data_exploration_task2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "from pprint import pprint\n",
    "\n",
    "import jsonlines\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from tqdm import tqdm, trange\n",
    "from datasets import load_dataset\n",
    "\n",
    "code_to_lang = {\n",
    "    \"en\": \"English\",\n",
    "    \"de\": \"German\",\n",
    "    \"es\": \"Spanish\",\n",
    "    \"fr\": \"French\",\n",
    "    \"ja\": \"Japanese\",\n",
    "    \"zh\": \"Chinese\",\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No config specified, defaulting to: amazon_reviews_multi/all_languages\n",
      "Found cached dataset amazon_reviews_multi (/home/matias/.cache/huggingface/datasets/amazon_reviews_multi/all_languages/1.0.0/724e94f4b0c6c405ce7e476a6c5ef4f87db30799ad49f765094cf9770e0f7609)\n",
      "No config specified, defaulting to: amazon_reviews_multi/all_languages\n",
      "Found cached dataset amazon_reviews_multi (/home/matias/.cache/huggingface/datasets/amazon_reviews_multi/all_languages/1.0.0/724e94f4b0c6c405ce7e476a6c5ef4f87db30799ad49f765094cf9770e0f7609)\n",
      "No config specified, defaulting to: amazon_reviews_multi/all_languages\n",
      "Found cached dataset amazon_reviews_multi (/home/matias/.cache/huggingface/datasets/amazon_reviews_multi/all_languages/1.0.0/724e94f4b0c6c405ce7e476a6c5ef4f87db30799ad49f765094cf9770e0f7609)\n"
     ]
    }
   ],
   "source": [
    "full_train_dataset = load_dataset(\"amazon_reviews_multi\", split=\"train\").with_format(\n",
    "    \"torch\"\n",
    ")\n",
    "full_valid_dataset = load_dataset(\"amazon_reviews_multi\", split=\"validation\").with_format(\n",
    "    \"torch\"\n",
    ")\n",
    "full_test_dataset = load_dataset(\"amazon_reviews_multi\", split=\"test\").with_format(\n",
    "    \"torch\"\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using custom data configuration default-a3faa38d34f6af55\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading and preparing dataset json/default to /home/matias/.cache/huggingface/datasets/json/default-a3faa38d34f6af55/0.0.0/e6070c77f18f01a5ad4551a8b7edfba20b8438b7cad4d94e6ad9378022ce4aab...\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e14669c6a9bc4125bd8ba4baa733f57a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b64d49aa892f481eaadec1566302a1dc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Extracting data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65f1ebcd3b9841fba21e0a831a16ac5d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0 tables [00:00, ? tables/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset json downloaded and prepared to /home/matias/.cache/huggingface/datasets/json/default-a3faa38d34f6af55/0.0.0/e6070c77f18f01a5ad4551a8b7edfba20b8438b7cad4d94e6ad9378022ce4aab. Subsequent calls will reuse this data.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using custom data configuration default-265673c1176f9c06\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading and preparing dataset json/default to /home/matias/.cache/huggingface/datasets/json/default-265673c1176f9c06/0.0.0/e6070c77f18f01a5ad4551a8b7edfba20b8438b7cad4d94e6ad9378022ce4aab...\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a6699f17d18540a881897f320b8a9e32",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "56fe33ff738443199165a4a931ea8c68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Extracting data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5395e755bc2b4a9cbdac4facd318790e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0 tables [00:00, ? tables/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset json downloaded and prepared to /home/matias/.cache/huggingface/datasets/json/default-265673c1176f9c06/0.0.0/e6070c77f18f01a5ad4551a8b7edfba20b8438b7cad4d94e6ad9378022ce4aab. Subsequent calls will reuse this data.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using custom data configuration default-5ed1e2abb728c315\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading and preparing dataset json/default to /home/matias/.cache/huggingface/datasets/json/default-5ed1e2abb728c315/0.0.0/e6070c77f18f01a5ad4551a8b7edfba20b8438b7cad4d94e6ad9378022ce4aab...