train_me.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PBiHryhhrq4X",
        "outputId": "eac1d9b8-96db-4892-b127-c374998291b4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Fri Jul 19 19:55:34 2024       \n",
            "+-----------------------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 555.52.01              Driver Version: 555.99         CUDA Version: 12.5     |\n",
            "|-----------------------------------------+------------------------+----------------------+\n",
            "| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |\n",
            "|                                         |                        |               MIG M. |\n",
            "|=========================================+========================+======================|\n",
            "|   0  NVIDIA GeForce RTX 4070 Ti     On  |   00000000:01:00.0 Off |                  N/A |\n",
            "|  0%   33C    P8             12W /  285W |    6934MiB /  12282MiB |      0%      Default |\n",
            "|                                         |                        |                  N/A |\n",
            "+-----------------------------------------+------------------------+----------------------+\n",
            "                                                                                         \n",
            "+-----------------------------------------------------------------------------------------+\n",
            "| Processes:                                                                              |\n",
            "|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |\n",
            "|        ID   ID                                                               Usage      |\n",
            "|=========================================================================================|\n",
            "|    0   N/A  N/A     38985      C   /python3.8                                  N/A      |\n",
            "+-----------------------------------------------------------------------------------------+\n"
          ]
        }
      ],
      "source": [
        "gpu_info = !nvidia-smi\n",
        "gpu_info = '\\n'.join(gpu_info)\n",
        "if gpu_info.find('failed') >= 0:\n",
        "  print('Not connected to a GPU')\n",
        "else:\n",
        "  print(gpu_info)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "id": "OwVuvjriq4ay"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "!pip install datasets\n",
        "!pip install --upgrade transformers\n",
        "!pip install torchaudio\n",
        "!pip install jiwer\n",
        "!pip install accelerate -U"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "%%capture\n",
        "!apt install git-lfs"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 145,
          "referenced_widgets": [
            "f26d748bd1ac44c8afd67ce029cba0cb",
            "3a389a0f8a144232924d19c73055e821",
            "2187ceff542d4548a0e593ebdec3807e",
            "6d770b8c4fe540ec8fa228aa3c99cdd3",
            "46c967b7e4804ba2a9f9b3d57dd0a0b5",
            "bd5cac7319be40dda5339e51f7ec1e17",
            "0bceadc4b165475185e9cdb084dc46a4",
            "bd5e74f835744aa4a57e00fb6d4f697f",
            "5bc6b1e4202044c7a32ab8a2a196f199",
            "49d234c33fb2440ea62eef42cb2c40c2",
            "72cd60f6ecb94a448d1b4b94266aaf43",
            "da4200d2f3f34f2e83965cb821900dd1",
            "43b635966d0f4509a07ed7b5d0a1b42d",
            "b08abeab897e4da98e1e00998286460c"
          ]
        },
        "id": "YwL0481orOdC",
        "outputId": "7a1abbe3-585f-49d7-9054-da1b7c1a5b46"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "ee8a92bbb48a4e82af9b5b3bd316a022",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "VBox(children=(HTML(value='<center> <img\\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.sv…"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from huggingface_hub import notebook_login\n",
        "notebook_login()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/",
          "height": 311,
          "referenced_widgets": [
            "5ce0a5dabce344429634fb0ff1b8ece9",
            "c8a1e5dbf15e46c6a3fab0a49eda9bca",
            "eb3a0e4f8cdd479aac3c84c2bb6b5cbc",
            "39db6d96b9e143aaa66f68dd7bae94e1",
            "f6d7d5bc46ab4d5ab69fe4e427db7b4a",
            "2291054c0c1f40ecb84ab48a0c8cfd32",
            "0d3f93a4fc2345b28ce8f353b34d3de7",
            "55ca041839bd4dbc97674f393bd3438b",
            "9b918ab414f84769b0c1ec7be59a1a95",
            "9fccf8ede61a45feabb319cb8f400bb3",
            "5851ee0cc17b43739383c1aa2c249f53",
            "da3e59f9ac684cc7a2072613f29bc801",
            "9b9ab70a5aed424fa7ef0ab5e12c81f4",
            "a41942b9e24b4b71adc496655870c36f",
            "34ba9b1d517946678be4d72ffaaf0bc8",
            "acbc6af33d70406a8b4ab431bb0c5c34",
            "6e29c0ae30c64266855bab1192843586",
            "49892946f8d34cb2b91458a9e6b76812",
            "1f0d145e17fc4a80b83f5a2b712c90c5",
            "ded4569fc4df4b97bfa631c2d27945b0",
            "7c4ff0a541be4a59802286a89e6b94a3",
            "d9f17359d205471b8e49643ff54d906b",
            "7372c621be0e4f12a48e1f333c9d355c",
            "d10749e6916448b291ad6b9efd7c4ff5",
            "d0cb97c746ab4f2593491b07f763d8ed",
            "811b368dda024f76a2c26731f05f0d7a",
            "b1bf1ee0fdaf448d93b63608503615b0",
            "77933009195a4126b51519c0817dfb25",
            "454570e965b8493f8a9d72f686f07ea0",
            "cdb93e1e689b44a3b679b83bd9fbf45f",
            "51963fdba2334a079fade7581ede98c6",
            "ea66228a400d4d498895b15c5c67eb94",
            "37b51e7ceec441a6a34bbfa0bccaf6d2",
            "712a2f8c01634b11abba9804926d65f9",
            "38ba58519f8841ef9f33c24be3bcbf60",
            "348fe98eff164ec984625e66b544472f",
            "a010622b6dfd49098fc2d67c849ce2fa",
            "337a822296e943bbbd3e6b450bdfe9ed",
            "68cc23d7a26340a28664039be2072e81",
            "cdcc80620ec54ddea56598075e2aa37f",
