U5.09- [TALLER] ELMo - Name Entity Recognition.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 9180,
     "status": "ok",
     "timestamp": 1588456543098,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "nu-c4z6VuXG2",
    "outputId": "6333c3e4-ef76-407b-aaf0-fd91fd725ca0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "replicating local resources\n"
     ]
    }
   ],
   "source": [
    "!wget --no-cache -O init.py -q https://raw.githubusercontent.com/rramosp/2020.deeplearning/master/init.py\n",
    "from init import init; init(force_download=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 10053,
     "status": "ok",
     "timestamp": 1588456546063,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "BqHaLt-PuniZ",
    "outputId": "35b8dbc7-70e5-4dad-9a46-29f0bb0299a2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "init.py  local\t__pycache__  sample_data\n"
     ]
    }
   ],
   "source": [
    "!dir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 15194,
     "status": "ok",
     "timestamp": 1588456551542,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "IH1ocrWtuXG6",
    "outputId": "4e9237b3-77f5-4f71-eee1-fed3c19c13a3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "setting tensorflow version in colab\n",
      "TensorFlow 1.x selected.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'1.15.2'"
      ]
     },
     "execution_count": 3,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import sys\n",
    "if 'google.colab' in sys.modules:\n",
    "    print (\"setting tensorflow version in colab\")\n",
    "    %tensorflow_version 1.x\n",
    "import tensorflow as tf\n",
    "tf.__version__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "yUTaRdfwuXG8"
   },
   "source": [
    "### ELMo: Embeddings from Language Models\n",
    "\n",
    "Word embeddings such as word2vec or GloVe provides an exact meaning to words. Eventhough they provided a great improvement to many NLP task, such \"constant\" meaning was a major drawback of this word embeddings as the meaning of words changes based on context, and thus this wasn't the best option for Language Modelling.\n",
    "\n",
    "For instance, after we train word2vec/Glove on a corpus we get as output one vector representation for, say the word **cell**. So even if we had a sentence like \"He went to the prison **cell** with his **cell** phone to extract blood **cell** samples from inmates\", where the word **cell** has different meanings based on the sentence context, these models just collapse them all into one vector for **cell** in their output [source](https://www.quora.com/What-are-the-main-differences-between-the-word-embeddings-of-ELMo-BERT-Word2vec-and-GloVe).\n",
    "\n",
    "Unlike most widely used word embeddings ELMo word representations are functions of the entire input sentence, instead of the single word. They are computed on top of two-layer Bidirectional Language Models (biLMs) with character convolutions, as a linear function of the internal network states. Therefore, the same word can have different word vectors under different contexts.\n",
    "\n",
    "[Deep contextualized word representations](https://arxiv.org/pdf/1802.05365.pdf)\n",
    "\n",
    "[Character-level Convolutional Networks](https://papers.nips.cc/paper/5782-character-level-convolutional-networks-for-text-classification.pdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 557
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4176,
     "status": "ok",
     "timestamp": 1588379330908,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "7lsKUGopuXG9",
    "outputId": "3817bdd7-ec04-4320-eb0b-be8306c3783d"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 5,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "#Image(filename='local/imgs/ELMo.gif', width=1200)\n",
    "Image(open('local/imgs/ELMo.gif','rb').read())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_RCGWtB1uXHA"
   },
   "source": [
    "[Image taken from here](https://www.analyticsvidhya.com/blog/2019/03/learn-to-use-elmo-to-extract-features-from-text/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "i82xk9FHuXHB"
   },
   "source": [
    "Given $T$ tokens $(x_1,x_2,\\cdots,x_T)$, a forward language model computes the probability of the sequence by modeling the probability of token $x_k$ given the history $(x_1,\\cdots, x_{k-1})$. This formulation has been addressed in the state of the art using many different approach, and more recently including some approximation based on Bidirectional Recurrent Networks.\n",
