
      "--------------------------------- ---------------\n",
      "absl-py                           2.0.0\n",
      "aiofiles                          22.1.0\n",
      "albumentations                    1.3.1\n",
      "altair                            5.0.1\n",
      "annotated-types                   0.6.0\n",
      "anyio                             4.2.0\n",
      "argon2-cffi                       23.1.0\n",
      "argon2-cffi-bindings              21.2.0\n",
      "arrow                             1.3.0\n",
      "asttokens                         2.4.1\n",
      "astunparse                        1.6.3\n",
      "async-lru                         2.0.4\n",
      "attrs                             23.2.0\n",
      "autoviz                           0.1.806\n",
      "Babel                             2.14.0\n",
      "beautifulsoup4                    4.12.2\n",
      "bleach                            6.1.0\n",
      "bokeh                             2.4.3\n",
      "Bottleneck                        1.3.5\n",
      "Brotli                            1.0.9\n",
      "cached-property                   1.5.2\n",
      "cachetools                        5.3.2\n",
      "certifi                           2023.11.17\n",
      "cffi                              1.16.0\n",
      "chardet                           5.2.0\n",
      "charset-normalizer                2.0.4\n",
      "click                             8.1.7\n",
      "colorama                          0.4.6\n",
      "colorcet                          3.0.1\n",
      "coloredlogs                       15.0.1\n",
      "comm                              0.2.1\n",
      "contourpy                         1.2.0\n",
      "cryptography                      41.0.7\n",
      "cycler                            0.11.0\n",
      "dacite                            1.8.1\n",
      "debugpy                           1.8.0\n",
      "decorator                         5.1.1\n",
      "defusedxml                        0.7.1\n",
      "emoji                             2.10.1\n",
      "entrypoints                       0.4\n",
      "et-xmlfile                        1.1.0\n",
      "exceptiongroup                    1.2.0\n",
      "executing                         2.0.1\n",
      "fastapi                           0.103.0\n",
      "fastjsonschema                    2.19.1\n",
      "ffmpy                             0.3.0\n",
      "filelock                          3.13.1\n",
      "flatbuffers                       23.5.26\n",
      "fonttools                         4.25.0\n",
      "fqdn                              1.5.1\n",
      "fsspec                            2023.10.0\n",
      "gast                              0.4.0\n",
      "gdown                             5.1.0\n",
      "gmpy2                             2.1.2\n",
      "google-auth                       2.26.1\n",
      "google-auth-oauthlib              0.4.6\n",
      "google-pasta                      0.2.0\n",
      "gradio                            4.16.0\n",
      "gradio_client                     0.8.1\n",
      "grpcio                            1.60.0\n",
      "h11                               0.14.0\n",
      "h2                                4.0.0\n",
      "h5py                              3.10.0\n",
      "holoviews                         1.14.9\n",
      "hpack                             4.0.0\n",
      "htmlmin                           0.1.12\n",
      "httpcore                          1.0.2\n",
      "httpx                             0.26.0\n",
      "huggingface_hub                   0.20.2\n",
      "humanfriendly                     10.0\n",
      "hvplot                            0.7.3\n",
      "hyperframe                        6.0.1\n",
      "idna                              3.6\n",
      "ImageHash                         4.3.1\n",
      "imageio                           2.33.1\n",
      "imgviz                            1.7.5\n",
      "importlib-metadata                7.0.1\n",
      "importlib-resources               6.1.1\n",
      "ipykernel                         6.28.0\n",
      "ipython                           8.18.1\n",
      "ipython-genutils                  0.2.0\n",
      "ipywidgets                        8.1.1\n",
      "isoduration                       20.11.0\n",
      "jedi                              0.19.1\n",
      "Jinja2                            3.1.2\n",
      "joblib                            1.3.2\n",
      "json5                             0.9.14\n",
      "jsonpointer                       2.4\n",
      "jsonschema                        4.20.0\n",
      "jsonschema-specifications         2023.12.1\n",
      "jupyter                           1.0.0\n",
      "jupyter_client                    7.4.9\n",
      "jupyter-console                   6.6.3\n",
      "jupyter-contrib-core              0.4.0\n",
      "jupyter-contrib-nbextensions      0.7.0\n",
      "jupyter_core                      5.7.0\n",
      "jupyter-events                    0.9.0\n",
      "jupyter-highlight-selected-word   0.2.0\n",
      "jupyter-latex-envs                1.4.6\n",
      "jupyter-lsp                       2.2.1\n",
      "jupyter-nbextensions-configurator 0.6.1\n",
      "jupyter_server                    2.12.2\n",
      "jupyter_server_terminals          0.5.1\n",
      "jupyterlab                        4.0.10\n",
      "jupyterlab_pygments               0.3.0\n",
      "jupyterlab_server                 2.25.2\n",
      "jupyterlab-widgets                3.0.9\n",