\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3ba71adc64824cf2be875622128990f7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9b0e3f00e9a74647b3432856f052963f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Extracting data files:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cd90e2a9086641be8384a408ac676cfc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0 tables [00:00, ? tables/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset json downloaded and prepared to /home/matias/.cache/huggingface/datasets/json/default-5ed1e2abb728c315/0.0.0/e6070c77f18f01a5ad4551a8b7edfba20b8438b7cad4d94e6ad9378022ce4aab. Subsequent calls will reuse this data.\n"
     ]
    }
   ],
   "source": [
    "train_dataset_tokenized = (\n",
    "    load_dataset(\n",
    "        \"json\", data_files={\"train\": \"ARC/train_dataset_bert_tokenized.jsonl\"}, split=\"train\"\n",
    "    )\n",
    ")\n",
    "valid_dataset_tokenized = (\n",
    "    load_dataset(\n",
    "        \"json\", data_files={\"train\": \"ARC/valid_dataset_bert_tokenized.jsonl\"}, split=\"train\"\n",
    "    )\n",
    ")\n",
    "test_dataset_tokenized = (\n",
    "    load_dataset(\n",
    "        \"json\", data_files={\"train\": \"ARC/test_dataset_bert_tokenized.jsonl\"}, split=\"train\"\n",
    "    )\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_train_id = np.array(full_train_dataset[\"review_id\"])\n",
    "small_train_id = np.array(train_dataset_tokenized[\"review_id\"])\n",
    "idx_train = np.isin(full_train_id, small_train_id)\n",
    "idx_train = np.arange(len(full_train_id))[idx_train]\n",
    "train_dataset = full_train_dataset.select(idx_train)\n",
    "\n",
    "full_valid_id = np.array(full_valid_dataset[\"review_id\"])\n",
    "small_valid_id = np.array(valid_dataset_tokenized[\"review_id\"])\n",
    "idx_valid = np.isin(full_valid_id, small_valid_id)\n",
    "idx_valid = np.arange(len(full_valid_id))[idx_valid]\n",
    "valid_dataset = full_valid_dataset.select(idx_valid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def utf8len(s):\n",
    "    return len(s.encode(\"utf-8\"))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 30000/30000 [00:14<00:00, 2031.44it/s]\n",
      "100%|██████████| 1000/1000 [00:00<00:00, 2218.24it/s]\n"
     ]
    }
   ],
   "source": [
    "def extract_metrics(dataset, language=\"English\"):\n",
    "    length_shortest_sample = np.inf\n",
    "    samples_with_shortest_sample = []\n",
    "\n",
    "    length_longest_sample = 0\n",
    "    samples_with_longest_sample = []\n",
    "\n",
    "    metrics = []\n",
    "    for sample in tqdm(dataset):\n",
    "\n",
    "        modified_language = code_to_lang[sample[\"language\"]]\n",
    "        modified_product_category = sample[\"product_category\"].replace(\"_\", \" \")\n",
    "        modified_product_category = modified_product_category[0].upper() + modified_product_category[1:]\n",
    "\n",
    "        if (\n",
    "            len(sample[\"review_body\"]) == length_shortest_sample\n",
    "            and modified_language == language\n",
    "        ):\n",
    "            samples_with_shortest_sample += [sample]\n",
    "        if (\n",
    "            len(sample[\"review_body\"]) < length_shortest_sample\n",
    "            and modified_language == language\n",
    "        ):\n",
    "            length_shortest_sample = len(sample[\"review_body\"])\n",
    "            samples_with_shortest_sample = [sample]\n",
    "\n",
    "        if (\n",
    "            len(sample[\"review_body\"]) == length_longest_sample\n",
    "            and modified_language == language\n",
    "        ):\n",
    "            samples_with_longest_sample += [sample]\n",
    "        if (\n",
    "            len(sample[\"review_body\"]) > length_longest_sample\n",
    "            and modified_language == language\n",
    "        ):\n",
    "            length_longest_sample = len(sample[\"review_body\"])\n",
    "            samples_with_longest_sample = [sample]\n",
    "\n",
    "        metric = {\n",
    "            \"document_length\": len(sample[\"review_body\"]),\n",