    "\n",
    "ELMo is inspired in the Language Modelling problem, which has the advantage of being a **self-supervised** task.\n",
    "\n",
    "A practical implication of this difference is that we can use word2vec and Glove vectors trained on a large corpus directly for downstream tasks. All we need is the vectors for the words. There is no need for the model itself that was used to train these vectors.\n",
    "\n",
    "However, in the case of ELMo and BERT (we will see it in a forthcoming lecture), since they are context dependent, we need the model that was used to train the vectors even after training, since the models generate the vectors for a word based on context. [source](https://www.quora.com/What-are-the-main-differences-between-the-word-embeddings-of-ELMo-BERT-Word2vec-and-GloVe)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xSkeYlYuuXHB"
   },
   "source": [
    "## Download the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Z5HzJ0l3uXHB"
   },
   "source": [
    "Let's load ELMo model. This will take some time because the model is over 350 Mb in size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "V34RjciouXHC"
   },
   "outputs": [],
   "source": [
    "import tensorflow_hub as hub\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "GqFLILYduXHE"
   },
   "outputs": [],
   "source": [
    "elmo_model = hub.Module(\"https://tfhub.dev/google/elmo/2\", trainable=True)\n",
    "# In TEnsorFlow 2 the statement should be somethin like this\n",
    "#elmo_model = hub.KerasLayer(\"https://tfhub.dev/google/elmo/2\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16758,
     "status": "ok",
     "timestamp": 1588456560314,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "iFcrxno_uXHG",
    "outputId": "2232bff2-ac71-426d-821a-2067d5c076ff"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Saver not created because there are no variables in the graph to restore\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Saver not created because there are no variables in the graph to restore\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 7, 1024)\n"
     ]
    }
   ],
   "source": [
    "embeddings = elmo_model([\"i like green eggs and ham\",\n",
    "                   \"would you eat them in a box\"],\n",
    "                  signature=\"default\",\n",
    "                  as_dict=True)[\"elmo\"]\n",
    "print(embeddings.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "cNgFAsKduXHI"
   },
   "source": [
    "# Name Entity Recognition (NER)\n",
    "\n",
    "NER is a sequential labeling problem where the aim is to label every word in a sentence of pargraph, according to a list of \"entity\" classes. This is different from part of Speech (POS) tagging, which explains how a word is used in a sentence [For more information about POS](https://medium.com/greyatom/learning-pos-tagging-chunking-in-nlp-85f7f811a8cb). \n",
    "\n",
    "Tipical labels in NER are:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 289
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2251,
     "status": "ok",
     "timestamp": 1588457014778,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "edtD37_juXHJ",
    "outputId": "04a7ffb2-0a00-4eb9-98b6-714f7ba161c7"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package maxent_ne_chunker to\n",
      "[nltk_data]     /home/julian/nltk_data...\n",
      "[nltk_data]   Package maxent_ne_chunker is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['B-FACILITY',\n",
       " 'B-GPE',\n",
       " 'B-GSP',\n",
       " 'B-LOCATION',\n",
       " 'B-ORGANIZATION',\n",
       " 'B-PERSON',\n",
       " 'I-FACILITY',\n",
       " 'I-GPE',\n",
       " 'I-GSP',\n",
       " 'I-LOCATION',\n",
       " 'I-ORGANIZATION',\n",
       " 'I-PERSON',\n",
       " 'O']"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "nltk.download('maxent_ne_chunker')\n",
    "chunker=nltk.data.load(nltk.chunk._MULTICLASS_NE_CHUNKER)\n",
    "sorted(chunker._tagger._classifier.labels())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "09jedUmkuXHL"
   },
   "source": [
    "GPE stands for Geo-Political Entity, GSP stands for Geographical-Social-Political Entity. and there are other definitions of labels that include more classes [Detailed info can be found here](https://www.ldc.upenn.edu/sites/www.ldc.upenn.edu/files/english-entities-guidelines-v6.6.pdf). \n",
    "\n",