      "keras                             2.10.0\n",
      "Keras-Preprocessing               1.1.2\n",
      "kiwisolver                        1.4.4\n",
      "labelme                           5.4.1\n",
      "lazy_loader                       0.3\n",
      "libclang                          16.0.6\n",
      "llvmlite                          0.41.1\n",
      "lxml                              5.1.0\n",
      "Markdown                          3.5.1\n",
      "markdown-it-py                    2.2.0\n",
      "MarkupSafe                        2.1.3\n",
      "matplotlib                        3.7.4\n",
      "matplotlib-inline                 0.1.6\n",
      "mdurl                             0.1.0\n",
      "mediapipe                         0.10.9\n",
      "mistune                           3.0.2\n",
      "mkl-fft                           1.3.8\n",
      "mkl-random                        1.2.4\n",
      "mkl-service                       2.4.0\n",
      "mpmath                            1.3.0\n",
      "multimethod                       1.11\n",
      "munkres                           1.1.4\n",
      "mysql-connector-python            8.3.0\n",
      "natsort                           8.4.0\n",
      "nbclassic                         1.0.0\n",
      "nbclient                          0.8.0\n",
      "nbconvert                         7.14.0\n",
      "nbformat                          5.9.2\n",
      "nest-asyncio                      1.5.8\n",
      "networkx                          3.1\n",
      "nltk                              3.8.1\n",
      "notebook                          6.5.6\n",
      "notebook_shim                     0.2.3\n",
      "numba                             0.58.1\n",
      "numexpr                           2.8.7\n",
      "numpy                             1.24.4\n",
      "oauthlib                          3.2.2\n",
      "onnxruntime                       1.17.0\n",
      "opencv-contrib-python             4.9.0.80\n",
      "opencv-python                     4.9.0.80\n",
      "opencv-python-headless            4.9.0.80\n",
      "openpyxl                          3.1.2\n",
      "opt-einsum                        3.3.0\n",
      "orjson                            3.9.10\n",
      "overrides                         7.4.0\n",
      "packaging                         23.2\n",
      "pandas                            2.1.4\n",
      "pandas-dq                         1.29\n",
      "pandas-profiling                  3.6.6\n",
      "pandocfilters                     1.5.0\n",
      "panel                             0.14.4\n",
      "param                             1.13.0\n",
      "parso                             0.8.3\n",
      "patsy                             0.5.6\n",
      "phik                              0.12.4\n",
      "pickleshare                       0.7.5\n",
      "Pillow                            10.0.1\n",
      "pip                               23.3.1\n",
      "pkgutil_resolve_name              1.3.10\n",
      "platformdirs                      4.1.0\n",
      "plotly                            5.18.0\n",
      "prometheus-client                 0.19.0\n",
      "prompt-toolkit                    3.0.42\n",
      "protobuf                          3.19.6\n",
      "psutil                            5.9.7\n",
      "pure-eval                         0.2.2\n",
      "pyamg                             5.0.1\n",
      "pyarrow                           15.0.0\n",
      "pyasn1                            0.5.1\n",
      "pyasn1-modules                    0.3.0\n",
      "pycparser                         2.21\n",
      "pyct                              0.5.0\n",
      "pydantic                          2.5.3\n",
      "pydantic_core                     2.14.6\n",
      "pydub                             0.25.1\n",
      "Pygments                          2.17.2\n",
      "pyOpenSSL                         23.2.0\n",
      "pyparsing                         3.0.9\n",
      "PyQt5                             5.15.10\n",
      "PyQt5-Qt5                         5.15.2\n",
      "PyQt5-sip                         12.13.0\n",
      "pyreadline3                       3.4.1\n",
      "PySocks                           1.7.1\n",
      "python-dateutil                   2.8.2\n",
      "python-json-logger                2.0.7\n",
      "python-multipart                  0.0.6\n",
      "pytz                              2023.3.post1\n",
      "pyviz_comms                       3.0.1\n",
      "PyWavelets                        1.5.0\n",
      "pywin32                           306\n",
      "pywinpty                          2.0.12\n",
      "PyYAML                            6.0.1\n",
      "pyzmq                             24.0.1\n",
      "qtconsole                         5.5.1\n",
      "QtPy                              2.4.1\n",
      "qudida                            0.0.4\n",
      "referencing                       0.32.1\n",
      "regex                             2023.12.25\n",
      "requests                          2.31.0\n",
      "requests-oauthlib                 1.3.1\n",
      "rfc3339-validator                 0.1.4\n",
      "rfc3986-validator                 0.1.1\n",
      "rich                              13.3.5\n",
      "rpds-py                           0.16.2\n",
      "rsa                               4.9\n",
      "ruff                              0.2.0\n",
      "scikit-image                      0.22.0\n",
      "scikit-learn                      1.3.2\n",