            "3f15a9285d3146e9a0b881f22bc4647e",
            "2614e13f16fa4708954da60f78789658",
            "0d3c8574648a4cf3b7dbab5377db255a",
            "5350dfb0c4294c13a7f979e7c88e29fc"
          ]
        },
        "id": "SnEACAEvrOhO",
        "outputId": "932f37bb-55c9-40fc-ed60-ebf81d7aa33e"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "d91eb1572ab84da1bb8eebaa21ec225b",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Resolving data files:   0%|          | 0/31 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "d492b262ff5b47c5977c65069957df2d",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Loading dataset shards:   0%|          | 0/30 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from datasets import load_dataset, load_metric, Audio\n",
        "import os\n",
        "name=\"nateraw/fsd50k\"\n",
        "train = load_dataset(name,  split=\"train\", use_auth_token=True) \n",
        "train = train.remove_columns([\"title\", \"mids\", \"description\", \"tags\", \"license\", \"uploader\",\"split\"])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "6-WUy9qMrOqm",
        "outputId": "01bedafa-1903-4194-888a-589f3a50b8c0"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>labels</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Gunshot_and_gunfire,Explosion</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Fire</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Piano,Keyboard_(musical),Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Gasp,Breathing,Respiratory_sounds</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Trumpet,Brass_instrument,Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Guitar,Plucked_string_instrument,Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Wind_instrument_and_woodwind_instrument,Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Marimba_and_xylophone,Mallet_percussion,Percussion,Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Wind_instrument_and_woodwind_instrument,Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Wind_instrument_and_woodwind_instrument,Musical_instrument,Music</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import random\n",
        "import pandas as pd\n",
        "from IPython.display import display, HTML\n",
        "\n",
        "def show_random_elements(dataset, num_examples=10):\n",
        "    assert num_examples <= len(dataset), \"Can't pick more elements than there are in the dataset.\" \n",
        "    picks = []\n",
        "    for _ in range(num_examples):\n",
        "        pick = random.randint(0, len(dataset)-1)\n",
        "        while pick in picks:\n",
        "            pick = random.randint(0, len(dataset)-1)\n",
        "        picks.append(pick)\n",
        "\n",
        "    df = pd.DataFrame(dataset[picks])\n",
        "    display(HTML(df.to_html()))\n",
        "\n",
        "show_random_elements(train.remove_columns([\"audio\"]), num_examples=10)\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "background_save": true,
          "referenced_widgets": [
            "94cdb693a4dd433ab37384b7012263da",
            "995654c80003475686f74594c2c063cb"
          ]
        },
        "id": "aAB7f0fzrOzG",
        "outputId": "a65676e4-ee3c-4f93-831b-b4798b44aba4"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>labels</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Female speech and woman speaking Speech Human voice</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Keyboard musical Musical instrument Music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Insect Wild animals Animal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Laughter Human voice</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Liquid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Female speech and woman speaking Human voice Speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Gasp Breathing Respiratory sounds</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Burping and eructation</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Dishes and pots and pans Chink and clink Domestic sounds and home sounds Glass</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Human voice Cough Respiratory sounds</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "import re\n",
        "def clean_labels(batch):\n",
        "    batch[\"labels\"] = [re.sub(r'_', ' ', item) for item in batch[\"labels\"]]\n",
        "    batch[\"labels\"] = [re.sub(r',', ' ', item) for item in batch[\"labels\"]]\n",
        "    batch[\"labels\"] = [re.sub(r'\\)', '', item) for item in batch[\"labels\"]]\n",
        "    batch[\"labels\"] = [re.sub(r\"\\(\", '', item) for item in batch[\"labels\"]]\n",
        "\n",
        "    return batch\n",
        "train = train.map(clean_labels, batched=True, batch_size=1000) \n",
        "\n",
        "show_random_elements(train.remove_columns([\"audio\"]), num_examples=10)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "IoZuUzuorPEL"
      },
      "outputs": [],
      "source": [
        "train = train.cast_column(\"audio\", Audio(sampling_rate=16000))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "4ch1hydvrPKZ",
        "outputId": "5b79d5d7-9259-461b-9fe9-9d1c36895909"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "                <audio  controls=\"controls\" autoplay=\"autoplay\">\n",
              "                    <source src=\"data:audio/wav;base64,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\" type=\"audio/wav\" />\n",
              "                    Your browser does not support the audio element.\n",
              "                </audio>\n",
              "              "
            ],
            "text/plain": [
              "<IPython.lib.display.Audio object>"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import IPython.display as ipd\n",
        "import numpy as np\n",
        "import random\n",
        "rand_int = random.randint(0, len(train)-1)\n",
        "\n",
