translate_vi_target.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Translation between low-resource languages\n",
        "This notebook implements the translation between two low-resource languages with an intermediate English translation using pretrained models (second part of Extension I). In particular, the available language pairs are:\n",
        "* Vietnamese - Khmer\n",
        "* Vietnamese - Indonesian"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nyi2y3sPAtep"
      },
      "source": [
        "## Get ALT dataset\n",
        "Download ALT dataset (the documentation can be found at [this link](https://www2.nict.go.jp/astrec-att/member/mutiyama/ALT/)), containing translations to different Asian languages. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hcdzn4BrAC7Z"
      },
      "outputs": [],
      "source": [
        "# Download dataset\n",
        "!wget https://www2.nict.go.jp/astrec-att/member/mutiyama/ALT/ALT-Parallel-Corpus-20191206.zip\n",
        "# Unzip dataset\n",
        "!unzip ALT-Parallel-Corpus-20191206.zip\n",
        "! rm ALT-Parallel-Corpus-20191206.zip"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h2GILajOrWJE"
      },
      "source": [
        "## Preparation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ZOMGrqXorolA"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Define folders containing the pretrained models\n",
        "###\n",
        "\n",
        "checkpoint_vi_en = \"CLAck/vi-en\"\n",
        "checkpoint_en_target = \"CLAck/en-indo\"\n",
        "\n",
        "# Define target language (\"khm\" or \"indo\")\n",
        "target_lang = \"khm\"\n",
        "\n",
        "if target_lang == \"khm\":\n",
        "    mbart_lang = \"km_KH\"\n",
        "elif target_lang == \"indo\":\n",
        "    mbart_lang = \"id_ID\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tQC7ka5H7B94"
      },
      "outputs": [],
      "source": [
        "## Install useful libraries\n",
        "%%capture\n",
        "!pip install transformers datasets metrics sacrebleu transformers[sentencepiece]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Zkq5uZ5NH7lG"
      },
      "outputs": [],
      "source": [
        "#CHANGE the name in /content/ALT-Parallel-Corpus-20191206/ of data_id.txt to data_indo.txt \n",
        "# Needed to avoid confusion between id (-> name of Indonesian language) and id (-> name of index column in dataframe)\n",
        "import os\n",
        "os.rename('/content/ALT-Parallel-Corpus-20191206/data_id.txt', '/content/ALT-Parallel-Corpus-20191206/data_indo.txt')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "FKXcmZ8Y727a",
        "outputId": "488b8e6d-1c10-4827-c703-a9db761da486"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Create dataframe with english, target language and vietnamese\n",
        "###\n",
        "\n",
        "import pandas as pd\n",
        "\n",
        "df_en = pd.read_csv(\"/content/ALT-Parallel-Corpus-20191206/data_en.txt\", sep='\\t', header=None, names=[\"id\", \"en\"])\n",
        "df_vi = pd.read_csv(\"/content/ALT-Parallel-Corpus-20191206/data_vi.txt\", sep='\\t', header=None, names=[\"id\", \"vi\"])\n",
        "df_target = pd.read_csv(\"/content/ALT-Parallel-Corpus-20191206/data_\" + target_lang + \".txt\", sep='\\t', header=None, names=[\"id\", target_lang])\n",
        "df_en = df_en.set_index(\"id\")\n",
        "df_vi = df_vi.set_index(\"id\")\n",
        "df_target = df_target.set_index(\"id\")\n",
        "\n",
        "# These will be useful for the tokenizer\n",
        "df_en_vi = df_en.join(df_vi)\n",
        "df_en_vi.dropna(inplace=True)\n",
        "df_en_target = df_en.join(df_target)\n",
        "df_en_target.dropna(inplace=True)\n",
        "\n",
        "# This will be used for evaluation\n",
        "df_en_vi_target = df_en_vi.join(df_target)\n",
        "df_en_vi_target.dropna(inplace=True)\n",
        "df_en_vi_target.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "GNyfA42HojW0",
        "outputId": "fffb60c8-d97d-4134-f00a-dbb5b3b1137f"
      },
      "outputs": [],
      "source": [
        "### \n",
        "# Split dataset into training, evaluation and test\n",
        "# To make sure that test sentences have not been seen by the models during training,\n",
        "# the exact splits used for training are replicated, and are then merged together\n",