    "NER task is commonly viewed as a sequential prediction problem in which we aim at assigning the correct label for each token. Different ways of encoding information in a set of labels make different chunk representation. The two most popular schemes are BIO and BILOU [source](https://natural-language-understanding.fandom.com/wiki/Named_entity_recognition).\n",
    "\n",
    "- BIO stands for Beginning, Inside and Outside (of a text segment).\n",
    "- Similar but more detailed than BIO, BILOU encode the Beginning, the Inside and Last token of multi-token chunks while differentiate them from Unit-length chunks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 421
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1275,
     "status": "ok",
     "timestamp": 1588457018627,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "elatP0ZjuXHL",
    "outputId": "900e657a-92bd-4ec1-ea8c-7ced7d070734"
   },
   "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>Sentence #</th>\n",
       "      <th>Word</th>\n",
       "      <th>Tag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1048563</th>\n",
       "      <td>Sentence: 47958</td>\n",
       "      <td>exploded</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048564</th>\n",
       "      <td>Sentence: 47958</td>\n",
       "      <td>upon</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048565</th>\n",
       "      <td>Sentence: 47958</td>\n",
       "      <td>impact</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048566</th>\n",
       "      <td>Sentence: 47958</td>\n",
       "      <td>.</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048567</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>Indian</td>\n",
       "      <td>B-gpe</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048568</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>forces</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048569</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>said</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048570</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>they</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048571</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>responded</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048572</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>to</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048573</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>the</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048574</th>\n",
       "      <td>Sentence: 47959</td>\n",
       "      <td>attack</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Sentence #       Word    Tag\n",
       "1048563  Sentence: 47958   exploded      O\n",
       "1048564  Sentence: 47958       upon      O\n",
       "1048565  Sentence: 47958     impact      O\n",
       "1048566  Sentence: 47958          .      O\n",
       "1048567  Sentence: 47959     Indian  B-gpe\n",
       "1048568  Sentence: 47959     forces      O\n",
       "1048569  Sentence: 47959       said      O\n",
       "1048570  Sentence: 47959       they      O\n",
       "1048571  Sentence: 47959  responded      O\n",
       "1048572  Sentence: 47959         to      O\n",
       "1048573  Sentence: 47959        the      O\n",
       "1048574  Sentence: 47959     attack      O"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import math\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use(\"ggplot\")\n",
    "\n",
    "data = pd.read_csv(\"local/data/ner_dataset.csv\", encoding=\"latin1\")\n",
    "data = data.drop(['POS'], axis =1)\n",
    "data = data.fillna(method=\"ffill\")\n",
    "data.tail(12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1510,
     "status": "ok",
     "timestamp": 1588457022349,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "015Pz5tkuXHN",
    "outputId": "e45d63a8-babd-4ade-8fc7-87bf86cbf180"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "35179"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "words = set(list(data['Word'].values))\n",
    "words.add('PADword')\n",
    "n_words = len(words)\n",
    "n_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 967,
     "status": "ok",
     "timestamp": 1588457022915,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "RZ6nx-m8uXHQ",
    "outputId": "7518b42e-80f4-499b-c0ce-21ad50f8aee9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tags = list(set(data[\"Tag\"].values))\n",
    "n_tags = len(tags)\n",
    "n_tags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 306
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 559,
     "status": "ok",