      "scipy                             1.11.4\n",
      "seaborn                           0.12.2\n",
      "semantic-version                  2.10.0\n",
      "Send2Trash                        1.8.2\n",
      "setuptools                        68.2.2\n",
      "shellingham                       1.5.0\n",
      "six                               1.16.0\n",
      "sniffio                           1.3.0\n",
      "sounddevice                       0.4.6\n",
      "soupsieve                         2.5\n",
      "stack-data                        0.6.2\n",
      "starlette                         0.27.0\n",
      "statsmodels                       0.14.1\n",
      "sympy                             1.12\n",
      "tangled-up-in-unicode             0.2.0\n",
      "tenacity                          8.2.3\n",
      "tensorboard                       2.10.1\n",
      "tensorboard-data-server           0.6.1\n",
      "tensorboard-plugin-wit            1.8.1\n",
      "tensorflow                        2.10.1\n",
      "tensorflow-estimator              2.10.0\n",
      "tensorflow-io                     0.31.0\n",
      "tensorflow-io-gcs-filesystem      0.31.0\n",
      "termcolor                         2.4.0\n",
      "terminado                         0.18.0\n",
      "textblob                          0.18.0.post0\n",
      "threadpoolctl                     3.2.0\n",
      "tifffile                          2023.12.9\n",
      "tinycss2                          1.2.1\n",
      "tomli                             2.0.1\n",
      "tomlkit                           0.12.0\n",
      "toolz                             0.12.0\n",
      "torch                             2.2.0\n",
      "torchaudio                        2.2.0\n",
      "torchvision                       0.17.0\n",
      "tornado                           6.3.3\n",
      "tqdm                              4.65.0\n",
      "traitlets                         5.14.1\n",
      "typeguard                         4.1.5\n",
      "typer                             0.9.0\n",
      "types-python-dateutil             2.8.19.20240106\n",
      "typing_extensions                 4.9.0\n",
      "typing-utils                      0.1.0\n",
      "tzdata                            2023.3\n",
      "uri-template                      1.3.0\n",
      "urllib3                           1.26.18\n",
      "uvicorn                           0.20.0\n",
      "visions                           0.7.5\n",
      "wcwidth                           0.2.13\n",
      "webcolors                         1.13\n",
      "webencodings                      0.5.1\n",
      "websocket-client                  1.7.0\n",
      "websockets                        10.4\n",
      "Werkzeug                          3.0.1\n",
      "wheel                             0.41.2\n",
      "widgetsnbextension                4.0.9\n",
      "win-inet-pton                     1.1.0\n",
      "wordcloud                         1.9.3\n",
      "wrapt                             1.16.0\n",
      "xgboost                           1.6.2\n",
      "xlrd                              2.0.1\n",
      "ydata-profiling                   4.6.4\n",
      "zipp                              3.17.0\n"
     ]
    }
   ],
   "source": [
    "!pip list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3f4550f2-44ea-4ff1-9633-e24863527e04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "print(tf.config.list_physical_devices('GPU'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b39d2c54-524a-48a8-b9fd-92ee41261810",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers import TextVectorization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3090c51f-db77-4044-b9fb-30d79ea66341",
   "metadata": {},
   "outputs": [],
   "source": [
    "x=df[\"comment_text\"]\n",
    "y=df[df.columns[2:]].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "31d8d254-45d1-4181-b288-f246db35e040",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'comment_text', 'toxic', 'severe_toxic', 'obscene', 'threat',\n",
       "       'insult', 'identity_hate'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0a6f0e34-0880-4714-a7ef-0f2a203c2bf3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         Explanation\\nWhy the edits made under my usern...\n",
       "1         D'aww! He matches this background colour I'm s...\n",
       "2         Hey man, I'm really not trying to edit war. It...\n",
       "3         \"\\nMore\\nI can't make any real suggestions on ...\n",
       "4         You, sir, are my hero. Any chance you remember...\n",
       "                                ...                        \n",
       "159566    \":::::And for the second time of asking, when ...\n",
       "159567    You should be ashamed of yourself \\n\\nThat is ...\n",
       "159568    Spitzer \\n\\nUmm, theres no actual article for ...\n",
       "159569    And it looks like it was actually you who put ...\n",
       "159570    \"\\nAnd ... I really don't think you understand...\n",
       "Name: comment_text, Length: 159571, dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3e4710c6-7bf9-417b-ad21-dcf64dc7c60f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       ...,\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0]], dtype=int64)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4d545805-499d-4728-be21-997699fc0e17",
   "metadata": {},
   "outputs": [],
   "source": [
    "MAX_FEATURES=20000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "9337c062-952b-4b11-bcd1-758952ae8806",