        "###\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Indonesian\n",
        "train_df, test_df_en_target = train_test_split(df_en_target, test_size=0.2, random_state=42)\n",
        "\n",
        "# Vietnamese\n",
        "train_df, test_df_en_vi = train_test_split(df_en_vi, test_size=0.2, random_state=42)\n",
        "\n",
        "test_df = test_df_en_target.join(test_df_en_vi[\"vi\"])\n",
        "test_df.dropna(inplace=True)\n",
        "test_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "jZwgfO6_bcA5"
      },
      "outputs": [],
      "source": [
        "####\n",
        "# Load tokenizers: basic Marian tokenizer + pretrained mBart tokenizer for both languages\n",
        "####\n",
        "\n",
        "from transformers import AutoTokenizer, MBart50TokenizerFast, MarianTokenizer\n",
        "import random, transformers\n",
        "\n",
        "transformers.logging.set_verbosity(transformers.logging.CRITICAL)\n",
        "\n",
        "mbart_tokenizer_vi = MBart50TokenizerFast.from_pretrained(\"facebook/mbart-large-50\", src_lang=\"en_XX\", tgt_lang=\"vi_VN\")\n",
        "mbart_tokenizer_target = MBart50TokenizerFast.from_pretrained(\"facebook/mbart-large-50\", src_lang=\"en_XX\", tgt_lang=mbart_lang)\n",
        "\n",
        "marian_tokenizer_vi_en = MarianTokenizer.from_pretrained(\"Helsinki-NLP/opus-mt-zh-en\")\n",
        "marian_tokenizer_en_target = MarianTokenizer.from_pretrained(\"Helsinki-NLP/opus-mt-en-zh\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rFrFyhQBbpsb",
        "outputId": "2004f682-d839-4fc2-bf4c-97b682d5f80b"
      },
      "outputs": [],
      "source": [
        "####\n",
        "# Extend Marian tokenizer by adding tokens from target language according\n",
        "# to how mBart tokenizes sentences in the dataset\n",
        "####\n",
        "\n",
        "import transformers\n",
        "from tqdm import tqdm\n",
        "\n",
        "transformers.logging.set_verbosity(transformers.logging.CRITICAL)\n",
        "\n",
        "## Add vietnamese tokens to vi-en tokenizer\n",
        "tokensZoo = []\n",
        "\n",
        "for sentence in tqdm(list(df_en_vi[\"vi\"])):\n",
        "  tokenized_sentence = mbart_tokenizer_vi(sentence)\n",
        "  for t in mbart_tokenizer_vi.convert_ids_to_tokens(tokenized_sentence[\"input_ids\"]):\n",
        "    if t.lstrip(\"▁\") not in tokensZoo and t.lstrip(\"▁\") != \"\":\n",
        "      tokensZoo.append(t.lstrip(\"▁\"))\n",
        "    \n",
        "\n",
        "print(f\"{len(tokensZoo)} tokens to be added.\")\n",
        "print(f\"initial vocab size: {len(marian_tokenizer_vi_en)}\")\n",
        "initial_len = len(marian_tokenizer_vi_en)\n",
        "marian_tokenizer_vi_en.add_tokens(tokensZoo, special_tokens=True)\n",
        "marian_tokenizer_vi_en.add_tokens([\"<zh>\", \"<vi>\"], special_tokens=True)\n",
        "print(f\"final vocab size: {len(marian_tokenizer_vi_en)}\")\n",
        "added_tokens = len(marian_tokenizer_vi_en) - initial_len\n",
        "\n",
        "\n",
        "## Add target language tokens to en-target tokenizer\n",
        "tokensZoo = []\n",
        "\n",
        "for sentence in tqdm(list(df_en_target[target_lang])):\n",
        "  tokenized_sentence = mbart_tokenizer_target(sentence)\n",
        "  for t in mbart_tokenizer_target.convert_ids_to_tokens(tokenized_sentence[\"input_ids\"]):\n",
        "    if t.lstrip(\"▁\") not in tokensZoo and t.lstrip(\"▁\") != \"\":\n",
        "      tokensZoo.append(t.lstrip(\"▁\"))\n",
        "    \n",
        "\n",
        "print(f\"{len(tokensZoo)} tokens to be added.\")\n",
        "print(f\"initial vocab size: {len(marian_tokenizer_en_target)}\")\n",
        "initial_len = len(marian_tokenizer_en_target)\n",
        "marian_tokenizer_en_target.add_tokens(tokensZoo, special_tokens=True)\n",
        "marian_tokenizer_en_target.add_tokens([\"<2zh>\", \"<2\" + target_lang +\">\"], special_tokens=True)\n",
        "print(f\"final vocab size: {len(marian_tokenizer_en_target)}\")\n",
        "added_tokens = len(marian_tokenizer_en_target) - initial_len"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-P6lnB54mbkl"
      },
      "source": [
        "# First Phase\n",
        "Vietnamese -> English"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dsoPvNSDLu-T"
      },
      "outputs": [],
      "source": [