     "timestamp": 1588457023238,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "_5SMlsDXuXHS",
    "outputId": "9b2d9927-05d8-40ad-e6b8-4e3bec0ac29e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['I-per',\n",
       " 'B-tim',\n",
       " 'B-gpe',\n",
       " 'B-nat',\n",
       " 'O',\n",
       " 'I-gpe',\n",
       " 'B-art',\n",
       " 'I-eve',\n",
       " 'I-nat',\n",
       " 'B-org',\n",
       " 'I-geo',\n",
       " 'I-art',\n",
       " 'I-org',\n",
       " 'I-tim',\n",
       " 'B-eve',\n",
       " 'B-geo',\n",
       " 'B-per']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7FIXAVV2uXHT"
   },
   "outputs": [],
   "source": [
    "class SentenceGetter(object):\n",
    "    \n",
    "    def __init__(self, data):\n",
    "        self.n_sent = 1\n",
    "        self.data = data\n",
    "        self.empty = False\n",
    "        agg_func = lambda s: [(w, t) for w, t in zip(s[\"Word\"].values.tolist(),s[\"Tag\"].values.tolist())]\n",
    "        self.grouped = self.data.groupby(\"Sentence #\").apply(agg_func)\n",
    "        self.sentences = [s for s in self.grouped]\n",
    "    \n",
    "    def get_next(self):\n",
    "        try:\n",
    "            s = self.grouped[\"Sentence: {}\".format(self.n_sent)]\n",
    "            self.n_sent += 1\n",
    "            return s\n",
    "        except:\n",
    "            return None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6869,
     "status": "ok",
     "timestamp": 1588457039753,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "oEk-6J6IuXHV",
    "outputId": "7f8edd40-3ce8-4ce7-ce11-c793e3fd4a8c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('Thousands', 'O'), ('of', 'O'), ('demonstrators', 'O'), ('have', 'O'), ('marched', 'O'), ('through', 'O'), ('London', 'B-geo'), ('to', 'O'), ('protest', 'O'), ('the', 'O'), ('war', 'O'), ('in', 'O'), ('Iraq', 'B-geo'), ('and', 'O'), ('demand', 'O'), ('the', 'O'), ('withdrawal', 'O'), ('of', 'O'), ('British', 'B-gpe'), ('troops', 'O'), ('from', 'O'), ('that', 'O'), ('country', 'O'), ('.', 'O')]\n"
     ]
    }
   ],
   "source": [
    "getter = SentenceGetter(data)\n",
    "sent = getter.get_next()\n",
    "print(sent)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5662,
     "status": "ok",
     "timestamp": 1588457039756,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "K0UVYWA2uXHX",
    "outputId": "e3fa7f2e-1820-493d-b529-dd9a5589791b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "47959\n"
     ]
    }
   ],
   "source": [
    "sentences = getter.sentences\n",
    "print(len(sentences))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4939,
     "status": "ok",
     "timestamp": 1588457039758,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "JEBSIzsnuXHZ",
    "outputId": "76bd5fbf-327d-468e-c053-f9717b09eada"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "biggest sentence has 104 words\n"
     ]
    }
   ],
   "source": [
    "largest_sen = max(len(sen) for sen in sentences)\n",
    "print('biggest sentence has {} words'.format(largest_sen))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4363,
     "status": "ok",
     "timestamp": 1588457040076,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "iVMCNZ56uXHa",
    "outputId": "7abe2e58-bcaa-4956-def9-7151d63a2efd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "plt.hist([len(sen) for sen in sentences], bins= 50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3469,
     "status": "ok",
     "timestamp": 1588457040078,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "gjXNZY-quXHc",
    "outputId": "5aba085b-4ad6-407e-a713-d52cc801f6a2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14333\n",
      "15\n"
     ]
    }
   ],
   "source": [
    "words2index = {w:i for i,w in enumerate(words)}\n",
    "tags2index = {t:i for i,t in enumerate(tags)}\n",
    "print(words2index['London'])\n",
    "print(tags2index['B-geo'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "GvBG6ZXsE6_-"
   },
   "source": [
    "ELMo embedding represents every word as 1024 feature vector. Therefore, in order to reduce the computational complexity and based on the former histogram, we can truncate the sequences to a maximum length of 50."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "VWwaxjO3uXHe"
   },
   "outputs": [],
   "source": [
    "max_len = 50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ElMo uses a character based embedding, so the padding process requires the use a default word for  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 867
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2020,