   "metadata": {},
   "outputs": [],
   "source": [
    "vectorizer = TextVectorization(max_tokens=MAX_FEATURES,\n",
    "                               output_sequence_length=1800,\n",
    "                               output_mode='int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "1eb42f41-2f61-4fb1-9e21-dd08f1af04aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "vectorizer.adapt(x.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "04605d4e-da1d-491c-8a1c-7e7afb85ca50",
   "metadata": {},
   "outputs": [],
   "source": [
    "vectorized_text=vectorizer(x.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f465f7e2-f711-497b-bb4b-e51193859f72",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(159571, 1800), dtype=int64, numpy=\n",
       "array([[ 643,   76,    2, ...,    0,    0,    0],\n",
       "       [   1,   54, 2506, ...,    0,    0,    0],\n",
       "       [ 425,  440,   70, ...,    0,    0,    0],\n",
       "       ...,\n",
       "       [   1, 7329,  383, ...,    0,    0,    0],\n",
       "       [   5,   12,  533, ...,    0,    0,    0],\n",
       "       [   5,    8,  130, ...,    0,    0,    0]], dtype=int64)>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectorized_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b192ac0d-0950-47e8-985d-ca85ad310a35",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = tf.data.Dataset.from_tensor_slices((vectorized_text, y))\n",
    "dataset = dataset.cache()\n",
    "dataset = dataset.shuffle(160000)\n",
    "dataset = dataset.batch(16)\n",
    "dataset = dataset.prefetch(8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "41c8fd13-1743-43c3-b48b-e6915d9222e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = dataset.take(int(len(dataset)*.7))\n",
    "val = dataset.skip(int(len(dataset)*.7)).take(int(len(dataset)*.2))\n",
    "test = dataset.skip(int(len(dataset)*.9)).take(int(len(dataset)*.1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "272197f7-adf4-4256-a0ed-04c81284ddd5",
   "metadata": {},
   "source": [
    "## Sequence Model Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f6760988-f546-4345-ad8b-e42b495ff4c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import LSTM, Dropout, Bidirectional, Dense, Embedding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "c7c073ae-3b7b-4660-8ffc-a81269bb1534",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "#  embedding layer \n",
    "model.add(Embedding(MAX_FEATURES+1, 32))\n",
    "# Bidirectional LSTM Layer\n",
    "model.add(Bidirectional(LSTM(32, activation='tanh')))\n",
    "#  Fully connected layers\n",
    "model.add(Dense(128, activation='relu'))\n",
    "model.add(Dense(256, activation='relu'))\n",
    "model.add(Dense(128, activation='relu'))\n",
    "# Final layer \n",
    "model.add(Dense(6, activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "c6131463-a535-497f-ad62-dd5418036e62",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='BinaryCrossentropy', optimizer='Adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "bdbc040d-1e96-4bbe-98cb-4d1c82037d76",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding (Embedding)       (None, None, 32)          640032    \n",
      "                                                                 \n",
      " bidirectional (Bidirectiona  (None, 64)               16640     \n",
      " l)                                                              \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               8320      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 256)               33024     \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 128)               32896     \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 6)                 774       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 731,686\n",
      "Trainable params: 731,686\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "8a01bb33-9c82-4c6b-bdc5-a84651bb4faf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "6981/6981 [==============================] - 1240s 177ms/step - loss: 0.0628 - val_loss: 0.0480\n",
      "Epoch 2/10\n",
      "6981/6981 [==============================] - 1322s 189ms/step - loss: 0.0483 - val_loss: 0.0441\n",
      "Epoch 3/10\n",
      "6981/6981 [==============================] - 1252s 179ms/step - loss: 0.0444 - val_loss: 0.0424\n",
      "Epoch 4/10\n",
      "6981/6981 [==============================] - 1206s 173ms/step - loss: 0.0412 - val_loss: 0.0370\n",
      "Epoch 5/10\n",
      "6981/6981 [==============================] - 1321s 189ms/step - loss: 0.0379 - val_loss: 0.0339\n",
      "Epoch 6/10\n",
      "6981/6981 [==============================] - 1183s 169ms/step - loss: 0.0355 - val_loss: 0.0316\n",
      "Epoch 7/10\n",
      "6981/6981 [==============================] - 1438s 206ms/step - loss: 0.0330 - val_loss: 0.0293\n",
      "Epoch 8/10\n",
      "6981/6981 [==============================] - 1161s 166ms/step - loss: 0.0307 - val_loss: 0.0285\n",
      "Epoch 9/10\n",
      "6981/6981 [==============================] - 1204s 172ms/step - loss: 0.0291 - val_loss: 0.0259\n",
      "Epoch 10/10\n",
      "6981/6981 [==============================] - 1286s 184ms/step - loss: 0.0264 - val_loss: 0.0236\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(train, epochs=10, validation_data=val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "0e9f6474-c703-41b5-bd30-3e99ccc89f97",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': [0.06278979033231735,\n",
       "  0.04830821231007576,\n",