        "####\n",
        "# Load vietnamese - english model \n",
        "####\n",
        "from transformers import AutoModelForSeq2SeqLM, DataCollatorForSeq2Seq, Seq2SeqTrainingArguments, Seq2SeqTrainer\n",
        "\n",
        "model_vi_en = AutoModelForSeq2SeqLM.from_pretrained(checkpoint_vi_en)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "90ba8e9c63ec4e1f878ef30d3f277a2c",
            "02a8ad4bf5524174bf664b337475bd03",
            "797ef0e56b584c3b81b7bf73ad447a4d",
            "aa5dc610fe4842278e043bc9b1ad6def",
            "ba4f9433d6824adf8e330e3e1087c787",
            "c89ab218dd4048c294f669b29c5d45a2",
            "c7fb557172374bae84ed90f81fc8913a",
            "41f56499ff5f494f9a78e469ca8135a5",
            "3fbec5804cfe48aeaf63a87b4dde033d",
            "f571536339634855bcc0240631f8020d",
            "4d23daee9f5a4039afca41b513fa5ea7"
          ]
        },
        "id": "ZdxLSTdtSKqc",
        "outputId": "6b5386d5-56d5-4e8d-8ae9-73518d1d492f"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Tokenize sentences in Vietnamese and English (using only test set) and create first dataset\n",
        "###\n",
        "\n",
        "from datasets import Dataset\n",
        "\n",
        "ds_vi = Dataset.from_pandas(test_df)\n",
        "batch_size = 64\n",
        "\n",
        "def preprocess_function(examples):\n",
        "    inputs = [s for s in examples[\"vi\"]]\n",
        "    targets = [s for s in examples[\"en\"]]\n",
        "    model_inputs = marian_tokenizer_vi_en(inputs, max_length=64, truncation=True, padding=\"max_length\")\n",
        "\n",
        "    with marian_tokenizer_vi_en.as_target_tokenizer():\n",
        "        labels = marian_tokenizer_vi_en(targets, max_length=64, truncation=True, padding=\"max_length\")\n",
        "\n",
        "    model_inputs[\"labels\"] = labels[\"input_ids\"]\n",
        "\n",
        "    return model_inputs\n",
        "\n",
        "columns_to_return = ['input_ids', 'labels', 'attention_mask']\n",
        "ds_vi = ds_vi.map(preprocess_function, batched=True, batch_size=16)\n",
        "ds_vi.set_format(type='torch', columns=columns_to_return)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VtW9ZlD7k-MT",
        "outputId": "0cb0aa3a-7def-434a-95ff-ecf986827b32"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Define trainer arguments to translate from Vietnamese to English\n",
        "###\n",
        "\n",
        "args = Seq2SeqTrainingArguments(\n",
        "  \"./\", \n",
        "  evaluation_strategy = \"epoch\",\n",
        "  save_strategy=\"epoch\",\n",
        "  learning_rate=2e-5,\n",
        "  per_device_train_batch_size=batch_size,\n",
        "  per_device_eval_batch_size=batch_size,\n",
        "  weight_decay=0.01,\n",
        "  save_total_limit=1,\n",
        "  num_train_epochs=5,\n",
        "  predict_with_generate=True,\n",
        "  fp16=True, #CUDA purposes,\n",
        "  disable_tqdm=False,\n",
        ")\n",
        "\n",
        "data_collator = DataCollatorForSeq2Seq(marian_tokenizer_vi_en, model=model_vi_en)\n",
        "\n",
        "trainer = Seq2SeqTrainer(\n",
        "    model_vi_en,\n",
        "    args,\n",
        "    data_collator=data_collator,\n",
        "    tokenizer=marian_tokenizer_vi_en,\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 184
        },
        "id": "yrU1Xk7Vlzry",
        "outputId": "4c8e3b27-6912-4bb3-c3b9-453dee20b537"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Translate from Vietnamese to English and decode predicted tokens\n",
        "# to get translated sentences\n",
        "###\n",
        "\n",
        "predictions = trainer.predict(ds_vi)\n",
        "\n",
        "translated_en_sentences = []\n",
        "for p in tqdm(predictions[0]):\n",
        "  translated_en_sentences.append(marian_tokenizer_vi_en.decode(p, skip_special_tokens=True))\n",
        "\n",
        "print(translated_en_sentences[:3])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2WZh5HrQmLmo"
      },
      "source": [
        "# Second Phase\n",
        "English -> Target language"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "3fnDr1RKmIVh",
        "outputId": "28e910e4-12f9-4262-969a-20cfe283b3ec"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Create dataframe with English translations + target language\n",