     "status": "ok",
     "timestamp": 1588457042366,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "j-vDaeHKuXHg",
    "outputId": "08fbac6c-a75a-413a-d490-be32f3e525f8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Israeli',\n",
       " 'officials',\n",
       " 'say',\n",
       " 'Prime',\n",
       " 'Minister',\n",
       " 'Ariel',\n",
       " 'Sharon',\n",
       " 'will',\n",
       " 'undergo',\n",
       " 'a',\n",
       " 'medical',\n",
       " 'procedure',\n",
       " 'Thursday',\n",
       " 'to',\n",
       " 'close',\n",
       " 'a',\n",
       " 'tiny',\n",
       " 'hole',\n",
       " 'in',\n",
       " 'his',\n",
       " 'heart',\n",
       " 'discovered',\n",
       " 'during',\n",
       " 'treatment',\n",
       " 'for',\n",
       " 'a',\n",
       " 'minor',\n",
       " 'stroke',\n",
       " 'suffered',\n",
       " 'last',\n",
       " 'month',\n",
       " '.',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword',\n",
       " 'PADword']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = [[w[0]for w in s] for s in sentences]\n",
    "new_X = []\n",
    "for seq in X:\n",
    "    new_seq = []\n",
    "    for i in range(max_len):\n",
    "        try:\n",
    "            new_seq.append(seq[i])\n",
    "        except:\n",
    "            new_seq.append(\"PADword\")\n",
    "    new_X.append(new_seq)\n",
    "new_X[15]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 102
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1284,
     "status": "ok",
     "timestamp": 1588457049283,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "DBUoDD4wuXHh",
    "outputId": "5753d7b1-07d5-4b7e-ab59-e31a802ad147"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([ 2,  4,  4, 16,  0,  0,  0,  4,  4,  4,  4,  4,  1,  4,  4,  4,  4,\n",
       "        4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,\n",
       "        4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4],\n",
       "      dtype=int32)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.preprocessing.sequence import pad_sequences\n",
    "y = [[tags2index[w[1]] for w in s] for s in sentences]\n",
    "y = pad_sequences(maxlen=max_len, sequences=y, padding=\"post\", value=tags2index[\"O\"])\n",
    "y[15]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "TYJVZaGkuXHj"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_tr, X_te, y_tr, y_te = train_test_split(new_X, y, test_size=0.1, random_state=2018)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "h3wWkU2zuXHl"
   },
   "outputs": [],
   "source": [
    "batch_size = 32\n",
    "import tensorflow as tf\n",
    "import tensorflow_hub as hub\n",
    "from keras import backend as K\n",
    "sess = tf.Session()\n",
    "K.set_session(sess)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "t8aN0JvtuXHm"
   },
   "outputs": [],
   "source": [
    "sess.run(tf.global_variables_initializer())\n",
    "sess.run(tf.tables_initializer())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ZgNluKywFuzi"
   },
   "source": [
    "This is how we have to define the layer in order to be used in the model (In TF 2.x this should not be necessary [take a look to this guide](https://www.tensorflow.org/hub)):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MzJWb2i6uXHo"
   },
   "outputs": [],
   "source": [
    "def ElmoEmbedding(x):\n",
    "    return elmo_model(inputs={\n",
    "                            \"tokens\": tf.squeeze(tf.cast(x, tf.string)),\n",
    "                            \"sequence_len\": tf.constant(batch_size*[max_len])\n",
    "                      },\n",
    "                      signature=\"tokens\",\n",
    "                      as_dict=True)[\"elmo\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "3sH7WhkZuXHq"
   },
   "outputs": [],
   "source": [
    "from keras.models import Model, Input\n",
    "from keras.layers.merge import add\n",
    "from keras.layers import LSTM, Embedding, Dense, TimeDistributed, Dropout, Bidirectional, Lambda"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "2dUV8HYiGcwd"
   },
   "source": [
    "## Excercise 1: Complete the model arquitecture\n",
    "\n",
    "After the Embedding layer include two Bidirectional LSTM layers with 512 cells each. The second layer must be on top of the first one. As output layer include a Dense layer which takes as input the sum of the two Bidirectional layers output's. Remember that in NER the output must be a sequence of same length  as the input. Use also an Adam optimizer and include accuracy in the metrics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5010,
