SaveEmbeddings.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create the embeddings and save them to file\n",
    "Savce the embeddings so that they can be reused later"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import the necessary libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import tensorflow_hub as hub\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import json"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Set the paths and files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "use_path = \"dataset/\"\n",
    "baseline_file = \"baseline_questions.csv\"\n",
    "quora_file = \"quora_questions.csv\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Download the USE module"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Using /tmp/tfhub_modules to cache modules.\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_0:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_0\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_1:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_1\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_10:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_10\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_11:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_11\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_12:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_12\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_13:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_13\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_14:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_14\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_15:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_15\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_16:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_16\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_2:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_2\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_3:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_3\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_4:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_4\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_5:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_5\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_6:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_6\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_7:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_7\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_8:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_8\n",
      "INFO:tensorflow:Initialize variable module/Embeddings_en/sharded_9:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Embeddings_en/sharded_9\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv1/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv1/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv1/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv1/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv2/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv2/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv2/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/conv2/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/ffn/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/k/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/k/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/output_transform/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/output_transform/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/q/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/q/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/v/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_0/self_attention/multihead_attention/v/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv1/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv1/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv1/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv1/kernel\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv2/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv2/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv2/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/conv2/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/ffn/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/k/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/k/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/output_transform/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/output_transform/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/q/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/q/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/v/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_1/self_attention/multihead_attention/v/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv1/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv1/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv1/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv1/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv2/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv2/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv2/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/conv2/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/ffn/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/k/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/k/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/output_transform/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/output_transform/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/q/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/q/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/v/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_2/self_attention/multihead_attention/v/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv1/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv1/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv1/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv1/kernel\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv2/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv2/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv2/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/conv2/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/ffn/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/k/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/k/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/output_transform/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/output_transform/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/q/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/q/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/v/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_3/self_attention/multihead_attention/v/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv1/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv1/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv1/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv1/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv2/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv2/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv2/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/conv2/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/ffn/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/k/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/k/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/output_transform/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/output_transform/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/q/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/q/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/v/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_4/self_attention/multihead_attention/v/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv1/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv1/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv1/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv1/kernel\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv2/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv2/