        "# and add special token to identify target language\n",
        "###\n",
        "\n",
        "test_df_en_target = pd.DataFrame(translated_en_sentences)\n",
        "test_df_en_target[target_lang] = test_df[target_lang].values\n",
        "test_df_en_target = test_df_en_target.rename(columns={0:\"en\"})\n",
        "test_df_en_target[\"en\"] = \"<2\" + target_lang + \"> \" + test_df_en_target[\"en\"]\n",
        "test_df_en_target.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "b3ae8b60f20f4a1baced297f723261a3",
            "a2394b5dec184c56872f5c8323c4da3f",
            "52e141f90ce54bc9be86c5903ccc5688",
            "6ab1ae48d09a4e1b9d5eb446c3ca388c",
            "c692fd69c6c047d7b806617f01cae05d",
            "ffa99ffe0eac453585f135509686baad",
            "fca0ba9c83724fabbfc344ef7c4284aa",
            "4cfdb6a1f7dc4137ba4d410375fb25ff",
            "1f81fa7986db41f98b8edfe133ed27b2",
            "58093ae89b33423a9aa03911b3f53c71",
            "2fe1f6e175564bd5ae78c6d04ce7fdc3"
          ]
        },
        "id": "9OOetIhRNoBq",
        "outputId": "f46591fb-23a3-4a00-d14f-51c10c80a3d3"
      },
      "outputs": [],
      "source": [
        "####\n",
        "# Tokenize the sentences in the dataset\n",
        "####\n",
        "\n",
        "from transformers import AutoTokenizer, MBart50TokenizerFast, MarianTokenizer\n",
        "import random\n",
        "from tqdm import tqdm\n",
        "import transformers    \n",
        "\n",
        "\n",
        "transformers.logging.set_verbosity(transformers.logging.CRITICAL)\n",
        "\n",
        "# create dataset objects\n",
        "test_ds_en_target = Dataset.from_pandas(test_df_en_target)\n",
        "\n",
        "max_input_length = 64\n",
        "max_target_length = 64\n",
        "batch_size = 16\n",
        "\n",
        "# To tokenize English, use a basic Marian tokenizer (the extended version\n",
        "# has some problems when tokenizing English)\n",
        "pure_marian_tokenizer = AutoTokenizer.from_pretrained(\"Helsinki-NLP/opus-mt-en-zh\")\n",
        "pure_marian_tokenizer.add_tokens([\"<2zh>\", \"<2\"+target_lang+\">\"], special_tokens=True)\n",
        "\n",
        "def preprocess_function(examples):\n",
        "    targets = [s for s in examples[target_lang]]\n",
        "    inputs = [s for s in examples[\"en\"]]\n",
        "\n",
        "    model_inputs = pure_marian_tokenizer(inputs, max_length=max_input_length, truncation=True, padding=\"max_length\")\n",
        "    # Setup the tokenizer for targets\n",
        "    with marian_tokenizer_en_target.as_target_tokenizer():\n",
        "        labels = marian_tokenizer_en_target(targets, max_length=max_target_length, truncation=True, padding=\"max_length\")\n",
        "\n",
        "    model_inputs[\"labels\"] = labels[\"input_ids\"]\n",
        "    \n",
        "    return model_inputs\n",
        "\n",
        "\n",
        "test_ds_en_target = test_ds_en_target.map(preprocess_function, batched=True, batch_size=batch_size)\n",
        "test_ds_en_target.set_format(type='torch', columns=columns_to_return)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "d5a1c42ec1e04c7fa628b8883e1c5c44",
            "350fa20c50f648488e14df0d23921c67",
            "ea6c1ea23b564584bee1489edead44b5",
            "26a28ade5afd45b19467311acee0429d",
            "ed7ef335cdcf44f18f042082517ed153",
            "3425293e285b47b996210a7b694a7d57",
            "37a54adf9bf14af5ac08c33831a11879",
            "3831e042e50647c88e95ff66e90e9ffa",
            "27b784a5c4504601b8ea1796577cbb78",
            "1067054c612643928d923d09e720d6e3",
            "bb04204b2bf5472483427f1ff4728f26"
          ]
        },
        "id": "IhIdjF_L59ju",
        "outputId": "a92144ab-834c-4b7f-a644-4639d75a4056"
      },
      "outputs": [],
      "source": [
        "####\n",
        "# Define the function to compute the BLEU score during training\n",
        "####\n",
        "from datasets import load_metric\n",
        "import numpy as np\n",
        "metric = load_metric(\"sacrebleu\")\n",
        "\n",
        "def postprocess_text(preds, labels):\n",
        "    preds = [pred.strip() for pred in preds]\n",
        "    labels = [[label.strip()] for label in labels]\n",
        "    return preds, labels\n",
        "    \n",
        "def compute_metrics(eval_preds):\n",
        "    preds, labels = eval_preds\n",