        "print(train[rand_int][\"labels\"])\n",
        "ipd.Audio(data=train[rand_int][\"audio\"][\"array\"], autoplay=True, rate=16000)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "G7GzhtbfrPMW",
        "outputId": "2f2e9eed-b773-4bdc-dc35-d954d9879214"
      },
      "outputs": [],
      "source": [
        "rand_int = random.randint(0, len(train)-1)\n",
        "\n",
        "print(\"Target text:\", train[rand_int][\"labels\"])\n",
        "print(\"Input array shape:\", train[rand_int][\"audio\"][\"array\"].shape)\n",
        "print(\"Sampling rate:\", train[rand_int][\"audio\"][\"sampling_rate\"])\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "b2BIdybCuRL5"
      },
      "outputs": [],
      "source": [
        "from transformers import SeamlessM4TFeatureExtractor\n",
        "feature_extractor = SeamlessM4TFeatureExtractor.from_pretrained(\"facebook/w2v-bert-2.0\",stride = 2)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {},
      "outputs": [],
      "source": [
        "import torch\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "2fd05b118e88418f8eaa382e1df8ba9b",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Loading pipeline components...:   0%|          | 0/6 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "text/plain": [
              "CLIPTextModel(\n",
              "  (text_model): CLIPTextTransformer(\n",
              "    (embeddings): CLIPTextEmbeddings(\n",
              "      (token_embedding): Embedding(49408, 1024)\n",
              "      (position_embedding): Embedding(77, 1024)\n",
              "    )\n",
              "    (encoder): CLIPEncoder(\n",
              "      (layers): ModuleList(\n",
              "        (0): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (1): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (2): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (3): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (4): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (5): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (6): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (7): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (8): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (9): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (10): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (11): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (12): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (13): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (14): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (15): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (16): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (17): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (18): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (19): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (20): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (21): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (22): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "      )\n",
              "    )\n",
              "    (final_layer_norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "  )\n",
              ")"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from diffusers import StableDiffusionPipeline\n",
        "pipe=StableDiffusionPipeline.from_pretrained(\"stabilityai/stable-diffusion-2-1\")\n",
        "text_encoder=pipe.components[\"text_encoder\"]\n",
        "tokenizer=pipe.components[\"tokenizer\"]\n",
        "print(tokenizer)\n",
        "text_encoder.to(device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "mCoOLo84rPPx"
      },
      "outputs": [],
      "source": [
        "def prepare_dataset(batch):\n",
        "    audio = batch[\"audio\"]\n",
        "    labels= batch[\"labels\"]\n",
        "    features=feature_extractor(audio[\"array\"], sampling_rate=audio[\"sampling_rate\"], truncation=True, padding='max_length', max_length=375)\n",
        "    batch[\"input_features\"],batch[\"attention_mask\"] = features.input_features[0], features.attention_mask[0]\n",
        "    batch[\"input_length\"] = len(batch[\"input_features\"])\n",
        "    tokens=tokenizer(labels,return_tensors=\"pt\", truncation=True, padding='max_length', max_length=48)\n",
        "    with torch.no_grad():\n",
        "        batch[\"label\"] = text_encoder(**tokens.to(device)).last_hidden_state.squeeze(0).cpu()    \n",
        "    return batch\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "background_save": true,
          "referenced_widgets": [
            "2846c53a8fff4b4fa5eccec9fba2dff8",
            "3a5d309da97d4674af91003f1f4822ed"
          ]
        },
        "id": "jtzlzMsxrPTF",
        "outputId": "f733454a-39bc-46e1-a4f1-1f83e06e08ec"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "CLIPTextModel(\n",
              "  (text_model): CLIPTextTransformer(\n",
              "    (embeddings): CLIPTextEmbeddings(\n",
              "      (token_embedding): Embedding(49408, 1024)\n",
              "      (position_embedding): Embedding(77, 1024)\n",
              "    )\n",
              "    (encoder): CLIPEncoder(\n",
              "      (layers): ModuleList(\n",
              "        (0): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (1): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (2): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (3): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (4): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (5): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (6): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (7): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (8): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (9): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (10): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (11): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (12): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (13): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (14): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (15): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (16): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (17): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (18): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (19): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (20): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (21): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "        (22): CLIPEncoderLayer(\n",
