Created
August 16, 2022 11:52
-
-
Save kacperlukawski/dbd2e80592a80d2f397c79e391b6b845 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "cells": [ | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "# Wine reviews dataset\n", | |
| "\n", | |
| "Kaggle's wine reviews dataset is a great choice to create a recommendation system." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 1, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:09:58.688155Z", | |
| "start_time": "2022-08-04T11:09:56.725326Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "wine-reviews.zip: Skipping, found more recently modified local copy (use --force to force download)\r\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "!kaggle datasets download zynicide/wine-reviews" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 2, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:09:59.031535Z", | |
| "start_time": "2022-08-04T11:09:58.695943Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "import pandas as pd\n", | |
| "from zipfile import ZipFile" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 3, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:00.682055Z", | |
| "start_time": "2022-08-04T11:09:59.040031Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/html": [ | |
| "<div>\n", | |
| "<style scoped>\n", | |
| " .dataframe tbody tr th:only-of-type {\n", | |
| " vertical-align: middle;\n", | |
| " }\n", | |
| "\n", | |
| " .dataframe tbody tr th {\n", | |
| " vertical-align: top;\n", | |
| " }\n", | |
| "\n", | |
| " .dataframe thead th {\n", | |
| " text-align: right;\n", | |
| " }\n", | |
| "</style>\n", | |
| "<table border=\"1\" class=\"dataframe\">\n", | |
| " <thead>\n", | |
| " <tr style=\"text-align: right;\">\n", | |
| " <th></th>\n", | |
| " <th>country</th>\n", | |
| " <th>description</th>\n", | |
| " <th>designation</th>\n", | |
| " <th>points</th>\n", | |
| " <th>price</th>\n", | |
| " <th>province</th>\n", | |
| " <th>region_1</th>\n", | |
| " <th>region_2</th>\n", | |
| " <th>taster_name</th>\n", | |
| " <th>taster_twitter_handle</th>\n", | |
| " <th>title</th>\n", | |
| " <th>variety</th>\n", | |
| " <th>winery</th>\n", | |
| " </tr>\n", | |
| " </thead>\n", | |
| " <tbody>\n", | |
| " <tr>\n", | |
| " <th>4323</th>\n", | |
| " <td>US</td>\n", | |
| " <td>Grown in the Sebastopol Hills area, this Pinot...</td>\n", | |
| " <td>Umino Vineyard</td>\n", | |
| " <td>89</td>\n", | |
| " <td>48.0</td>\n", | |
| " <td>California</td>\n", | |
| " <td>Sonoma Coast</td>\n", | |
| " <td>Sonoma</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>W.H. Smith 2007 Umino Vineyard Pinot Noir (Son...</td>\n", | |
| " <td>Pinot Noir</td>\n", | |
| " <td>W.H. Smith</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>50398</th>\n", | |
| " <td>Spain</td>\n", | |
| " <td>Coarse on the nose, this has scratchy aromas o...</td>\n", | |
| " <td>Limited Edition</td>\n", | |
| " <td>86</td>\n", | |
| " <td>17.0</td>\n", | |
| " <td>Northern Spain</td>\n", | |
| " <td>Rioja</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>Michael Schachner</td>\n", | |
| " <td>@wineschach</td>\n", | |
| " <td>Ramón Bilbao 2009 Limited Edition (Rioja)</td>\n", | |
| " <td>Tempranillo</td>\n", | |
| " <td>Ramón Bilbao</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>74324</th>\n", | |
| " <td>France</td>\n", | |
| " <td>A new-wood-dominated wine, this is very toasty...</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>89</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>Bordeaux</td>\n", | |
| " <td>Saint-Julien</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>Roger Voss</td>\n", | |
| " <td>@vossroger</td>\n", | |
| " <td>Château Saint-Pierre 2010 Saint-Julien</td>\n", | |
| " <td>Bordeaux-style Red Blend</td>\n", | |
| " <td>Château Saint-Pierre</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>24764</th>\n", | |
| " <td>France</td>\n", | |
| " <td>Orange peel perfumes the rich pear notes and i...</td>\n", | |
| " <td>Zellberg</td>\n", | |
| " <td>93</td>\n", | |
| " <td>58.0</td>\n", | |
| " <td>Alsace</td>\n", | |
| " <td>Alsace</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>Anne Krebiehl MW</td>\n", | |
| " <td>@AnneInVino</td>\n", | |
| " <td>Domaine Ostertag 2013 Zellberg Pinot Gris (Als...</td>\n", | |
| " <td>Pinot Gris</td>\n", | |
| " <td>Domaine Ostertag</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>78009</th>\n", | |
| " <td>US</td>\n", | |
| " <td>Spicy nutmeg meets a reckoning of lemon zest a...</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>91</td>\n", | |
| " <td>32.0</td>\n", | |
| " <td>California</td>\n", | |
| " <td>Russian River Valley</td>\n", | |
| " <td>Sonoma</td>\n", | |
| " <td>Virginie Boone</td>\n", | |
| " <td>@vboone</td>\n", | |
| " <td>Jordan 2014 Chardonnay (Russian River Valley)</td>\n", | |
| " <td>Chardonnay</td>\n", | |
| " <td>Jordan</td>\n", | |
| " </tr>\n", | |
| " </tbody>\n", | |
| "</table>\n", | |
| "</div>" | |
| ], | |
| "text/plain": [ | |
| " country description \\\n", | |
| "4323 US Grown in the Sebastopol Hills area, this Pinot... \n", | |
| "50398 Spain Coarse on the nose, this has scratchy aromas o... \n", | |
| "74324 France A new-wood-dominated wine, this is very toasty... \n", | |
| "24764 France Orange peel perfumes the rich pear notes and i... \n", | |
| "78009 US Spicy nutmeg meets a reckoning of lemon zest a... \n", | |