    "            \"document_length_byte\": utf8len(sample[\"review_body\"]),\n",
    "            \"unique_caracters_document\": len(np.unique(list(sample[\"review_body\"]))),\n",
    "            \"product_category\": modified_product_category,\n",
    "            \"title_length\": len(sample[\"review_title\"]),\n",
    "            \"title_length_byte\": utf8len(sample[\"review_title\"]),\n",
    "            \"unique_caracters_title\": len(np.unique(list(sample[\"review_title\"]))),\n",
    "            \"language\": modified_language,\n",
    "            \"stars\": sample[\"stars\"].item()\n",
    "        }\n",
    "        metrics.append(metric)\n",
    "    metrics = pd.DataFrame.from_records(metrics).sort_values(\"language\", ascending=False)\n",
    "    return metrics, samples_with_shortest_sample, samples_with_longest_sample\n",
    "\n",
    "(\n",
    "    train_metrics,\n",
    "    train_samples_with_shortest_sample,\n",
    "    train_samples_with_longest_sample,\n",
    ") = extract_metrics(train_dataset)\n",
    "(\n",
    "    valid_metrics,\n",
    "    valid_samples_with_shortest_sample,\n",
    "    valid_samples_with_longest_sample,\n",
    ") = extract_metrics(valid_dataset)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'review_id': 'en_0902148',\n",
       "  'product_id': 'product_en_0595978',\n",
       "  'reviewer_id': 'reviewer_en_0983661',\n",
       "  'stars': tensor(3),\n",
       "  'review_body': 'I am 5\\'9, 162 pounds, and typically wear size 10 tall/long in pants. After reading all the reviews that these pants run a little large - I decided to stick with my size as a 10 because I wanted something that was a little more loose fitting and forgiving. I am glad I didn\\'t order a size smaller because both pair I received did NOT run small. As a matter of fact - if I were to order again, I would size UP 1 size in order to have a little more room. The size 10 fit, but are a little tight everywhere except the waist. And as far as the length goes - I ordered the regular instead of the long, again because of reviews. They are BARELY long enough. The fit is more of a flare leg, not a straight leg. This wouldn\\'t be that big of a deal if they were longer, but coming at mid-ankle and being flared just looks all kinds of wrong. I may just have to send these back and order the long size and have them altered. I wish there was a choice in between a 32\" inseam and a 36\" inseam. The fit is mid to low rise and comes about 2\" below my belly button. The zip pocket in the front fits my Galaxy 7 phone, but since the legs are tight I wouldn\\'t put my phone here as it would be uncomfortable, look bulgy, and would somewhat restrict movement. I agree with others that the drawstring is fairly useless, and I also agree that it might be a good idea to double stitch the buttons once you get the pants. These are made in China and the manufacturing quality is probably not in line with the price point of these pants. They did however include an extra button stitched to the inside tag of the pants in case you lose one. At this point - 3 stars is the best I could offer these pants. I will return the ones I have for a 12 Long and update my review at that time. Hope this helps some of you that are taller than 5\\'7 but shorter than 6 foot!',\n",
       "  'review_title': 'Who are all these people leaving 5 star reviews?',\n",
       "  'language': 'en',\n",
       "  'product_category': 'other'}]"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_samples_with_longest_sample"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Smallest and longest question samples in the dataset for English and Arabic\n",
    "\n",
    "### English samples:\n",
    "\n",
    "- **Title**: Who are all these people leaving 5 star reviews?\n",
    "    \n",