              "          (self_attn): CLIPAttention(\n",
              "            (k_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (v_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (q_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "            (out_proj): Linear(in_features=1024, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "          (mlp): CLIPMLP(\n",
              "            (activation_fn): GELUActivation()\n",
              "            (fc1): Linear(in_features=1024, out_features=4096, bias=True)\n",
              "            (fc2): Linear(in_features=4096, out_features=1024, bias=True)\n",
              "          )\n",
              "          (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "        )\n",
              "      )\n",
              "    )\n",
              "    (final_layer_norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n",
              "  )\n",
              ")"
            ]
          },
          "execution_count": 14,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train = train.map(prepare_dataset,remove_columns=train.column_names,batch_size=100)\n",
        "text_encoder.to('cpu')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {},
      "outputs": [],
      "source": [
        "del pipe\n",
        "del text_encoder\n",
        "import gc\n",
        "gc.collect()\n",
        "torch.cuda.empty_cache()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "dict_keys(['input_features', 'attention_mask', 'input_length', 'label'])"
            ]
          },
          "execution_count": 16,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train[0].keys()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {},
      "outputs": [],
      "source": [
        "split_dataset = train.train_test_split(test_size=0.1)\n",
        "train_subset = split_dataset['train']\n",
        "test_subset = split_dataset['test']\n",
        "print(f\"Train subset size: {len(train_subset)}\")\n",
        "print(f\"Test subset size: {len(test_subset)}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "GRib3ZJFrPWe"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "from dataclasses import dataclass\n",
        "@dataclass\n",
        "class DataCollator:\n",
        "\n",
        "    def __call__(self, features): \n",
        "        input_features = [torch.tensor(feature[\"input_features\"], dtype=torch.float32) for feature in features]\n",
        "        attention_mask = [torch.tensor(feature[\"attention_mask\"], dtype=torch.float32) for feature in features]  \n",
        "        label_features = [torch.tensor(feature[\"label\"], dtype=torch.float32) for feature in features] \n",
        "\n",
        "\n",
        "        batch = {\n",
        "            \"input_features\": torch.stack(input_features),\n",
        "            \"attention_mask\": torch.stack(attention_mask), \n",
        "            \"text_embeddings\": torch.stack(label_features)[:, 1:, :]\n",
        "        }\n",
        "    \n",
        "\n",
        "        return batch\n",
        "data_collator = DataCollator()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "background_save": true,
          "referenced_widgets": [
            "460c0924a5b940f7adfd315ad910b456",
            "1cd903b686bd403aa905ca70ffd0cf16"
          ]
        },
        "id": "MQ3vK2bzEqSJ",
        "outputId": "edbf4536-b49d-4e70-b876-94b3cec5562f"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Some weights of Wav2Vec2BertModel were not initialized from the model checkpoint at facebook/w2v-bert-2.0 and are newly initialized: ['adapter.layers.0.ffn.intermediate_dense.bias', 'adapter.layers.0.ffn.intermediate_dense.weight', 'adapter.layers.0.ffn.output_dense.bias', 'adapter.layers.0.ffn.output_dense.weight', 'adapter.layers.0.ffn_layer_norm.bias', 'adapter.layers.0.ffn_layer_norm.weight', 'adapter.layers.0.residual_conv.bias', 'adapter.layers.0.residual_conv.weight', 'adapter.layers.0.residual_layer_norm.bias', 'adapter.layers.0.residual_layer_norm.weight', 'adapter.layers.0.self_attn.linear_k.bias', 'adapter.layers.0.self_attn.linear_k.weight', 'adapter.layers.0.self_attn.linear_out.bias', 'adapter.layers.0.self_attn.linear_out.weight', 'adapter.layers.0.self_attn.linear_q.bias', 'adapter.layers.0.self_attn.linear_q.weight', 'adapter.layers.0.self_attn.linear_v.bias', 'adapter.layers.0.self_attn.linear_v.weight', 'adapter.layers.0.self_attn_conv.bias', 'adapter.layers.0.self_attn_conv.weight', 'adapter.layers.0.self_attn_layer_norm.bias', 'adapter.layers.0.self_attn_layer_norm.weight', 'adapter.layers.1.ffn.intermediate_dense.bias', 'adapter.layers.1.ffn.intermediate_dense.weight', 'adapter.layers.1.ffn.output_dense.bias', 'adapter.layers.1.ffn.output_dense.weight', 'adapter.layers.1.ffn_layer_norm.bias', 'adapter.layers.1.ffn_layer_norm.weight', 'adapter.layers.1.residual_conv.bias', 'adapter.layers.1.residual_conv.weight', 'adapter.layers.1.residual_layer_norm.bias', 'adapter.layers.1.residual_layer_norm.weight', 'adapter.layers.1.self_attn.linear_k.bias', 'adapter.layers.1.self_attn.linear_k.weight', 'adapter.layers.1.self_attn.linear_out.bias', 'adapter.layers.1.self_attn.linear_out.weight', 'adapter.layers.1.self_attn.linear_q.bias', 'adapter.layers.1.self_attn.linear_q.weight', 'adapter.layers.1.self_attn.linear_v.bias', 'adapter.layers.1.self_attn.linear_v.weight', 'adapter.layers.1.self_attn_conv.bias', 'adapter.layers.1.self_attn_conv.weight', 'adapter.layers.1.self_attn_layer_norm.bias', 'adapter.layers.1.self_attn_layer_norm.weight']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ]