| "\n", | |
| " designation points price province region_1 \\\n", | |
| "4323 Umino Vineyard 89 48.0 California Sonoma Coast \n", | |
| "50398 Limited Edition 86 17.0 Northern Spain Rioja \n", | |
| "74324 NaN 89 NaN Bordeaux Saint-Julien \n", | |
| "24764 Zellberg 93 58.0 Alsace Alsace \n", | |
| "78009 NaN 91 32.0 California Russian River Valley \n", | |
| "\n", | |
| " region_2 taster_name taster_twitter_handle \\\n", | |
| "4323 Sonoma NaN NaN \n", | |
| "50398 NaN Michael Schachner @wineschach \n", | |
| "74324 NaN Roger Voss @vossroger \n", | |
| "24764 NaN Anne Krebiehl MW @AnneInVino \n", | |
| "78009 Sonoma Virginie Boone @vboone \n", | |
| "\n", | |
| " title \\\n", | |
| "4323 W.H. Smith 2007 Umino Vineyard Pinot Noir (Son... \n", | |
| "50398 Ramón Bilbao 2009 Limited Edition (Rioja) \n", | |
| "74324 Château Saint-Pierre 2010 Saint-Julien \n", | |
| "24764 Domaine Ostertag 2013 Zellberg Pinot Gris (Als... \n", | |
| "78009 Jordan 2014 Chardonnay (Russian River Valley) \n", | |
| "\n", | |
| " variety winery \n", | |
| "4323 Pinot Noir W.H. Smith \n", | |
| "50398 Tempranillo Ramón Bilbao \n", | |
| "74324 Bordeaux-style Red Blend Château Saint-Pierre \n", | |
| "24764 Pinot Gris Domaine Ostertag \n", | |
| "78009 Chardonnay Jordan " | |
| ] | |
| }, | |
| "execution_count": 3, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "with ZipFile(\"wine-reviews.zip\") as zipf:\n", | |
| " source_file = zipf.open(\"winemag-data-130k-v2.csv\")\n", | |
| " wine_df = pd.read_csv(source_file, index_col=0)\n", | |
| "\n", | |
| "wine_df.sample(n=5)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 4, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:00.691029Z", | |
| "start_time": "2022-08-04T11:10:00.686692Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "(129971, 13)" | |
| ] | |
| }, | |
| "execution_count": 4, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "wine_df.shape" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "# Designed system\n", | |
| "\n", | |
| "\n", | |
| "\n", | |
| "\n", | |
| "The variety of wine species and producers is so rich that nobody can drink them all. Indeed, some try, but for a majority of us, that's not feasible. However, if we would like to maximize the probability we'll enjoy the wine we just bought, similarity learning is a great choice to help.\n", | |
| "\n", | |
| "If we already have some experience with wine and know the ones we liked, then we could make an attempt to create a system that will find some different ones that we should also like. Our dataset has a column with a detailed description of a particular wine, and that's quite easy to convert those descriptions into embeddings and perform a semantic search based on them. However, that won't include our own preferences. We may like both Chardonnay and Shiraz, but the taste and smell perception is totally different.\n", | |
| "\n", | |
| "" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "## Embeddings\n", | |
| "\n", | |
| "First, let's create embeddings of the wine descriptions using a pretrained model from SentenceTransformers. We're going to use `all-MiniLM-L6-v2`, but it actually doesn't matter that much, as we're still going to fine tune the created model." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 5, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:02.641732Z", | |
| "start_time": "2022-08-04T11:10:00.692730Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "from sentence_transformers import SentenceTransformer" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 6, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:03.164184Z", | |
| "start_time": "2022-08-04T11:10:02.644696Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "model = SentenceTransformer(\"all-MiniLM-L6-v2\")" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 7, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:03.178605Z", | |
| "start_time": "2022-08-04T11:10:03.168902Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "['The enticing tartness of red and green apples dominates the appetizing nose. These crisp notes are heightened by a very lively frothing mousse on a light dry palate that shows both autolysis and freshness. The finish is lip-smacking and very refreshing and has a lasting hint of yeast.',\n", | |
| " 'This five-grape blend straddles the gamut of aromas and flavors. It starts with a flat, aged cheese and tobacco nose before opening to show cassis, raspberry and other solid fruit flavors. The finish is sturdy and lengthy, and overall it delivers a round feel and good expression.',\n", | |
| " 'Earthy, tart cherry and strawberry paint a pretty picture in this aromatic, inviting wine, a bold, fruity offering accented by luscious streaks of herb. Well-integrated and composed, dustings of cocoa and cola spice build steam on the finish, the overall package just lightly oaked.',\n", | |
| " 'Dark volcanic influences are evident here and shape this impenetrable and intense Aglianico. Aromas include blackberry, chocolate fudge, spice, smoke and rhubarb. The wine is elegant and tight with dusty tannins and sour cherries on the close. Drink after 2010.',\n", | |
| " 'A powerfully concentrated wine, its ripe yellow fruits smoothly filling the mouth with their unctuous honeyed character. Ginger and other spices add an edge to this gorgeous, full wine that needs aging over many years.']" | |