    "    **Body**: I am 5'9, 162 pounds, and typically wear size 10 tall/long in pants. After reading all the reviews that these pants run a little large - I decided to stick with my size as a 10 because I wanted something that was a little more loose fitting and forgiving. I am glad I didn't order a size smaller because both pair I received did NOT run small. As a matter of fact - if I were to order again, I would size UP 1 size in order to have a little more room. The size 10 fit, but are a little tight everywhere except the waist. And as far as the length goes - I ordered the regular instead of the long, again because of reviews. They are BARELY long enough. The fit is more of a flare leg, not a straight leg. This wouldn't be that big of a deal if they were longer, but coming at mid-ankle and being flared just looks all kinds of wrong. I may just have to send these back and order the long size and have them altered. I wish there was a choice in between a 32\" inseam and a 36\" inseam. The fit is mid to low rise and comes about 2\" below my belly button. The zip pocket in the front fits my Galaxy 7 phone, but since the legs are tight I wouldn't put my phone here as it would be uncomfortable, look bulgy, and would somewhat restrict movement. I agree with others that the drawstring is fairly useless, and I also agree that it might be a good idea to double stitch the buttons once you get the pants. These are made in China and the manufacturing quality is probably not in line with the price point of these pants. They did however include an extra button stitched to the inside tag of the pants in case you lose one. At this point - 3 stars is the best I could offer these pants. I will return the ones I have for a 12 Long and update my review at that time. Hope this helps some of you that are taller than 5'7 but shorter than 6 foot!\n",
    "    \n",
    "- **Title**: One Star\n",
    "    \n",
    "    **Body**: I never received it.\n",
    "\n",
    "- **Title**: Coucou\n",
    "    \n",
    "    **Body**: Usuel et sport Déçu\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_metrics_per_language = train_metrics.groupby(\"language\").agg(\n",
    "    {\n",
    "        \"document_length\": \"mean\",\n",
    "        \"document_length_byte\": \"mean\",\n",
    "        \"unique_caracters_document\": \"mean\",\n",
    "        \"title_length\": \"mean\",\n",
    "        \"title_length_byte\": \"mean\",\n",
    "        \"unique_caracters_title\": \"mean\",\n",
    "        # \"product_category\": \"mean\",\n",
    "    }\n",
    ")\n",
    "train_metrics_per_language[\"count\"] = train_metrics.groupby(\"language\").count()[\n",
    "    \"document_length\"\n",
    "]\n",
    "\n",
    "valid_metrics_per_language = valid_metrics.groupby(\"language\").agg(\n",
    "    {\n",
    "        \"document_length\": \"mean\",\n",
    "        \"document_length_byte\": \"mean\",\n",
    "        \"unique_caracters_document\": \"mean\",\n",
    "        \"title_length\": \"mean\",\n",
    "        \"title_length_byte\": \"mean\",\n",
    "        \"unique_caracters_title\": \"mean\",\n",
    "        # \"product_category\": \"mean\",\n",
    "    }\n",
    ")\n",
    "valid_metrics_per_language[\"count\"] = valid_metrics.groupby(\"language\").count()[\n",
    "    \"document_length\"\n",
    "]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 1, figsize=(8, 8))\n",
    "\n",
    "document_number_metric = pd.DataFrame(\n",
    "    data=[train_metrics_per_language[\"count\"], valid_metrics_per_language[\"count\"]],\n",
    "    index=[\"Train\", \"Valid\"],\n",
    ").T.sort_values(\"language\", ascending=False)\n",
    "\n",
    "document_number_metric.plot(\n",
    "    kind=\"barh\",\n",
    "    stacked=True,\n",
    "    color=[\n",
    "        \"cornflowerblue\",\n",
    "        \"coral\",\n",
    "        \"lightgreen\"\n",
    "    ],\n",
    "    ax=axs,\n",
    ")\n",
    "axs.legend(loc=\"upper right\")\n",
    "axs.set_ylabel(\"Language\")\n",
    "axs.set_xlabel(\"# of documents\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"ARC/number_documents.jpg\", dpi=300)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 2, figsize=(16, 8))\n",
    "\n",
    "ordered_languages = train_metrics_per_language.index.sort_values(ascending=True)\n",
    "\n",
    "sns.barplot(\n",
    "    train_metrics,\n",
    "    x=\"document_length\",\n",
    "    y=\"language\",\n",
    "    ax=axs[0],\n",