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv2/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/conv2/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/ffn/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/k/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/k/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/output_transform/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/output_transform/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/q/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/q/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/v/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_5/self_attention/multihead_attention/v/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_prepostprocess/layer_norm/layer_norm_bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_prepostprocess/layer_norm/layer_norm_bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_prepostprocess/layer_norm/layer_norm_scale:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/TransformerEncodeFast/encoder/layer_prepostprocess/layer_norm/layer_norm_scale\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/dense/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/dense/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/Transformer/dense/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/Transformer/dense/kernel\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/hidden_layers/tanh_layer_0/dense/bias:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/hidden_layers/tanh_layer_0/dense/bias\n",
      "INFO:tensorflow:Initialize variable module/Encoder_en/hidden_layers/tanh_layer_0/dense/kernel:0 from checkpoint b'/tmp/tfhub_modules/96e8f1d3d4d90ce86b2db128249eb8143a91db73/variables/variables' with Encoder_en/hidden_layers/tanh_layer_0/dense/kernel\n"
     ]
    }
   ],
   "source": [
    "module_url = \"https://tfhub.dev/google/universal-sentence-encoder-large/3\" \n",
    "embed = hub.Module(module_url)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create an interactive session to use later when running the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "session = tf.InteractiveSession()\n",
    "session.run(tf.global_variables_initializer())\n",
    "session.run(tf.tables_initializer())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a dataframe of quora question pairs and answer labels "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_quora_qs():\n",
    "    quora_path = use_path + quora_file\n",
    "    quora_qs = pd.read_csv(quora_path)\n",
    "    return(quora_qs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate embeddings for each of the questions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "sts_input1 = tf.placeholder(tf.string, shape=(None))\n",
    "\n",
    "# For evaluation we use exactly normalized rather than\n",
    "# approximately normalized.\n",
    "sts_encode1 = tf.nn.l2_normalize(embed(sts_input1), axis=1)\n",
    "\n",
    "def get_embeds(session, text):\n",
    "    \"\"\"Returns the similarity scores\"\"\"\n",
    "    embed = session.run(\n",
    "        [sts_encode1],\n",
    "        feed_dict={\n",
    "            sts_input1: text\n",
    "        })\n",
    "    return(embed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a dataframe of queries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>Astrology: I am a Capricorn Sun Cap moon and c...</td>\n",
       "      <td>I'm a triple Capricorn (Sun, Moon and ascendan...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>16</td>\n",
       "      <td>How can I be a good geologist?</td>\n",
       "      <td>What should I do to be a great geologist?</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11</td>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "      <td>How do I read and find my YouTube comments?</td>\n",
       "      <td>How can I see all my Youtube comments?</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12</td>\n",
       "      <td>25</td>\n",
       "      <td>26</td>\n",
       "      <td>What can make Physics easy to learn?</td>\n",
       "      <td>How can you make physics easy to learn?</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13</td>\n",
       "      <td>27</td>\n",
       "      <td>28</td>\n",
       "      <td>What was your first sexual experience like?</td>\n",
       "      <td>What was your first sexual experience?</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  qid1  qid2                                          question1  \\\n",
       "0   5    11    12  Astrology: I am a Capricorn Sun Cap moon and c...   \n",
       "1   7    15    16                     How can I be a good geologist?   \n",
       "2  11    23    24        How do I read and find my YouTube comments?   \n",
       "3  12    25    26               What can make Physics easy to learn?   \n",
       "4  13    27    28        What was your first sexual experience like?   \n",
       "\n",
       "                                           question2  is_duplicate  \n",
       "0  I'm a triple Capricorn (Sun, Moon and ascendan...             1  \n",
       "1          What should I do to be a great geologist?             1  \n",
       "2             How can I see all my Youtube comments?             1  \n",
       "3            How can you make physics easy to learn?             1  \n",
       "4             What was your first sexual experience?             1  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quora_df = get_quora_qs()\n",
    "quora_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Combine both question columns into df with one column\n",
    "q_1_2 = pd.concat([quora_df['question1'], quora_df['question2']])\n",
    "questions = [e for e in q_1_2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3min 10s, sys: 12.4 s, total: 3min 22s\n",
      "Wall time: 27 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "embeddings = get_embeds(session, questions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "res = (embeddings[0]).tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "q_df = pd.DataFrame(questions, columns=['questions'])\n",
    "q_df['embeddings'] = res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>questions</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Astrology: I am a Capricorn Sun Cap moon and c...</td>\n",
       "      <td>[-0.043053239583969116, 0.05912359803915024, -...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>How can I be a good geologist?</td>\n",
       "      <td>[0.04468201845884323, 0.0032161467242985964, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>How do I read and find my YouTube comments?</td>\n",
       "      <td>[0.031296346336603165, -0.002411748981103301, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What can make Physics easy to learn?</td>\n",
       "      <td>[0.0734066367149353, 0.055123668164014816, 0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>What was your first sexual experience like?</td>\n",
       "      <td>[0.05466366186738014, -0.003594229696318507, -...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           questions  \\\n",
       "0  Astrology: I am a Capricorn Sun Cap moon and c...   \n",
       "1                     How can I be a good geologist?   \n",
       "2        How do I read and find my YouTube comments?   \n",
       "3               What can make Physics easy to learn?   \n",
       "4        What was your first sexual experience like?   \n",
       "\n",
       "                                          embeddings  \n",
       "0  [-0.043053239583969116, 0.05912359803915024, -...  \n",
       "1  [0.04468201845884323, 0.0032161467242985964, 0...  \n",
       "2  [0.031296346336603165, -0.002411748981103301, ...  \n",
       "3  [0.0734066367149353, 0.055123668164014816, 0.0...  \n",
       "4  [0.05466366186738014, -0.003594229696318507, -...  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "q_df.to_pickle(\"/home/ec2-user/sentence_embeddings/embeddings.pkl\") "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Environment (conda_tensorflow_p36)",
   "language": "python",
   "name": "conda_tensorflow_p36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}