       "  0.04440246522426605,\n",
       "  0.04121015593409538,\n",
       "  0.03785306587815285,\n",
       "  0.03552129864692688,\n",
       "  0.03299447149038315,\n",
       "  0.030697394162416458,\n",
       "  0.02911706268787384,\n",
       "  0.026411259546875954],\n",
       " 'val_loss': [0.048018988221883774,\n",
       "  0.04406905546784401,\n",
       "  0.042407866567373276,\n",
       "  0.03697800636291504,\n",
       "  0.03387845680117607,\n",
       "  0.031604524701833725,\n",
       "  0.0292844045907259,\n",
       "  0.02852766588330269,\n",
       "  0.025872504338622093,\n",
       "  0.023609673604369164]}"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history.history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "987c82f7-7e52-4f46-87f7-c31502ec98b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "e2b234ea-9f6d-4037-a10b-f51cad2d6ab2",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 800x500 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,5))\n",
    "pd.DataFrame(history.history).plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7f09c2a-6231-4c8a-9196-899c822e7147",
   "metadata": {},
   "source": [
    "## Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "be7fa73d-1cdd-4338-a413-c708e94e2776",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_text = vectorizer('You freaking suck! I am going to bit you.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "5c93e426-2115-4747-a321-0f2064286e2e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(1800,), dtype=int64, numpy=array([   7, 7113,  396, ...,    0,    0,    0], dtype=int64)>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "input_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "83619237-1063-41b4-975c-1bfa6f3f9d3f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['toxic', 'severe_toxic', 'obscene', 'threat', 'insult',\n",
       "       'identity_hate'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns[2:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "dd8e87b8-6584-4f0b-8169-1f1d8c05979a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 1s 569ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.99719256, 0.08500203, 0.95627314, 0.00187306, 0.95939726,\n",
       "        0.03799162]], dtype=float32)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict(np.expand_dims(input_text,0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "19c00a6a-99de-4b6b-8cf2-5cb89f6dc224",
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_X, batch_y = test.as_numpy_iterator().next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "2d3daf99-dbac-4f73-bf01-fcd81ccc930b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 177ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 1, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0],\n",
       "       [0, 0, 0, 0, 0, 0]])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(model.predict(batch_X) > 0.5).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "11b29e86-7223-433d-8189-b5c7c8fc9793",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 172ms/step\n"
     ]
    }
   ],
   "source": [
    "res=model.predict(batch_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "dd8add0b-1d63-403a-8958-a62c5d4ef236",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16, 6)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "633ca92e-b542-457b-8e33-37dc5ccf9fd2",
   "metadata": {},
   "source": [
    "## Evaluate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "67529c57-3321-40a2-b174-4e1f66b7ef3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.metrics import Precision, Recall, CategoricalAccuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "739dae41-3bd8-41db-a2c2-f38d51270ef5",
   "metadata": {},
   "outputs": [],
   "source": [
    "pre=Precision()\n",
    "re=Recall()\n",
    "ca=CategoricalAccuracy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "65ef580d-508a-4c22-b352-6a424e1a3121",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 169ms/step\n",
      "1/1 [==============================] - 0s 167ms/step\n",
      "1/1 [==============================] - 0s 129ms/step\n",
      "1/1 [==============================] - 0s 115ms/step\n",
      "1/1 [==============================] - 0s 120ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 75ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 118ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 111ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 108ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 117ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 75ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 112ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 112ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 110ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 74ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 109ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 118ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 124ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 118ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 111ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 110ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 111ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 114ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 108ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 108ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 118ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 126ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 111ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 115ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 126ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 117ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 111ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 138ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 108ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 81ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 115ms/step\n",