    "    color=\"cornflowerblue\",\n",
    "    order=ordered_languages,\n",
    ")\n",
    "axs[0].set_xlabel(\"Document length (# of caracters)\")\n",
    "axs[0].set_ylabel(\"Language\")\n",
    "sns.barplot(\n",
    "    train_metrics,\n",
    "    x=\"title_length\",\n",
    "    y=\"language\",\n",
    "    ax=axs[1],\n",
    "    color=\"cornflowerblue\",\n",
    "    order=ordered_languages,\n",
    ")\n",
    "axs[1].set_xlabel(\"Title length (# of caracters)\")\n",
    "axs[1].set_ylabel(\"Language\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"ARC/train_doc_q_length.jpg\", dpi=300)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_metrics_per_product = train_metrics.groupby(\"product_category\").agg(\n",
    "    {\n",
    "        \"document_length\": \"mean\",\n",
    "        \"document_length_byte\": \"mean\",\n",
    "        \"unique_caracters_document\": \"mean\",\n",
    "        \"title_length\": \"mean\",\n",
    "        \"title_length_byte\": \"mean\",\n",
    "        \"unique_caracters_title\": \"mean\",\n",
    "        # \"product_category\": \"mean\",\n",
    "    }\n",
    ")\n",
    "train_metrics_per_product[\"count\"] = train_metrics.groupby(\"product_category\").count()[\n",
    "    \"document_length\"\n",
    "]\n",
    "\n",
    "valid_metrics_per_product = valid_metrics.groupby(\"product_category\").agg(\n",
    "    {\n",
    "        \"document_length\": \"mean\",\n",
    "        \"document_length_byte\": \"mean\",\n",
    "        \"unique_caracters_document\": \"mean\",\n",
    "        \"title_length\": \"mean\",\n",
    "        \"title_length_byte\": \"mean\",\n",
    "        \"unique_caracters_title\": \"mean\",\n",
    "        # \"product_category\": \"mean\",\n",
    "    }\n",
    ")\n",
    "valid_metrics_per_product[\"count\"] = valid_metrics.groupby(\"product_category\").count()[\n",
    "    \"document_length\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 1, figsize=(8, 8))\n",
    "\n",
    "document_number_metric = pd.DataFrame(\n",
    "    data=[train_metrics_per_product[\"count\"], valid_metrics_per_product[\"count\"]],\n",
    "    index=[\"Train\", \"Valid\"],\n",
    ").T.sort_values(\"product_category\", ascending=False)\n",
    "\n",
    "a = document_number_metric.index\n",
    "new_index = []\n",
    "for i in document_number_metric.index:\n",
    "    c = i.replace(\"_\", \" \")\n",
    "    new_index.append(c[0].upper() + c[1:])\n",
    "document_number_metric.index = new_index\n",
    "\n",
    "document_number_metric.plot(\n",
    "    kind=\"barh\",\n",
    "    stacked=True,\n",
    "    color=[\n",
    "        \"cornflowerblue\",\n",
    "        \"coral\",\n",
    "        \"lightgreen\"\n",
    "    ],\n",
    "    ax=axs,\n",
    ")\n",
    "\n",
    "axs.set_ylabel(\"Product\")\n",
    "axs.set_xlabel(\"# of documents\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"ARC/number_product.jpg\", dpi=300)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_metrics_per_star = train_metrics.groupby(\"stars\").agg(\n",
    "    {\n",
    "        \"document_length\": \"mean\",\n",
    "        \"document_length_byte\": \"mean\",\n",
    "        \"unique_caracters_document\": \"mean\",\n",
    "        \"title_length\": \"mean\",\n",
    "        \"title_length_byte\": \"mean\",\n",
    "        \"unique_caracters_title\": \"mean\",\n",
    "        # \"product_category\": \"mean\",\n",
    "    }\n",
    ")\n",
    "train_metrics_per_star[\"count\"] = train_metrics.groupby(\"stars\").count()[\n",
    "    \"document_length\"\n",
    "]\n",
    "\n",
    "valid_metrics_per_star = valid_metrics.groupby(\"stars\").agg(\n",
    "    {\n",
    "        \"document_length\": \"mean\",\n",
    "        \"document_length_byte\": \"mean\",\n",
    "        \"unique_caracters_document\": \"mean\",\n",
    "        \"title_length\": \"mean\",\n",
    "        \"title_length_byte\": \"mean\",\n",
    "        \"unique_caracters_title\": \"mean\",\n",
    "        # \"product_category\": \"mean\",\n",
    "    }\n",
    ")\n",
    "valid_metrics_per_star[\"count\"] = valid_metrics.groupby(\"stars\").count()[\n",