| ] | |
| }, | |
| "execution_count": 7, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "sample_descriptions = wine_df[\"description\"].sample(n=5).tolist()\n", | |
| "sample_descriptions" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 8, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:05.911313Z", | |
| "start_time": "2022-08-04T11:10:03.180907Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "array([[ 0.07002985, -0.07527184, 0.06776675, ..., -0.00525439,\n", | |
| " 0.01998687, -0.01869105],\n", | |
| " [-0.04183391, -0.06389872, -0.02885838, ..., -0.0331034 ,\n", | |
| " 0.06085657, -0.00752696],\n", | |
| " [-0.00212462, -0.00276716, 0.07620955, ..., -0.04412899,\n", | |
| " 0.05942521, 0.00020054],\n", | |
| " [ 0.04090948, -0.01545571, 0.0628666 , ..., -0.02850977,\n", | |
| " -0.0013344 , -0.05113691],\n", | |
| " [-0.0294616 , 0.00158572, -0.02667119, ..., 0.01966581,\n", | |
| " 0.00921486, -0.00683407]], dtype=float32)" | |
| ] | |
| }, | |
| "execution_count": 8, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "sample_embeddings = model.encode(sample_descriptions)\n", | |
| "sample_embeddings" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "# Pipeline implementation\n", | |
| "\n", | |
| "Right now we've just seen how to create the embeddings using pretrained model, but that has nothing to do with the similarity learning. All its magic starts here. The text model we just used should be already able to capture the characteristics of each wine, but hasn't included our preferences yet. We'll use Quaterion in order to perform the fine-tuning.\n", | |
| "\n", | |
| "\n", | |
| "\n", | |
| "At first, we need to structure the encoder a little bit, so it can be used for further training. `Encoder` in Quaterion is a part of the network responsible for generating the embeddings, and we need to implement it so it perform full conversion of given text into the embeddings space.\n", | |
| "\n", | |
| "\n" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 9, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:05.926206Z", | |
| "start_time": "2022-08-04T11:10:05.914499Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "import os\n", | |
| "from torch import Tensor, nn\n", | |
| "from sentence_transformers.models import Transformer, Pooling\n", | |
| "from quaterion_models.types import TensorInterchange, CollateFnType\n", | |
| "from quaterion_models.encoders import Encoder" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 10, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.000361Z", | |
| "start_time": "2022-08-04T11:10:05.928786Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "class WineDescriptionEncoder(Encoder):\n", | |
| " def __init__(self, transformer, pooling):\n", | |
| " super().__init__()\n", | |
| " self.transformer = transformer\n", | |
| " self.pooling = pooling\n", | |
| " self.encoder = nn.Sequential(self.transformer, self.pooling)\n", | |
| "\n", | |
| " @property\n", | |
| " def trainable(self) -> bool:\n", | |
| " # Defines if we want to train encoder itself, or head layer only\n", | |
| " return False\n", | |
| "\n", | |
| " @property\n", | |
| " def embedding_size(self) -> int:\n", | |
| " return self.transformer.get_word_embedding_dimension()\n", | |
| "\n", | |
| " def forward(self, batch: TensorInterchange) -> Tensor:\n", | |
| " return self.encoder(batch)[\"sentence_embedding\"]\n", | |
| "\n", | |
| " def get_collate_fn(self) -> CollateFnType:\n", | |
| " # `collate_fn` is a function that converts input samples into Tensor(s) for use as encoder input.\n", | |
| " return self.transformer.tokenize\n", | |
| "\n", | |
| " @staticmethod\n", | |
| " def _transformer_path(path: str) -> str:\n", | |
| " # just an additional method to reduce amount of repeated code\n", | |
| " return os.path.join(path, \"transformer\")\n", | |
| "\n", | |
| " @staticmethod\n", | |
| " def _pooling_path(path: str) -> str:\n", | |
| " return os.path.join(path, \"pooling\")\n", | |
| "\n", | |
| " def save(self, output_path: str):\n", | |
| " # to provide correct saving of encoder layers we need to implement it manually\n", | |
| " transformer_path = self._transformer_path(output_path)\n", | |
| " os.makedirs(transformer_path, exist_ok=True)\n", | |
| "\n", | |
| " pooling_path = self._pooling_path(output_path)\n", | |
| " os.makedirs(pooling_path, exist_ok=True)\n", | |
| "\n", | |
| " self.transformer.save(transformer_path)\n", | |
| " self.pooling.save(pooling_path)\n", | |
| "\n", | |
| " @classmethod\n", | |
| " def load(cls, input_path: str) -> Encoder:\n", | |
| " transformer = Transformer.load(cls._transformer_path(input_path))\n", | |
| " pooling = Pooling.load(cls._pooling_path(input_path))\n", | |
| " return cls(transformer=transformer, pooling=pooling)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "Another important concept is a `TrainableModel` that handles the training phase. It is an extended version of the `LightningModule` from PyTorch Lightning. We're going to attach an additional layer, so-called head, and only train this new one, but the encoder won't be updated at all." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 11, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.691010Z", | |
| "start_time": "2022-08-04T11:10:06.001982Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "from quaterion.eval.attached_metric import AttachedMetric\n", | |