        "    if isinstance(preds, tuple):\n",
        "        preds = preds[0]\n",
        "\n",
        "    decoded_preds = marian_tokenizer_en_target.batch_decode(preds, skip_special_tokens=True)\n",
        "\n",
        "    # Replace -100 in the labels as we can't decode them.\n",
        "    labels = np.where(labels != -100, labels, marian_tokenizer_en_target.pad_token_id)\n",
        "    decoded_labels = marian_tokenizer_en_target.batch_decode(labels, skip_special_tokens=True)\n",
        "\n",
        "    # Some simple post-processing\n",
        "    decoded_preds, decoded_labels = postprocess_text(decoded_preds, decoded_labels)\n",
        "    print(\"Decoded preds: \", decoded_preds[0:3])\n",
        "    print(\"Decoded labels: \", decoded_labels[0:3])\n",
        "\n",
        "    result = metric.compute(predictions=decoded_preds, references=decoded_labels)\n",
        "    result = {\"bleu\": result[\"score\"]}\n",
        "\n",
        "    prediction_lens = [np.count_nonzero(pred != marian_tokenizer_en_target.pad_token_id) for pred in preds]\n",
        "    result[\"gen_len\"] = np.mean(prediction_lens)\n",
        "    result = {k: round(v, 4) for k, v in result.items()}\n",
        "    return result"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "imU4CMYhjQAu"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Download the pretrained en-target language model from the hub\n",
        "###\n",
        "\n",
        "model_en_target = AutoModelForSeq2SeqLM.from_pretrained(checkpoint_en_target)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CTuB-IGrV6nI",
        "outputId": "e605eb7b-14f8-4dba-811e-ad1fadc689c4"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Define training arguments\n",
        "###\n",
        "\n",
        "args = Seq2SeqTrainingArguments(\n",
        "  \".\",\n",
        "  evaluation_strategy = \"epoch\",\n",
        "  save_strategy=\"epoch\",\n",
        "  learning_rate=2e-5,\n",
        "  per_device_train_batch_size=batch_size,\n",
        "  per_device_eval_batch_size=batch_size,\n",
        "  weight_decay=0.01,\n",
        "  save_total_limit=1,\n",
        "  num_train_epochs=5,\n",
        "  predict_with_generate=True,\n",
        "  fp16=True, #CUDA purposes,\n",
        "  disable_tqdm=False,\n",
        ")\n",
        "\n",
        "data_collator = DataCollatorForSeq2Seq(marian_tokenizer_en_target, model=model_en_target)\n",
        "\n",
        "trainer = Seq2SeqTrainer(\n",
        "    model_en_target,\n",
        "    args,\n",
        "    data_collator=data_collator,\n",
        "    tokenizer=marian_tokenizer_en_target,\n",
        "    compute_metrics=compute_metrics\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "bVorp02rlYyK",
        "outputId": "bc912fe2-6a23-4084-a2d4-42f9825707f4"
      },
      "outputs": [],
      "source": [
        "###\n",
        "# Translate from English to target language and evaluate predictions\n",
        "###\n",
        "\n",
        "print(trainer.predict(test_ds_en_target))"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "name": "Low_resource_MT_vi_indo.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "02a8ad4bf5524174bf664b337475bd03": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1067054c612643928d923d09e720d6e3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "1f81fa7986db41f98b8edfe133ed27b2": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "26a28ade5afd45b19467311acee0429d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_27b784a5c4504601b8ea1796577cbb78",
            "max": 2356,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_3831e042e50647c88e95ff66e90e9ffa",
            "value": 2356
          }
        },
        "27b784a5c4504601b8ea1796577cbb78": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2fe1f6e175564bd5ae78c6d04ce7fdc3": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3425293e285b47b996210a7b694a7d57": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "350fa20c50f648488e14df0d23921c67": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "37a54adf9bf14af5ac08c33831a11879": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3831e042e50647c88e95ff66e90e9ffa": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "3fbec5804cfe48aeaf63a87b4dde033d": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "41f56499ff5f494f9a78e469ca8135a5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "4cfdb6a1f7dc4137ba4d410375fb25ff": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "4d23daee9f5a4039afca41b513fa5ea7": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "52e141f90ce54bc9be86c5903ccc5688": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fca0ba9c83724fabbfc344ef7c4284aa",