     "status": "ok",
     "timestamp": 1588457248754,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "kefkZDIKuXHs",
    "outputId": "8aa17201-8da4-4601-8c21-13fee29bb8cc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Saver not created because there are no variables in the graph to restore\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Saver not created because there are no variables in the graph to restore\n"
     ]
    }
   ],
   "source": [
    "input_text = Input(shape=(max_len,), dtype=tf.string)\n",
    "embedding = Lambda(ElmoEmbedding, output_shape=(max_len, 1024))(input_text)\n",
    "....\n",
    "# Complete the model architecture here:\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "mIp44ksIa37_"
   },
   "source": [
    "Split data into training and validation subsets, use both during training to evaluate training and validation accuracies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "IjF1DUEGuXHw"
   },
   "outputs": [],
   "source": [
    "X_tr, X_val = X_tr[:1213*batch_size], X_tr[-135*batch_size:]\n",
    "y_tr, y_val = y_tr[:1213*batch_size], y_tr[-135*batch_size:]\n",
    "y_tr = y_tr.reshape(y_tr.shape[0], y_tr.shape[1], 1)\n",
    "y_val = y_val.reshape(y_val.shape[0], y_val.shape[1], 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "skr8b7fvllLU"
   },
   "source": [
    "Train the model using a batch_size = 32 and 5 epochs. Use verbose=1 to see the evolution of the training process. This is what you should get:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2708364,
     "status": "ok",
     "timestamp": 1588460011504,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "JNS-L6BluXHy",
    "outputId": "5013d7fa-1c8b-467a-d277-c9aafae2414e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 38816 samples, validate on 4320 samples\n",
      "Epoch 1/5\n",
      "38816/38816 [==============================] - 542s 14ms/step - loss: 0.0584 - accuracy: 0.9828 - val_loss: 0.0415 - val_accuracy: 0.9867\n",
      "Epoch 2/5\n",
      "38816/38816 [==============================] - 542s 14ms/step - loss: 0.0413 - accuracy: 0.9867 - val_loss: 0.0329 - val_accuracy: 0.9889\n",
      "Epoch 3/5\n",
      "38816/38816 [==============================] - 542s 14ms/step - loss: 0.0347 - accuracy: 0.9884 - val_loss: 0.0267 - val_accuracy: 0.9907\n",
      "Epoch 4/5\n",
      "38816/38816 [==============================] - 540s 14ms/step - loss: 0.0291 - accuracy: 0.9899 - val_loss: 0.0208 - val_accuracy: 0.9927\n",
      "Epoch 5/5\n",
      "38816/38816 [==============================] - 541s 14ms/step - loss: 0.0242 - accuracy: 0.9914 - val_loss: 0.0169 - val_accuracy: 0.9943\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7f0a2f696a20>"
      ]
     },
     "execution_count": 33,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "yKFgpdvRuXH1"
   },
   "outputs": [],
   "source": [
    "idx2tag = {i: w for w, i in tags2index.items()}\n",
    "\n",
    "def pred2label(pred):\n",
    "    out = []\n",
    "    for pred_i in pred:\n",
    "        out_i = []\n",
    "        for p in pred_i:\n",
    "            p_i = np.argmax(p)\n",
    "            out_i.append(idx2tag[p_i].replace(\"PADword\", \"O\"))\n",
    "        out.append(out_i)\n",
    "    return out\n",
    "\n",
    "def test2label(pred):\n",
    "    out = []\n",
    "    for pred_i in pred:\n",
    "        out_i = []\n",
    "        for p in pred_i:\n",
    "            out_i.append(idx2tag[p].replace(\"PADword\", \"O\"))\n",
    "        out.append(out_i)\n",
    "    return out\n",
    "    \n",
    "pred_labels = pred2label(test_pred)\n",
    "test_labels = test2label(y_te[:149*32])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 48603,
     "status": "ok",
     "timestamp": 1588460087000,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "HnCv4YTKuXHz",
    "outputId": "8e11e62b-f85c-4c83-c023-a992c951f3e6"
   },
   "source": [
    "Use seqeval library: https://github.com/chakki-works/seqeval to estimate F1 score, precision and recall measures. You should see the following results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1288,
     "status": "ok",
     "timestamp": 1588460119383,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "9VjMrGfjuXH3",
    "outputId": "9ded37e0-5cee-4eb9-8ba1-da40c1abaebf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1-score: 82.6%\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 255
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3123,