        }
      ],
      "source": [
        "from transformers import Wav2Vec2BertModel\n",
        "\n",
        "model = Wav2Vec2BertModel.from_pretrained(\n",
        "    \"facebook/w2v-bert-2.0\",\n",
        "    add_adapter=True,\n",
        "    adapter_kernel_size=3,\n",
        "    adapter_stride=2,\n",
        "    num_adapter_layers=2,\n",
        "    layerdrop=0.0,\n",
        "    )\n",
        "\n",
        "print(next(model.parameters()).device)  \n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {},
      "outputs": [],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "def compute_metrics(pred):\n",
        "    predictions = pred.predictions\n",
        "    labels = pred.label_ids\n",
        "\n",
        "    tensor_predictions = torch.from_numpy(predictions)\n",
        "    tensor_labels = torch.from_numpy(labels)\n",
        "\n",
        "    cosine_similarities = torch.cosine_similarity(tensor_predictions, tensor_labels, dim=-1)\n",
        "    avg_cos_sim = cosine_similarities.mean()\n",
        "\n",
        "    mse_loss = nn.MSELoss()\n",
        "    E_mse = mse_loss(tensor_predictions, tensor_labels)    \n",
        "\n",
        "    euclidean_distances = torch.norm(tensor_predictions - tensor_labels, dim=-1)\n",
        "    avg_euclidean_dist = euclidean_distances.mean().item()\n",
        "    \n",
        "    return {\n",
        "        'cosine': avg_cos_sim,\n",
        "        'E_mse': E_mse,\n",
        "        'euclidean_distance': avg_euclidean_dist,\n",
        "    }"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "7kIjDR5yN4yY"
      },
      "outputs": [],
      "source": [
        "from transformers import Trainer\n",
        "import torch.nn.functional as F\n",
        "\n",
        "def cosine_similarity_loss(embeddings1, embeddings2, margin=0.5):\n",
        "    cos_sim = torch.cosine_similarity(embeddings1, embeddings2,dim=-1)\n",
        "    loss = torch.mean(1 - cos_sim + margin)\n",
        "    return loss   \n",
        "\n",
        "\n",
        "def Contrastive_loss(embeddings1, embeddings2, temperature=0.15):\n",
        " \n",
        "    cos_sim = torch.cosine_similarity(embeddings1.unsqueeze(1), embeddings2.unsqueeze(0), dim=-1)\n",
        "    cos_sim = cos_sim / temperature \n",
        "    labels = torch.arange(embeddings1.size(0)).unsqueeze(1).repeat(1, embeddings1.size(1)).to(embeddings1.device)    \n",
        "    loss = F.cross_entropy(cos_sim, labels)   \n",
        "    return loss\n",
        "\n",
        "class TrainBert(Trainer):\n",
        "    def __init__(self,*args,**kwargs):\n",
        "        super().__init__(*args,**kwargs)\n",
        "  \n",
        "    def compute_loss(self, model, inputs, return_outputs=False):\n",
        "        labels = inputs.pop(\"text_embeddings\")\n",
        "        outputs = model(**inputs)\n",
        "        outputs=outputs.last_hidden_state\n",
        "        loss = Contrastive_loss(outputs, labels)\n",
        "        outputs = (loss, outputs)\n",
        "        return (loss, outputs) if return_outputs else loss\n",
        "    \n",
        "    "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "D3f3MHveEqPW"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/youssef/miniconda3/envs/ldmmm/lib/python3.8/site-packages/transformers/training_args.py:1525: FutureWarning: `evaluation_strategy` is deprecated and will be removed in version 4.46 of 🤗 Transformers. Use `eval_strategy` instead\n",
            "  warnings.warn(\n"
          ]
        }
      ],
      "source": [
        "from transformers import TrainingArguments\n",
        "\n",
        "training_args = TrainingArguments(\n",
        "  output_dir=\"train/w2v-bert_sd_conditioning\",\n",
        "  per_device_train_batch_size=8,\n",
        "  per_device_eval_batch_size=8,\n",
        "  gradient_accumulation_steps=16,\n",
        "  evaluation_strategy=\"steps\",\n",
        "  num_train_epochs=8,\n",
        "  gradient_checkpointing=True,\n",
        "  fp16=True,\n",
        "  save_steps=288,\n",
        "  eval_steps=144,  \n",
        "  logging_steps=144,\n",
        "  learning_rate=2e-5,\n",
        "  warmup_steps=200,\n",
        "  save_total_limit=4,\n",
        "  push_to_hub=False,\n",
        "  label_names =[\"text_embeddings\"],\n",
        "  logging_dir='train/logs/run1_full_training_full_data_(contrastive_loss_T_015_adapter_3*3)',\n",
        "  dataloader_num_workers=10, \n",
        "  eval_accumulation_steps=10,\n",
        "  #load_best_model_at_end=True,\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "mAvAO0TfEqMU"
      },
      "outputs": [],
      "source": [
        "trainer = TrainBert(\n",
        "    model=model,\n",
        "    data_collator=data_collator,\n",
        "    args=training_args,\n",
        "    train_dataset=train_subset,\n",
        "    eval_dataset=test_subset,\n",
        "    tokenizer=feature_extractor,\n",
        "    compute_metrics=compute_metrics,\n",
        ")\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='2304' max='2304' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [2304/2304 5:20:53, Epoch 7/8]\n",
              "    </div>\n",
              "    <table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              " <tr style=\"text-align: left;\">\n",
              "      <th>Step</th>\n",
              "      <th>Training Loss</th>\n",
              "      <th>Validation Loss</th>\n",
              "      <th>Cosine</th>\n",
              "      <th>E Mse</th>\n",
              "      <th>Euclidean Distance</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>144</td>\n",
              "      <td>1.913700</td>\n",
              "      <td>1.734753</td>\n",
              "      <td>0.085479</td>\n",
              "      <td>1.756623</td>\n",