| "from torch.optim import Adam\n", | |
| "from quaterion import TrainableModel\n", | |
| "from quaterion.train.cache import CacheConfig, CacheType\n", | |
| "from quaterion.loss import MultipleNegativesRankingLoss, TripletLoss\n", | |
| "from sentence_transformers import SentenceTransformer\n", | |
| "from quaterion_models.heads.skip_connection_head import SkipConnectionHead\n", | |
| "from quaterion.eval.group import RetrievalRPrecision" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 12, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.701809Z", | |
| "start_time": "2022-08-04T11:10:06.693640Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "class WineDescriptionModel(TrainableModel):\n", | |
| " def __init__(self, lr=10e-5, *args, **kwargs):\n", | |
| " self.lr = lr\n", | |
| " super().__init__(*args, **kwargs)\n", | |
| "\n", | |
| " def configure_metrics(self):\n", | |
| " return [\n", | |
| " AttachedMetric(\n", | |
| " \"RetrievalRPrecision\",\n", | |
| " RetrievalRPrecision(),\n", | |
| " prog_bar=True,\n", | |
| " on_epoch=True\n", | |
| " ),\n", | |
| " ]\n", | |
| "\n", | |
| " def configure_optimizers(self):\n", | |
| " return Adam(self.model.parameters(), lr=self.lr)\n", | |
| "\n", | |
| " def configure_loss(self):\n", | |
| " return TripletLoss()\n", | |
| "\n", | |
| " def configure_encoders(self):\n", | |
| " pre_trained_model = SentenceTransformer(\"all-MiniLM-L6-v2\")\n", | |
| " transformer: Transformer = pre_trained_model[0]\n", | |
| " pooling: Pooling = pre_trained_model[1]\n", | |
| " encoder = WineDescriptionEncoder(transformer, pooling)\n", | |
| " return encoder\n", | |
| "\n", | |
| " def configure_head(self, input_embedding_size: int):\n", | |
| " return SkipConnectionHead(input_embedding_size)\n", | |
| "\n", | |
| " def configure_caches(self):\n", | |
| " return CacheConfig(CacheType.AUTO, batch_size=256)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "The last, yet still important thing is a dataset. And surprisingly, we are not going to label all the data we have. Instead, we're just going to create two groups - liked and hated. The names are quite obvious - the first one contains those wines we had and enjoyed, and the other one these we want to avoid. The dataset should be enclosed in a PyTorch `Dataset` subclass." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 13, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.781231Z", | |
| "start_time": "2022-08-04T11:10:06.704219Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "from torch.utils.data import Dataset\n", | |
| "from quaterion.dataset import (\n", | |
| " GroupSimilarityDataLoader,\n", | |
| " SimilarityGroupSample,\n", | |
| ")" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 14, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.804869Z", | |
| "start_time": "2022-08-04T11:10:06.782520Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "liked_mask = wine_df[\"variety\"].isin([\"Gewürztraminer\", \"Riesling\", \"Sauvignon Blanc\"])\n", | |
| "liked = wine_df[liked_mask][\"description\"].sample(50).tolist()\n", | |
| "\n", | |
| "hated_mask = wine_df[\"variety\"].isin([\"Cabernet Sauvignon\", \"Merlot\"])\n", | |
| "hated = wine_df[hated_mask][\"description\"].sample(50).tolist()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 15, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.811191Z", | |
| "start_time": "2022-08-04T11:10:06.806530Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "['Inviting notions of damask rose signal richness and aromatic opulence and continue on the rounded, plush and almost oily palate. It provides a textbook example of rich Alsace Gewurztraminer that shows rose-petal richness tamed by fresh orange-peel zestiness.',\n", | |
| " \"This lovely wine captures the floral, perfumed essence of the grape in a restrained, sippable style. It's dry and crisp, with lemony fruit and the delicate sensation that you are drinking fresh flowers.\",\n", | |
| " 'A soft style of Riesling, with some attractive aromatic white flower character. The fruit is soft, just off dry, freshened by acidity. Screwcap.',\n", | |
| " 'Still young, this has the potential to be a rich, rounded wine. Already the citrus-drenched yellow fruits are asserting themselves along with a warm, ripe texture. Give this wine until 2016.',\n", | |
| " 'A bright, herbaceous wine that is full of citrus and grapefruit flavors. These go with a tangy character, crisp and hinting at black currants and more tropical fruit.',\n", | |
| " \"Hay, mint and grass add a green touch to the nose, which also features nectarine. Citrus, but nothing really defined, dominates the flavor profile. More citrus and zippy acids run wild on the finish. It doesn't offer any one thing for you to hang your hat on.\",\n", | |
| " 'This wine is soft, although with plenty of ripe fruit. It has a touch of spice and light acidity, showing the richness of the white-wine vintage. Drink now.',\n", | |
| " 'The flint and sand soil of this vineyard brings out the minerally, steely character of this taut, tight wine. While it is herbaceous and flavors of pineapple and white fruit, the structure dominates. Give this impressive wine another year or two.',\n", | |
| " \"Aromas of toast, apricot, mint, honey and field dust make for a complex bouquet. The honeyed palate yields a trace of green herbs along with lemon rind and white pepper. Obviously there's a lot going on here, and for some it may be too “out there.” For others it will be a spot-on Kiwi offering with lots of individuality and character.\",\n", | |