            "placeholder": "​",
            "style": "IPY_MODEL_ffa99ffe0eac453585f135509686baad",
            "value": "100%"
          }
        },
        "58093ae89b33423a9aa03911b3f53c71": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6ab1ae48d09a4e1b9d5eb446c3ca388c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1f81fa7986db41f98b8edfe133ed27b2",
            "max": 53,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_4cfdb6a1f7dc4137ba4d410375fb25ff",
            "value": 53
          }
        },
        "797ef0e56b584c3b81b7bf73ad447a4d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c7fb557172374bae84ed90f81fc8913a",
            "placeholder": "​",
            "style": "IPY_MODEL_c89ab218dd4048c294f669b29c5d45a2",
            "value": "100%"
          }
        },
        "90ba8e9c63ec4e1f878ef30d3f277a2c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_797ef0e56b584c3b81b7bf73ad447a4d",
              "IPY_MODEL_aa5dc610fe4842278e043bc9b1ad6def",
              "IPY_MODEL_ba4f9433d6824adf8e330e3e1087c787"
            ],
            "layout": "IPY_MODEL_02a8ad4bf5524174bf664b337475bd03"
          }
        },
        "a2394b5dec184c56872f5c8323c4da3f": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "aa5dc610fe4842278e043bc9b1ad6def": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3fbec5804cfe48aeaf63a87b4dde033d",
            "max": 53,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_41f56499ff5f494f9a78e469ca8135a5",
            "value": 53
          }
        },
        "b3ae8b60f20f4a1baced297f723261a3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_52e141f90ce54bc9be86c5903ccc5688",
              "IPY_MODEL_6ab1ae48d09a4e1b9d5eb446c3ca388c",
              "IPY_MODEL_c692fd69c6c047d7b806617f01cae05d"
            ],
            "layout": "IPY_MODEL_a2394b5dec184c56872f5c8323c4da3f"
          }
        },
        "ba4f9433d6824adf8e330e3e1087c787": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4d23daee9f5a4039afca41b513fa5ea7",
            "placeholder": "​",
            "style": "IPY_MODEL_f571536339634855bcc0240631f8020d",
            "value": " 53/53 [00:02&lt;00:00, 23.42ba/s]"
          }
        },
        "bb04204b2bf5472483427f1ff4728f26": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c692fd69c6c047d7b806617f01cae05d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2fe1f6e175564bd5ae78c6d04ce7fdc3",
            "placeholder": "​",
            "style": "IPY_MODEL_58093ae89b33423a9aa03911b3f53c71",
            "value": " 53/53 [00:01&lt;00:00, 41.19ba/s]"
          }
        },
        "c7fb557172374bae84ed90f81fc8913a": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c89ab218dd4048c294f669b29c5d45a2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d5a1c42ec1e04c7fa628b8883e1c5c44": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_ea6c1ea23b564584bee1489edead44b5",
              "IPY_MODEL_26a28ade5afd45b19467311acee0429d",
              "IPY_MODEL_ed7ef335cdcf44f18f042082517ed153"
            ],
            "layout": "IPY_MODEL_350fa20c50f648488e14df0d23921c67"
          }
        },
        "ea6c1ea23b564584bee1489edead44b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_37a54adf9bf14af5ac08c33831a11879",
            "placeholder": "​",
            "style": "IPY_MODEL_3425293e285b47b996210a7b694a7d57",
            "value": "Downloading: "
          }
        },
        "ed7ef335cdcf44f18f042082517ed153": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_bb04204b2bf5472483427f1ff4728f26",
            "placeholder": "​",
            "style": "IPY_MODEL_1067054c612643928d923d09e720d6e3",
            "value": " 5.65k/? [00:00&lt;00:00, 116kB/s]"
          }
        },
        "f571536339634855bcc0240631f8020d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "fca0ba9c83724fabbfc344ef7c4284aa": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ffa99ffe0eac453585f135509686baad": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}