              "      <td>42.263260</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>288</td>\n",
              "      <td>1.680900</td>\n",
              "      <td>1.613458</td>\n",
              "      <td>0.116268</td>\n",
              "      <td>1.659532</td>\n",
              "      <td>41.095066</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>432</td>\n",
              "      <td>1.551000</td>\n",
              "      <td>1.506113</td>\n",
              "      <td>0.147984</td>\n",
              "      <td>1.553063</td>\n",
              "      <td>39.733353</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>576</td>\n",
              "      <td>1.475700</td>\n",
              "      <td>1.440376</td>\n",
              "      <td>0.156046</td>\n",
              "      <td>1.438201</td>\n",
              "      <td>38.234638</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>720</td>\n",
              "      <td>1.405200</td>\n",
              "      <td>1.405672</td>\n",
              "      <td>0.172788</td>\n",
              "      <td>1.314867</td>\n",
              "      <td>36.545670</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>864</td>\n",
              "      <td>1.378100</td>\n",
              "      <td>1.372734</td>\n",
              "      <td>0.181113</td>\n",
              "      <td>1.272299</td>\n",
              "      <td>35.930172</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1008</td>\n",
              "      <td>1.325900</td>\n",
              "      <td>1.350391</td>\n",
              "      <td>0.190992</td>\n",
              "      <td>1.223559</td>\n",
              "      <td>35.238632</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1152</td>\n",
              "      <td>1.307500</td>\n",
              "      <td>1.338883</td>\n",
              "      <td>0.198897</td>\n",
              "      <td>1.205876</td>\n",
              "      <td>34.976425</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1296</td>\n",
              "      <td>1.267900</td>\n",
              "      <td>1.323583</td>\n",
              "      <td>0.204958</td>\n",
              "      <td>1.178422</td>\n",
              "      <td>34.572132</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1440</td>\n",
              "      <td>1.259100</td>\n",
              "      <td>1.314878</td>\n",
              "      <td>0.202416</td>\n",
              "      <td>1.158960</td>\n",
              "      <td>34.271767</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1584</td>\n",
              "      <td>1.225000</td>\n",
              "      <td>1.306984</td>\n",
              "      <td>0.212688</td>\n",
              "      <td>1.125986</td>\n",
              "      <td>33.767563</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1728</td>\n",
              "      <td>1.214100</td>\n",
              "      <td>1.295199</td>\n",
              "      <td>0.207239</td>\n",
              "      <td>1.108922</td>\n",
              "      <td>33.509209</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1872</td>\n",
              "      <td>1.190000</td>\n",
              "      <td>1.294782</td>\n",
              "      <td>0.211648</td>\n",
              "      <td>1.077237</td>\n",
              "      <td>33.032864</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2016</td>\n",
              "      <td>1.187700</td>\n",
              "      <td>1.288203</td>\n",
              "      <td>0.221070</td>\n",
              "      <td>1.056833</td>\n",
              "      <td>32.714443</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2160</td>\n",
              "      <td>1.166700</td>\n",
              "      <td>1.288779</td>\n",
              "      <td>0.223201</td>\n",
              "      <td>1.040101</td>\n",
              "      <td>32.452045</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2304</td>\n",
              "      <td>1.168600</td>\n",
              "      <td>1.286108</td>\n",
              "      <td>0.219770</td>\n",
              "      <td>1.039849</td>\n",
              "      <td>32.444210</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><p>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "text/plain": [
              "TrainOutput(global_step=2304, training_loss=1.3572947515381708, metrics={'train_runtime': 19282.704, 'train_samples_per_second': 15.296, 'train_steps_per_second': 0.119, 'total_flos': 3.356981348459176e+19, 'train_loss': 1.3572947515381708, 'epoch': 7.998264265567368})"
            ]
          },
          "execution_count": 24,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "torch.cuda.empty_cache() \n",
        "import gc\n",
        "gc.collect()\n",
        "trainer.train()\n",
        "#trainer.train(resume_from_checkpoint=True)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "language": "python",
      "name": "python3"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "0bceadc4b165475185e9cdb084dc46a4": {
          "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
          }
        },
        "0d3c8574648a4cf3b7dbab5377db255a": {
          "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
          }
        },
        "0d3f93a4fc2345b28ce8f353b34d3de7": {
          "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": ""
          }
        },
        "1f0d145e17fc4a80b83f5a2b712c90c5": {
          "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
          }
        },
        "2187ceff542d4548a0e593ebdec3807e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "LabelModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "LabelModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "LabelView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5bc6b1e4202044c7a32ab8a2a196f199",
            "placeholder": "​",
            "style": "IPY_MODEL_49d234c33fb2440ea62eef42cb2c40c2",
            "value": "Your token has been saved in your configured git credential helpers (store)."