| " \"Secondary characteristics like mushrooms, hay, and nuts make this unusual but in a mostly charming way. Tropical fruits take a back seat on the palate; Green beans and peas are in the driver's seat. Texturally it has a surprising weight, but there's enough acidity to keep things fresh.\",\n", | |
| " \"A respectable effort, this offers attractive aromas of fig, melon and nectarine, but doesn't follow through with much intensity on the palate. It's light to medium in body, with hints of citrus pith and a bit of bitterness on the finish.\",\n", | |
| " \"Mandarin-orange zest, cut Asian pear, lemon blossom and slight beeswax show on the fresh nose of this bottling by young winemaker Mike Callahan. It's extremely dry on the palate, like licking limestone, with hints of nectarine blossom and lemon-skin flavor.\",\n", | |
| " \"With the snap and concentration of a tasty lemon-drop cocktail, this lovely Gewürztraminer pays homage to the floral aspect of the grape without drowning in it. Perfumed, not soapy, with superb fruit flavors of peaches and citrus, it's a sensational value.\",\n", | |
| " \"This old-vine Riesling matches medium sweetness (25g/L) to ample acidity. It's a dense and flavorful style, taut and fleshy at the same time. The peachy fruit hints at ripe banana as well, with a lick of butterscotch adorning the finish. Drink now through 2022.\",\n", | |
| " 'Tart in lemon peel and tangerine, this lively white also offers a magnitude of stony minerality that manifests as a gravelly complexion on the texture. Honeysuckle figures prominently on the nose.',\n", | |
| " \"A blend of Sauvignon Blanc, Sauvignon Gris and Sauvignon Musqué, this wine delights for its supple apricot and nectarine flavors. It's floral on the nose, with grapefruit and fennel scents, unfolding slowly to reveal a rich texture and fresh acidity.\",\n", | |
| " \"A rounded earthiness reminiscent of Bosc pear envelops this fluid, balanced and fruit-driven wine. It conveys effortless harmony with a creamy touch that's clear and beautiful. The light approach is admirable, the clean, apple-scented finish very satisfying.\",\n", | |
| " 'Beautifully lifted, almost sublimated apple and citrus fruit combine to make a perfumed, appetizing nose. That pure apple theme continues on the concentrated palate that sings with purity to leave a satisfying and honest impression of completeness. Pure and simply very delicious.',\n", | |
| " 'White pepper and celery aromas, along with distinctly herbaceous and tart flavors give a tight aspect. It is definitely light, crisp and mouth-cleansing; would be great with raw oysters or goat cheese on toasts.',\n", | |
| " 'Heady peach is joined by a tart notion of citrus foliage. This exquisite aromatic quality of the nose carries throughout the taut but generous body. Despite dryness, luscious peach and red-cheeked mirabelle plums are evoked in a very pure honest fashion. This is the bounty of 2015 displayed with a very light touch, with finesse and elegance. Drink now through 2027.',\n", | |
| " \"A fine, crisp and clean Sauvignon Blanc at a good price. It's basically dry, with a honeyed edge to the tropical fruit, lime and peach flavors, and shows a good bite of acidity.\",\n", | |
| " \"Luscious white peach and apricot flavors are anchored by a steely core of mineral in this easy-drinking but elegant wine. Light bodied but persistent, it's a lip-smacking, sweet wine that finishes with a touch of brambly herbaceousness. Drink now through 2020.\",\n", | |
| " \"Here's a Sauvignon Blanc (with 15% Chardonnay) from a warm Italian climate (Cortona, Tuscany) that offers broad and thick aromas of passionfruit, peach and a touch of mango. The wine is light and streamlined in the mouth and makes a good partner to light finger foods or appetizers.\",\n", | |
| " 'Now at its peak, this rich, rounded wine is full of spice and tropical fruits, and is full in the mouth. The explosion of ripe fruits is only lightly restrained by some final acidity. This joyous wine is ready to drink.',\n", | |
| " 'This has intriguing aromas of marzipan and white cherry, plus touches of apple and mineral. A medium-bodied wine, it offers acidity that becomes mouth-puckering on the finish. Pair this with some goat cheese to help tame the tartness.',\n", | |
| " \"Medium-gold in color. Complex and inviting nose layered with a strong earthy minerality, vibrant bright fruit and intriguing notes of petrol and honey. On the palate it is racy, taut and focused full of well defined ripe succulent citrus and stone fruit. Rich, viscous texture perfectly balanced by the wine's crispness and almost steely minerality. Very long, juicy, lingering finish.\",\n", | |
| " 'Aromas of tropical fruit and lime are restrained on the nose. The palate is soft and a bit flabby, with papaya and yellow-apple flavors that become more pithy and bitter on the easy finish.',\n", | |
| " \"Breathtaking scents of white peach and blossom introduce this pristinely peachy wine. It's invigoratingly fresh, laced by a gossamer fringe of sweetness and vivacious lemon-lime acidity that lingers on and on. Delicately textured yet deeply penetrating.\",\n", | |