     "status": "ok",
     "timestamp": 1588460126101,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "kAehxV2fuXH4",
    "outputId": "53778a7c-e792-4f92-9230-8ba4d9764341"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           precision    recall  f1-score   support\n",
      "\n",
      "      geo       0.85      0.90      0.87      3720\n",
      "      per       0.74      0.77      0.76      1677\n",
      "      tim       0.86      0.86      0.86      2148\n",
      "      gpe       0.96      0.94      0.95      1591\n",
      "      org       0.69      0.70      0.70      2061\n",
      "      eve       0.22      0.36      0.27        33\n",
      "      art       0.38      0.10      0.16        49\n",
      "      nat       0.33      0.18      0.24        22\n",
      "\n",
      "micro avg       0.82      0.84      0.83     11301\n",
      "macro avg       0.82      0.84      0.83     11301\n",
      "\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 731
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1322,
     "status": "ok",
     "timestamp": 1588460345452,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "JCtRF6rh3S8I",
    "outputId": "04b31c74-e100-4e26-8991-6e9b6a03bbff"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Word            Pred : (True)\n",
      "==============================\n",
      "Citing         :O     (O)\n",
      "a              :O     (O)\n",
      "draft          :O     (O)\n",
      "report         :O     (O)\n",
      "from           :O     (O)\n",
      "the            :O     (O)\n",
      "U.S.           :B-org (B-org)\n",
      "Government     :I-org (I-org)\n",
      "Accountability :I-org (O)\n",
      "office         :O     (O)\n",
      ",              :O     (O)\n",
      "The            :B-org (B-org)\n",
      "New            :I-org (I-org)\n",
      "York           :I-org (I-org)\n",
      "Times          :I-org (I-org)\n",
      "said           :O     (O)\n",
      "Saturday       :B-tim (B-tim)\n",
      "the            :O     (O)\n",
      "losses         :O     (O)\n",
      "amount         :O     (O)\n",
      "to             :O     (O)\n",
      "between        :O     (O)\n",
      "1,00,000       :O     (O)\n",
      "and            :O     (O)\n",
      "3,00,000       :O     (O)\n",
      "barrels        :O     (O)\n",
      "a              :O     (O)\n",
      "day            :O     (O)\n",
      "of             :O     (O)\n",
      "Iraq           :B-geo (B-geo)\n",
      "'s             :O     (O)\n",
      "declared       :O     (O)\n",
      "oil            :O     (O)\n",
      "production     :O     (O)\n",
      "over           :O     (O)\n",
      "the            :O     (O)\n",
      "past           :B-tim (B-tim)\n",
      "four           :I-tim (I-tim)\n",
      "years          :O     (O)\n",
      ".              :O     (O)\n"
     ]
    }
   ],
   "source": [
    "i = 390\n",
    "p = model.predict(np.array(X_te[i:i+batch_size]))[0]\n",
    "p = np.argmax(p, axis=-1)\n",
    "print(\"{:15} {:5}: ({})\".format(\"Word\", \"Pred\", \"True\"))\n",
    "print(\"=\"*30)\n",
    "for w, true, pred in zip(X_te[i], y_te[i], p):\n",
    "    if w != \"PADword\":\n",
    "        print(\"{:15}:{:5} ({})\".format(w, tags[pred], tags[true]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "fcoPgGwP3aBN"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "GFWesTq_nX0r"
   },
   "source": [
    "## Excercise 2 \n",
    "\n",
    "For the sake of comparison, create a model using a conventional Embedding layer, instead of ELMo model. Use an output dimension of 1024 for the Embedding layer and 10000 words dictionary for tokanization."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You should get the following results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1289,
     "status": "ok",
     "timestamp": 1588464339780,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "JZHWT9Lc1Idv",
    "outputId": "a8cf5e99-394d-4b5f-b7df-cf10220f075c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1-score: 54.5%\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 255
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3147,
     "status": "ok",
     "timestamp": 1588464363463,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "p8y6MFng2kX3",
    "outputId": "881e8ae0-35e2-4709-ed59-3d06425b08ea"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           precision    recall  f1-score   support\n",
      "\n",
      "      geo       0.63      0.52      0.57      3720\n",