          }
        },
        "2291054c0c1f40ecb84ab48a0c8cfd32": {
          "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
          }
        },
        "2614e13f16fa4708954da60f78789658": {
          "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": ""
          }
        },
        "337a822296e943bbbd3e6b450bdfe9ed": {
          "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
          }
        },
        "348fe98eff164ec984625e66b544472f": {
          "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": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3f15a9285d3146e9a0b881f22bc4647e",
            "max": 40966,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_2614e13f16fa4708954da60f78789658",
            "value": 5288
          }
        },
        "34ba9b1d517946678be4d72ffaaf0bc8": {
          "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_7c4ff0a541be4a59802286a89e6b94a3",
            "placeholder": "​",
            "style": "IPY_MODEL_d9f17359d205471b8e49643ff54d906b",
            "value": " 1.15k/1.15k [00:00&lt;00:00, 33.5kB/s]"
          }
        },
        "37b51e7ceec441a6a34bbfa0bccaf6d2": {
          "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": ""
          }
        },
        "38ba58519f8841ef9f33c24be3bcbf60": {
          "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_68cc23d7a26340a28664039be2072e81",
            "placeholder": "​",
            "style": "IPY_MODEL_cdcc80620ec54ddea56598075e2aa37f",
            "value": "Generating train split:  13%"
          }
        },
        "39db6d96b9e143aaa66f68dd7bae94e1": {
          "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_9fccf8ede61a45feabb319cb8f400bb3",
            "placeholder": "​",
            "style": "IPY_MODEL_5851ee0cc17b43739383c1aa2c249f53",
            "value": " 31/31 [00:00&lt;00:00,  5.87it/s]"
          }
        },
        "3a389a0f8a144232924d19c73055e821": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "LabelModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "LabelModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "LabelView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0bceadc4b165475185e9cdb084dc46a4",
            "placeholder": "​",
            "style": "IPY_MODEL_bd5e74f835744aa4a57e00fb6d4f697f",
            "value": "Token is valid (permission: write)."
          }
        },
        "3f15a9285d3146e9a0b881f22bc4647e": {
          "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
          }
        },
        "43b635966d0f4509a07ed7b5d0a1b42d": {
          "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
          }
        },
        "454570e965b8493f8a9d72f686f07ea0": {
          "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": ""
          }
        },
        "46c967b7e4804ba2a9f9b3d57dd0a0b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "LabelModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "LabelModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "LabelView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_43b635966d0f4509a07ed7b5d0a1b42d",
            "placeholder": "​",
            "style": "IPY_MODEL_b08abeab897e4da98e1e00998286460c",
            "value": "Login successful"
          }
        },
        "49892946f8d34cb2b91458a9e6b76812": {
          "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": ""
          }
        },
        "49d234c33fb2440ea62eef42cb2c40c2": {
          "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": ""
          }
        },
        "51963fdba2334a079fade7581ede98c6": {
          "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": ""
          }
        },
        "5350dfb0c4294c13a7f979e7c88e29fc": {
          "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": ""
          }
        },
        "55ca041839bd4dbc97674f393bd3438b": {
          "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
          }
        },
        "5851ee0cc17b43739383c1aa2c249f53": {
          "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": ""
          }
        },
        "5bc6b1e4202044c7a32ab8a2a196f199": {
          "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
          }
        },
        "5ce0a5dabce344429634fb0ff1b8ece9": {
          "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_c8a1e5dbf15e46c6a3fab0a49eda9bca",
              "IPY_MODEL_eb3a0e4f8cdd479aac3c84c2bb6b5cbc",
              "IPY_MODEL_39db6d96b9e143aaa66f68dd7bae94e1"
            ],
            "layout": "IPY_MODEL_f6d7d5bc46ab4d5ab69fe4e427db7b4a"
          }
        },
        "68cc23d7a26340a28664039be2072e81": {
          "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
          }
        },
        "6d770b8c4fe540ec8fa228aa3c99cdd3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "LabelModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "LabelModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "LabelView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_72cd60f6ecb94a448d1b4b94266aaf43",
            "placeholder": "​",
            "style": "IPY_MODEL_da4200d2f3f34f2e83965cb821900dd1",
            "value": "Your token has been saved to /root/.cache/huggingface/token"
          }
        },
        "6e29c0ae30c64266855bab1192843586": {
          "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
          }
        },
        "712a2f8c01634b11abba9804926d65f9": {
          "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_38ba58519f8841ef9f33c24be3bcbf60",
              "IPY_MODEL_348fe98eff164ec984625e66b544472f",
              "IPY_MODEL_a010622b6dfd49098fc2d67c849ce2fa"
            ],
            "layout": "IPY_MODEL_337a822296e943bbbd3e6b450bdfe9ed"
          }
        },
        "72cd60f6ecb94a448d1b4b94266aaf43": {
          "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
          }
        },
        "7372c621be0e4f12a48e1f333c9d355c": {
          "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_d10749e6916448b291ad6b9efd7c4ff5",
              "IPY_MODEL_d0cb97c746ab4f2593491b07f763d8ed",