    "    \"document_length\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 2, figsize=(16, 8))\n",
    "\n",
    "document_number_metric = pd.DataFrame(\n",
    "    data=[train_metrics_per_product[\"count\"], valid_metrics_per_product[\"count\"]],\n",
    "    index=[\"Train\", \"Valid\"],\n",
    ").T.sort_values(\"product_category\", ascending=False)\n",
    "\n",
    "a = document_number_metric.index\n",
    "new_index = []\n",
    "for i in document_number_metric.index:\n",
    "    c = i.replace(\"_\", \" \")\n",
    "    new_index.append(c[0].upper() + c[1:])\n",
    "document_number_metric.index = new_index\n",
    "\n",
    "document_number_metric.plot(\n",
    "    kind=\"barh\",\n",
    "    stacked=True,\n",
    "    color=[\n",
    "        \"cornflowerblue\",\n",
    "        \"coral\",\n",
    "        \"lightgreen\"\n",
    "    ],\n",
    "    ax=axs[0],\n",
    ")\n",
    "\n",
    "axs[0].set_ylabel(\"Product\")\n",
    "axs[0].set_xlabel(\"# of documents\")\n",
    "\n",
    "\n",
    "document_number_metric = pd.DataFrame(\n",
    "    data=[train_metrics_per_star[\"count\"], valid_metrics_per_star[\"count\"]],\n",
    "    index=[\"Train\", \"Valid\"],\n",
    ").T.sort_values(\"stars\", ascending=False)\n",
    "\n",
    "document_number_metric.plot(\n",
    "    kind=\"barh\",\n",
    "    stacked=True,\n",
    "    color=[\n",
    "        \"cornflowerblue\",\n",
    "        \"coral\",\n",
    "        \"lightgreen\"\n",
    "    ],\n",
    "    ax=axs[1],\n",
    ")\n",
    "axs[1].legend(loc=\"upper right\")\n",
    "axs[1].set_ylabel(\"Number of stars\")\n",
    "axs[1].set_xlabel(\"# of documents\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"ARC/number_product_star.jpg\", dpi=300)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = np.array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 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, 4, 4, 4, 4, 4, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 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, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4,\n",
    "       4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,\n",
    "       4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4,\n",
    "       4, 4, 4, 4, 4, 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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 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, 4, 4, 4, 4, 4, 4,\n",
    "       4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
    "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
    "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,\n",
    "       3, 3, 3, 4, 4, 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])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report\n",
    "\n",
    "labels = test_dataset_tokenized[\"label\"]\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(8, 8))\n",
    "cm = confusion_matrix(\n",
    "    labels,\n",
    "    predictions,\n",
    "    normalize=\"true\",\n",
    ")\n",
    "sns.heatmap(\n",
    "    cm,\n",
    "    annot=True,\n",
    "    fmt=\".1f\",\n",
    "    cmap=\"Blues\",\n",
    "    ax=ax,\n",
    ")\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "\"Column review not in the dataset. Current columns in the dataset: ['review_id', 'product_id', 'reviewer_id', 'stars', 'review_body', 'review_title', 'language', 'product_category', 'input_ids', 'token_type_ids', 'attention_mask', 'label']\"",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32m/home/matias/SLP-Canine/data_exploration_task2.ipynb Cell 19\u001b[0m in \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell://wsl%2Bubuntu/home/matias/SLP-Canine/data_exploration_task2.ipynb#X36sdnNjb2RlLXJlbW90ZQ%3D%3D?line=0'>1</a>\u001b[0m test_dataset_tokenized[\u001b[39m\"\u001b[39m\u001b[39mreview\u001b[39m\u001b[39m\"\u001b[39m]\n",