| " 'Sappy and deep, this brings a mouth-pleasing mix of citrus and stone fruits, broadening into a full midpalate. Full-bodied, it puts the emphasis on fruit, with just a trace of minerality.',\n", | |
| " 'This wine is fresh, crisp and still young. Crisp acidity dominates the strongly citrus character and gives a tight aftertaste of lime and pink grapefruit. This 100% Sauvignon Blanc was aged on the lees for six months to enhance aroma. Wait to drink until early 2016.',\n", | |
| " \"Cool, mineral elegance and pristine tangerine and quince flavors mark this zesty, sylphlike Riesling. It's dry and light bodied with a finish highlighted by lemony acidity. Drink now through 2019.\",\n", | |
| " \"At the risk of overgeneralizing, Martinborough Sauvignons typically show less overt herbal characters than their Marlborough counterparts. At least, that's the case for this medium-bodied wine. White grapefruit, stone fruit and underripe melon notes finish crisp and clean.\",\n", | |
| " 'The mellow scent of baked apple rises from the glass, cut by ripe, equally mellow freshness on the palate. This wine is mild for a Riesling but very enjoyable, clean and fruity, with a mellow apple note that really gets you in the end.',\n", | |
| " 'Gentle purity and fluidity characterize the palate. There is a vein of lemony, citric freshness but also a creamy note of yeast. This is sleek but comes with rounded edges and gentle citrus and yellow plum notes.',\n", | |
| " 'A single estate run by Joseph Mellot, this produces an impressive ripe style of Sauvignon Blanc, at this stage youthful with exuberant white fruits and grapefruit, and an obvious need to age for 1–2 years.',\n", | |
| " 'This is pure fruitiness, softly textured and ready to drink. It has an attractive line of grapefruit and orange fruit along with a warm ripe aftertaste. The wine is delicious now.',\n", | |
| " 'Imparting a hint of petrol and wax on the nose, this Riesling is thickly textured with flavors of ripe apple, peach and pear. With nice acidity, it finishes dry.',\n", | |
| " 'Jasmine, apricot and floral aromas are followed by medium-sweet stone-fruit flavors. Floral and jasmine flavors persist on the finish.',\n", | |
| " 'Made in the charmat method, this has a light mousse and subtle notes of tropical vanilla, lemon, lime and Golden Delicious apple. Semi-dry, the richness is offset by healthy acidity and a long finish.',\n", | |
| " \"There's a slight spritziness and ripe fruit sweetness at the core of this appealing wine. The mix of mango and papaya on the palate will show well with spicier foods or at casual get-togethers. Unfussy and well-priced, it's layered and creamy with a crisp finish.\",\n", | |
| " \"A straightforward example of Chilean SB, meaning there's citrus and mineral on the nose, juicy acidity and lime, grapefruit and tangerine flavors. It's clean and solid, with slight pithiness. Delivers a lot for $9; drink now.\",\n", | |
| " \"Crisp green aromas suggest lime and then passion fruit. It's lively and spritzy in the mouth, with high acidity yielding green flavors that transition to pithy and heavy. Lemon-lime mixed with celery notes come up on the finish.\",\n", | |
| " 'A ripe currant- and lime-flavored wine, this is rounded and soft. This full-bodied wine fills the mouth but maintains its delicate crispness. There are good bright highlights of green fruits and fresh acidity.',\n", | |
| " 'A fresh, yeasty nose suggests a young wine. On the palate the white currants and fresh fruit are youthful and juicy, the mineral character only just showing through at the end. The wine needs at least 3–4 years to develop.',\n", | |
| " \"Pungent and leafy at first, this also boasts honey and passion fruit aromas. It's medium in body, with tropical, pineapple flavors and a clean, refreshing finish. Drink up.\",\n", | |
| " 'From the heart of the Napa Valley, this estate white is dry in floral lemon and lime, in addition to more voluptuous helpings of apricot and peach. Stainless-steel fermented, it offers crisp acidity before taking on complexity on the finish, more weighty than on the palate.',\n", | |
| " \"Blossoms and pink grapefruits scent this fresh, fruity wine. Off dry in style, it's concentrated in plush, tropical flavors of guava and mango yet calibrated neatly by a vein of lime acidity. It's guzzably fresh and approachable but refined enough for elegant dining occasions. Drink now through 2019.\",\n", | |
| " 'This mature expression of Sauvignon opens with bright aromas of peach, honey, apricot and passion fruit. The mouthfeel is less intense with a lean, but crisp finish.',\n", | |
| " 'Apple and pear notes have a slightly savory tone in this brisk, lemony dry Riesling. Throughout a tasting of two samples, there are lingering tones of rubber and struck matches that blow off gradually with aeration.',\n", | |
| " 'Airy smelling, with citrus and grass aromas brewing underneath. Has a wet, simple, refreshing feel and rather intense flavors of lime, tangerine and other citrus fruits. Cuts a direct path.']" | |
| ] | |
| }, | |
| "execution_count": 15, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| } | |
| ], | |
| "source": [ | |
| "liked" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 16, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.815545Z", | |
| "start_time": "2022-08-04T11:10:06.812811Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "import itertools" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 17, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.821746Z", | |