              "IPY_MODEL_811b368dda024f76a2c26731f05f0d7a"
            ],
            "layout": "IPY_MODEL_b1bf1ee0fdaf448d93b63608503615b0"
          }
        },
        "77933009195a4126b51519c0817dfb25": {
          "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
          }
        },
        "7c4ff0a541be4a59802286a89e6b94a3": {
          "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
          }
        },
        "811b368dda024f76a2c26731f05f0d7a": {
          "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_ea66228a400d4d498895b15c5c67eb94",
            "placeholder": "​",
            "style": "IPY_MODEL_37b51e7ceec441a6a34bbfa0bccaf6d2",
            "value": " 31/31 [28:09&lt;00:00, 62.60s/files]"
          }
        },
        "9b918ab414f84769b0c1ec7be59a1a95": {
          "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": ""
          }
        },
        "9b9ab70a5aed424fa7ef0ab5e12c81f4": {
          "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_6e29c0ae30c64266855bab1192843586",
            "placeholder": "​",
            "style": "IPY_MODEL_49892946f8d34cb2b91458a9e6b76812",
            "value": "Downloading metadata: 100%"
          }
        },
        "9fccf8ede61a45feabb319cb8f400bb3": {
          "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
          }
        },
        "a010622b6dfd49098fc2d67c849ce2fa": {
          "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_0d3c8574648a4cf3b7dbab5377db255a",
            "placeholder": "​",
            "style": "IPY_MODEL_5350dfb0c4294c13a7f979e7c88e29fc",
            "value": " 5288/40966 [00:25&lt;02:43, 218.50 examples/s]"
          }
        },
        "a41942b9e24b4b71adc496655870c36f": {
          "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_1f0d145e17fc4a80b83f5a2b712c90c5",
            "max": 1149,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ded4569fc4df4b97bfa631c2d27945b0",
            "value": 1149
          }
        },
        "acbc6af33d70406a8b4ab431bb0c5c34": {
          "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
          }
        },
        "b08abeab897e4da98e1e00998286460c": {
          "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": ""
          }
        },
        "b1bf1ee0fdaf448d93b63608503615b0": {
          "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
          }
        },
        "bd5cac7319be40dda5339e51f7ec1e17": {
          "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": "center",
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": "flex",
            "flex": null,
            "flex_flow": "column",
            "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": "50%"
          }
        },
        "bd5e74f835744aa4a57e00fb6d4f697f": {
          "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": ""
          }
        },
        "c8a1e5dbf15e46c6a3fab0a49eda9bca": {
          "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_2291054c0c1f40ecb84ab48a0c8cfd32",
            "placeholder": "​",
            "style": "IPY_MODEL_0d3f93a4fc2345b28ce8f353b34d3de7",
            "value": "Resolving data files: 100%"
          }
        },
        "cdb93e1e689b44a3b679b83bd9fbf45f": {
          "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
          }
        },
        "cdcc80620ec54ddea56598075e2aa37f": {
          "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": ""
          }
        },
        "d0cb97c746ab4f2593491b07f763d8ed": {
          "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_cdb93e1e689b44a3b679b83bd9fbf45f",
            "max": 31,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_51963fdba2334a079fade7581ede98c6",
            "value": 31
          }
        },
        "d10749e6916448b291ad6b9efd7c4ff5": {
          "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_77933009195a4126b51519c0817dfb25",
            "placeholder": "​",
            "style": "IPY_MODEL_454570e965b8493f8a9d72f686f07ea0",
            "value": "Downloading data: 100%"
          }
        },
        "d9f17359d205471b8e49643ff54d906b": {
          "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": ""
          }
        },
        "da3e59f9ac684cc7a2072613f29bc801": {
          "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_9b9ab70a5aed424fa7ef0ab5e12c81f4",
              "IPY_MODEL_a41942b9e24b4b71adc496655870c36f",
              "IPY_MODEL_34ba9b1d517946678be4d72ffaaf0bc8"
            ],
            "layout": "IPY_MODEL_acbc6af33d70406a8b4ab431bb0c5c34"
          }
        },
        "da4200d2f3f34f2e83965cb821900dd1": {
          "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": ""
          }
        },
        "ded4569fc4df4b97bfa631c2d27945b0": {
          "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": ""
          }
        },
        "ea66228a400d4d498895b15c5c67eb94": {
          "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
          }
        },
        "eb3a0e4f8cdd479aac3c84c2bb6b5cbc": {
          "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_55ca041839bd4dbc97674f393bd3438b",
            "max": 31,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_9b918ab414f84769b0c1ec7be59a1a95",
            "value": 31
          }
        },
        "f26d748bd1ac44c8afd67ce029cba0cb": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "VBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "VBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "VBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_3a389a0f8a144232924d19c73055e821",
              "IPY_MODEL_2187ceff542d4548a0e593ebdec3807e",
              "IPY_MODEL_6d770b8c4fe540ec8fa228aa3c99cdd3",
              "IPY_MODEL_46c967b7e4804ba2a9f9b3d57dd0a0b5"
            ],
            "layout": "IPY_MODEL_bd5cac7319be40dda5339e51f7ec1e17"
          }
        },
        "f6d7d5bc46ab4d5ab69fe4e427db7b4a": {
          "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
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}