      "1/1 [==============================] - 0s 109ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 103ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 133ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 102ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 112ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 126ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 114ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 79ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 121ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 111ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 121ms/step\n",
      "1/1 [==============================] - 0s 113ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 99ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 122ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 88ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 84ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 78ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 86ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 95ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 75ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 98ms/step\n",
      "1/1 [==============================] - 0s 85ms/step\n",
      "1/1 [==============================] - 0s 100ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 80ms/step\n",
      "1/1 [==============================] - 0s 77ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 87ms/step\n",
      "1/1 [==============================] - 0s 96ms/step\n",
      "1/1 [==============================] - 0s 106ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 76ms/step\n",
      "1/1 [==============================] - 0s 82ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 101ms/step\n",
      "1/1 [==============================] - 0s 104ms/step\n",
      "1/1 [==============================] - 0s 107ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 83ms/step\n",
      "1/1 [==============================] - 0s 89ms/step\n",
      "1/1 [==============================] - 0s 105ms/step\n"
     ]
    }
   ],
   "source": [
    "for batch in test.as_numpy_iterator(): \n",
    "    # Unpack the batch \n",
    "    X_true, y_true = batch\n",
    "    # Make a prediction \n",
    "    yhat = model.predict(X_true)\n",
    "    \n",
    "    # Flatten the predictions\n",
    "    y_true = y_true.flatten()\n",
    "    yhat = yhat.flatten()\n",
    "    \n",
    "    pre.update_state(y_true, yhat)\n",
    "    re.update_state(y_true, yhat)\n",
    "    ca.update_state(y_true, yhat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "a592e846-8ee0-4c59-9691-5a92961def41",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.8819800615310669, Recall:0.8543859720230103, Accuracy:0.47442325949668884\n"
     ]
    }
   ],
   "source": [
    "print(f'Precision: {pre.result().numpy()}, Recall:{re.result().numpy()}, Accuracy:{ca.result().numpy()}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4d471fe-2524-40c9-a884-c3ec5e8c1e75",
   "metadata": {},
   "source": [
    "## Test "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "28efc844-2777-4f41-a805-e202b7410b15",
   "metadata": {},
   "outputs": [],
   "source": [
    "import gradio as gr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "464ef15c-cbb6-4799-9352-961b3110f24d",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('toxicity.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "f7584d97-10cd-44a7-98f5-daaae2057316",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "4dcbde83-dc26-4a44-b07d-edefd1f0495f",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_text = vectorizer('hey i freaken hate you!')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "082024d3-0ca2-4ac4-bfea-79d27fd6bd05",
   "metadata": {},
   "outputs": [],
   "source": [
    "def score_comment(comment):\n",
    "    vectorized_comment = vectorizer([comment])\n",
    "    results = model.predict(vectorized_comment)\n",
    "    \n",
    "    text = ''\n",
    "    for idx, col in enumerate(df.columns[2:]):\n",
    "        text += '{}: {}\\n'.format(col, results[0][idx]>0.5)\n",
    "    \n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "ddd92f02-76f8-438a-b4a4-5fe884cfaa17",
   "metadata": {},
   "outputs": [],
   "source": [
    "interface = gr.Interface(fn=score_comment, \n",
    "                         inputs=gr.Textbox(lines=2, placeholder=\"Comment here\"),\n",
    "                        outputs='text')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "290364f4-6c43-4566-a032-3f6d0c9d1b5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on local URL:  http://127.0.0.1:7860\n",
      "\n",
      "To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div><iframe src=\"http://127.0.0.1:7860/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 151ms/step\n"
     ]
    }
   ],
   "source": [
    "interface.launch()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "99897eec-2acf-4dd5-a0b0-69c077847ae9",
   "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}