| "start_time": "2022-08-04T11:10:06.817295Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "class WineDataset(Dataset):\n", | |
| "\n", | |
| " def __init__(self, liked, hated):\n", | |
| " self.examples = list((entry, \"liked\") for entry in liked) + \\\n", | |
| " list((entry, \"hated\") for entry in hated)\n", | |
| "\n", | |
| " def __getitem__(self, index) -> SimilarityGroupSample:\n", | |
| " obj, group = self.examples[index]\n", | |
| " return SimilarityGroupSample(obj=obj, group=hash(group))\n", | |
| "\n", | |
| " def __len__(self):\n", | |
| " return len(self.examples)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "# Training\n", | |
| "\n", | |
| "We're ready to train the network, so it creates new embeddings, this time with our preferences included." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 18, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:10:06.826185Z", | |
| "start_time": "2022-08-04T11:10:06.823648Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "import torch\n", | |
| "import pytorch_lightning as pl\n", | |
| "from quaterion import Quaterion" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 19, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T11:15:58.494110Z", | |
| "start_time": "2022-08-04T11:10:06.827876Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "Global seed set to 42\n", | |
| "Auto select gpus: [0]\n", | |
| "GPU available: True, used: True\n", | |
| "TPU available: False, using: 0 TPU cores\n", | |
| "IPU available: False, using: 0 IPUs\n", | |
| "HPU available: False, using: 0 HPUs\n", | |
| "2022-08-04 13:10:07.250 | DEBUG | quaterion.train.cache_mixin:_cache:168 - Using full cache\n", | |
| "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n" | |
| ] | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "146323728a3b4d71a103109959f258a2", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Predicting: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "2022-08-04 13:10:07.669 | DEBUG | quaterion.train.cache_mixin:_cache:220 - Caching has been successfully finished\n", | |
| "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", | |
| "\n", | |
| " | Name | Type | Params\n", | |
| "-------------------------------------------\n", | |
| "0 | _model | SimilarityModel | 22.9 M\n", | |
| "1 | _loss | TripletLoss | 0 \n", | |
| "-------------------------------------------\n", | |
| "148 K Trainable params\n", | |
| "22.7 M Non-trainable params\n", | |
| "22.9 M Total params\n", | |
| "91.446 Total estimated model params size (MB)\n" | |
| ] | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Sanity Checking: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "084a356c2a584e3db3b5bdce87600daf", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Training: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "application/vnd.jupyter.widget-view+json": { | |
| "model_id": "", | |
| "version_major": 2, | |
| "version_minor": 0 | |
| }, | |
| "text/plain": [ | |
| "Validation: 0it [00:00, ?it/s]" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| } | |
| ], | |
| "source": [ | |
| "def run(model, params):\n", | |
| " use_gpu = params.get(\"cuda\", torch.cuda.is_available())\n", | |
| "\n", | |
| " trainer = pl.Trainer(\n", | |
| " min_epochs=params.get(\"min_epochs\", 1),\n", | |
| " max_epochs=params.get(\"max_epochs\", 300),\n", | |
| " auto_select_gpus=use_gpu,\n", | |
| " log_every_n_steps=params.get(\"log_every_n_steps\", 1),\n", | |
| " gpus=int(use_gpu),\n", | |
| " num_sanity_val_steps=2,\n", | |
| " )\n", | |
| " liked_split_index, hated_split_index = int(0.8 * len(liked)), int(0.8 * len(hated))\n", | |
| " train_dataset = WineDataset(liked[:liked_split_index], hated[:hated_split_index])\n", | |
| " val_dataset = WineDataset(liked[liked_split_index:], hated[hated_split_index:])\n", | |
| " train_dataloader = GroupSimilarityDataLoader(train_dataset, batch_size=1024)\n", | |
| " val_dataloader = GroupSimilarityDataLoader(val_dataset, batch_size=1024)\n", | |
| " Quaterion.fit(model, trainer, train_dataloader, val_dataloader)\n", | |
| "\n", | |
| "\n", | |
| "pl.seed_everything(42, workers=True)\n", | |
| "wine_model = WineDescriptionModel()\n", | |
| "run(wine_model, {})" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:02:24.164441Z", | |
| "start_time": "2022-08-04T12:02:24.154394Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "# Neural search\n", | |
| "\n", | |
| "We've created a model that not only uses the descriptions, but also incorporates the fact if we liked that specific wine or not. We can now compare the results with the original sentence transformer." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 26, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:07:36.617616Z", | |
| "start_time": "2022-08-04T12:07:36.614985Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "from sentence_transformers import util" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 31, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:17:16.572835Z", | |
| "start_time": "2022-08-04T12:17:16.556800Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "wine_df_sample = wine_df[\"description\"].sample(frac=0.05).tolist()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 35, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:20:32.782794Z", | |