      "      per       0.51      0.46      0.48      1677\n",
      "      tim       0.59      0.50      0.54      2148\n",
      "      gpe       0.80      0.68      0.73      1591\n",
      "      org       0.49      0.38      0.42      2061\n",
      "      eve       0.20      0.12      0.15        33\n",
      "      art       0.08      0.02      0.03        49\n",
      "      nat       0.19      0.18      0.19        22\n",
      "\n",
      "micro avg       0.60      0.50      0.55     11301\n",
      "macro avg       0.60      0.50      0.54     11301\n",
      "\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 901
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 962,
     "status": "ok",
     "timestamp": 1588464461346,
     "user": {
      "displayName": "JULIAN DAVID ARIAS LONDOÑO",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjMHGw2psxBqZYpYNBjH-5eTIl5iXTwfxeq1M6EQZcBUgKr67y9o6hdlok-m-QIOfvvMHymbu48oDvqCnWsiujBa5ZnTzc_VuNhaC1f688kmIF2ZBOgqijCLqdf-RwSJKY4OeFmS-rl2nqSwV9TsSa0fPrmcQfugOob9GUbkrqHLhCoIGnLXtN7gmN-ShB8YfKVIDNSLfSzhFAyTrOjHd7vWAfgljQd3PPlolgrr6Z39apBVvVf0VKGgQcjJiwmX-Ds7GmOX0gkMlFgOK0UNAD70w4dwHzLXKcY2bv-LiyTvC6lbJVhzKU6AAICBxzHAiv9XfmPRfe6n0QYPjzQHxOzu8ArMTGPtp1RxlIgPTangmGQsePUC4yE7BC1LcGE6XshwSfkhunTu4TSuBhTI3l32_Tn7btVqwfy5M4mC4O0wqY6jO33BETBQuKYZl8d-esBTna-S0FYYRRNrbd926ueB-lV0HZPH4LoxP1gdD9IQyyerAqajruBBz9Lp2TnMJCiBriDJXHyH0uZChzuoWY3O5HqxB_7qF0C6dm2WQet24O0aeKRTmFQQPWdXYoTFtFm747ynWZcwkLIMoS9PZxt9d2rvqTWcmdb9nu_re9sDkTtOthYPFOeTAFYK0bNPjMNGbpxBWtIcL1shk_cd81C9xqKGsD-a9dLBcUZfK_EFGxzs34uPw6jN7ALQhxALDYVJG0lCFFuWm3ELCCrpyRgLZ5-pMK6tPGn7TO3c3FjlnA4v4lEE_ENIPtDsB5Tb3pl33k=s64",
      "userId": "14990390101324121504"
     },
     "user_tz": 300
    },
    "id": "HmYrg7wQ2psn",
    "outputId": "69dd593d-6d65-4495-fa37-5e87a3ba86cf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Word            Pred : (True)\n",
      "==============================\n",
      "           2327:O     (O)\n",
      "              5:O     (O)\n",
      "           1604:O     (O)\n",
      "            144:O     (O)\n",
      "             21:O     (O)\n",
      "              1:O     (O)\n",
      "             14:O     (B-org)\n",
      "             20:B-geo (I-org)\n",
      "             34:O     (O)\n",
      "           8632:I-org (O)\n",
      "            383:I-org (O)\n",
      "              1:O     (B-org)\n",
      "             49:B-org (I-org)\n",
      "            436:I-org (I-org)\n",
      "            784:I-org (I-org)\n",
      "             16:O     (O)\n",
      "             95:O     (B-tim)\n",
      "              1:O     (O)\n",
      "           1890:O     (O)\n",
      "           2436:O     (O)\n",
      "              4:O     (O)\n",
      "            109:O     (O)\n",
      "            155:O     (O)\n",
      "            583:O     (O)\n",
      "             88:O     (O)\n",
      "              6:O     (O)\n",
      "            356:O     (O)\n",
      "            583:O     (O)\n",
      "             88:O     (O)\n",
      "           1559:O     (B-geo)\n",
      "              5:O     (O)\n",
      "            125:O     (O)\n",
      "              3:O     (O)\n",
      "             60:O     (O)\n",
      "              7:O     (O)\n",
      "            827:O     (O)\n",
      "             90:O     (B-tim)\n",
      "            589:O     (I-tim)\n",
      "             79:O     (O)\n",
      "              1:I-tim (O)\n",
      "            369:I-tim (O)\n",
      "            137:O     (O)\n",
      "            116:O     (O)\n",
      "              0:O     (O)\n",
      "              0:O     (O)\n",
      "              0:O     (O)\n",
      "              0:O     (O)\n",
      "              0:O     (O)\n",
      "              0:O     (O)\n",
      "              0:O     (O)\n"
     ]
    }
   ],
   "source": [
    "i = 390\n",
    "p = model2.predict(np.array(X_te[i:i+batch_size]))[0]\n",
    "p = np.argmax(p, axis=-1)\n",
    "print(\"{:15} {:5}: ({})\".format(\"Word\", \"Pred\", \"True\"))\n",
    "print(\"=\"*30)\n",
    "for w, true, pred in zip(X_te[i], y_te[i], p):\n",
    "    if w != \"PADword\":\n",
    "        print(\"{:15}:{:5} ({})\".format(w, tags[pred], tags[true]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "dCjLIwwj2_r6"
   },
   "source": [
    "### Note how the performance dropped dawn because of the ELMo removal!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "U5.09- [TALLER] ELMo - Name Entity Recognition.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}