      "File \u001b[0;32m~/.pyenv/versions/3.9.8/envs/MVA/lib/python3.9/site-packages/datasets/arrow_dataset.py:2343\u001b[0m, in \u001b[0;36mDataset.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   2341\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__getitem__\u001b[39m(\u001b[39mself\u001b[39m, key):  \u001b[39m# noqa: F811\u001b[39;00m\n\u001b[1;32m   2342\u001b[0m     \u001b[39m\"\"\"Can be used to index columns (by string names) or rows (by integer index or iterable of indices or bools).\"\"\"\u001b[39;00m\n\u001b[0;32m-> 2343\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_getitem(\n\u001b[1;32m   2344\u001b[0m         key,\n\u001b[1;32m   2345\u001b[0m     )\n",
      "File \u001b[0;32m~/.pyenv/versions/3.9.8/envs/MVA/lib/python3.9/site-packages/datasets/arrow_dataset.py:2327\u001b[0m, in \u001b[0;36mDataset._getitem\u001b[0;34m(self, key, decoded, **kwargs)\u001b[0m\n\u001b[1;32m   2325\u001b[0m format_kwargs \u001b[39m=\u001b[39m format_kwargs \u001b[39mif\u001b[39;00m format_kwargs \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39melse\u001b[39;00m {}\n\u001b[1;32m   2326\u001b[0m formatter \u001b[39m=\u001b[39m get_formatter(format_type, features\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mfeatures, decoded\u001b[39m=\u001b[39mdecoded, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mformat_kwargs)\n\u001b[0;32m-> 2327\u001b[0m pa_subtable \u001b[39m=\u001b[39m query_table(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_data, key, indices\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_indices \u001b[39mif\u001b[39;49;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_indices \u001b[39mis\u001b[39;49;00m \u001b[39mnot\u001b[39;49;00m \u001b[39mNone\u001b[39;49;00m \u001b[39melse\u001b[39;49;00m \u001b[39mNone\u001b[39;49;00m)\n\u001b[1;32m   2328\u001b[0m formatted_output \u001b[39m=\u001b[39m format_table(\n\u001b[1;32m   2329\u001b[0m     pa_subtable, key, formatter\u001b[39m=\u001b[39mformatter, format_columns\u001b[39m=\u001b[39mformat_columns, output_all_columns\u001b[39m=\u001b[39moutput_all_columns\n\u001b[1;32m   2330\u001b[0m )\n\u001b[1;32m   2331\u001b[0m \u001b[39mreturn\u001b[39;00m formatted_output\n",
      "File \u001b[0;32m~/.pyenv/versions/3.9.8/envs/MVA/lib/python3.9/site-packages/datasets/formatting/formatting.py:460\u001b[0m, in \u001b[0;36mquery_table\u001b[0;34m(table, key, indices)\u001b[0m\n\u001b[1;32m    458\u001b[0m     _raise_bad_key_type(key)\n\u001b[1;32m    459\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(key, \u001b[39mstr\u001b[39m):\n\u001b[0;32m--> 460\u001b[0m     _check_valid_column_key(key, table\u001b[39m.\u001b[39;49mcolumn_names)\n\u001b[1;32m    461\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m    462\u001b[0m     size \u001b[39m=\u001b[39m indices\u001b[39m.\u001b[39mnum_rows \u001b[39mif\u001b[39;00m indices \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39melse\u001b[39;00m table\u001b[39m.\u001b[39mnum_rows\n",
      "File \u001b[0;32m~/.pyenv/versions/3.9.8/envs/MVA/lib/python3.9/site-packages/datasets/formatting/formatting.py:400\u001b[0m, in \u001b[0;36m_check_valid_column_key\u001b[0;34m(key, columns)\u001b[0m\n\u001b[1;32m    398\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_check_valid_column_key\u001b[39m(key: \u001b[39mstr\u001b[39m, columns: List[\u001b[39mstr\u001b[39m]) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m    399\u001b[0m     \u001b[39mif\u001b[39;00m key \u001b[39mnot\u001b[39;00m \u001b[39min\u001b[39;00m columns:\n\u001b[0;32m--> 400\u001b[0m         \u001b[39mraise\u001b[39;00m \u001b[39mKeyError\u001b[39;00m(\u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mColumn \u001b[39m\u001b[39m{\u001b[39;00mkey\u001b[39m}\u001b[39;00m\u001b[39m not in the dataset. Current columns in the dataset: \u001b[39m\u001b[39m{\u001b[39;00mcolumns\u001b[39m}\u001b[39;00m\u001b[39m\"\u001b[39m)\n",
      "\u001b[0;31mKeyError\u001b[0m: \"Column review not in the dataset. Current columns in the dataset: ['review_id', 'product_id', 'reviewer_id', 'stars', 'review_body', 'review_title', 'language', 'product_category', 'input_ids', 'token_type_ids', 'attention_mask', 'label']\""
     ]
    }
   ],
   "source": [
    "test_dataset_tokenized[\"review\"]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "MVA",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.8"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}