| "start_time": "2022-08-04T12:19:09.436354Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "corpus_original_embeddings = model.encode(wine_df_sample, convert_to_tensor=False)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 45, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:25:14.314067Z", | |
| "start_time": "2022-08-04T12:22:35.008516Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "corpus_finetuned_embeddings = wine_model._model.encode(wine_df_sample)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 80, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:58:57.343097Z", | |
| "start_time": "2022-08-04T12:58:57.338772Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "query_text = \"Floral notes with a touch of pomegranate. Perfect for summertime evening meal.\"" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 81, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:58:57.872766Z", | |
| "start_time": "2022-08-04T12:58:57.849076Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "query_embedding = model.encode(query_text)\n", | |
| "cos_scores = util.cos_sim(query_embedding, corpus_original_embeddings)[0]\n", | |
| "top_results = torch.topk(cos_scores, k=5)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 88, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:59:56.907880Z", | |
| "start_time": "2022-08-04T12:59:56.904632Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "1 This is a classically light rosé, fragrant in strawberry and raspberry. Traces of pomegranate on the back palate offers darker, riper fruit in the glass, getting prettier as it goes along. Deliciously dry and dancing in acidity, it has plenty of citrus on the finish. The wine is substantial enough for springtime and summertime meals, but will drink equally well on its own. (Score: 0.6777) \n", | |
| "\n", | |
| "2 This is a complex wine, jammy and funky in pomegranate and orange notes that are acid-driven but not lacking in fruit or spice components. Soft and silky on the palate, it conveys a twinge of sandalwood to round things out on the long finish. (Score: 0.6526) \n", | |
| "\n", | |
| "3 Aromas of spiced plum, pipe tobacco, French oak and a whiff of pressed purple flower slowly take shape on this firmly structured red. The palate is youthfully austere, offering dried black cherry, pomegranate, green tea and clove framed in fresh acidity and bracing fine-grained tannins that give the finish a tight grip. An espresso note wraps up the close. (Score: 0.5737) \n", | |
| "\n", | |
| "4 Il Poggiale Riserva opens with bold tones of blackberry pie, plum, currant, leather, rum cake and dried rosemary. There's a ripe, almost jammy, quality to the fruit, and the palate feels dense and richly textured. (Score: 0.5736) \n", | |
| "\n", | |
| "5 Fresh apple and white floral notes meld effortlessly into an elegant vanilla note on the nose and palate of this dry, full-bodied Chard. Nicely concentrated, yet brisk with acidity, it's an excellent food pairing companion. (Score: 0.5673) \n", | |
| "\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "for i, (score, idx) in enumerate(zip(top_results[0], top_results[1])):\n", | |
| " print(i + 1, wine_df_sample[idx], \"(Score: {:.4f})\".format(score), \"\\n\")" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%% md\n" | |
| } | |
| }, | |
| "source": [ | |
| "## Fine-tuned network" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 89, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T12:59:59.979456Z", | |
| "start_time": "2022-08-04T12:59:59.942578Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "finetuned_query_embedding = wine_model._model.encode(query_text)\n", | |
| "ft_cos_scores = util.cos_sim(finetuned_query_embedding, corpus_finetuned_embeddings)[0]\n", | |
| "ft_top_results = torch.topk(ft_cos_scores, k=5)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 90, | |
| "metadata": { | |
| "ExecuteTime": { | |
| "end_time": "2022-08-04T13:00:00.537270Z", | |
| "start_time": "2022-08-04T13:00:00.533241Z" | |
| }, | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "1 This is a classically light rosé, fragrant in strawberry and raspberry. Traces of pomegranate on the back palate offers darker, riper fruit in the glass, getting prettier as it goes along. Deliciously dry and dancing in acidity, it has plenty of citrus on the finish. The wine is substantial enough for springtime and summertime meals, but will drink equally well on its own. (Score: 0.6673) \n", | |
| "\n", | |
| "2 This is a complex wine, jammy and funky in pomegranate and orange notes that are acid-driven but not lacking in fruit or spice components. Soft and silky on the palate, it conveys a twinge of sandalwood to round things out on the long finish. (Score: 0.6420) \n", | |
| "\n", | |
| "3 Il Poggiale Riserva opens with bold tones of blackberry pie, plum, currant, leather, rum cake and dried rosemary. There's a ripe, almost jammy, quality to the fruit, and the palate feels dense and richly textured. (Score: 0.5578) \n", | |
| "\n", | |
| "4 Aromas of spiced plum, pipe tobacco, French oak and a whiff of pressed purple flower slowly take shape on this firmly structured red. The palate is youthfully austere, offering dried black cherry, pomegranate, green tea and clove framed in fresh acidity and bracing fine-grained tannins that give the finish a tight grip. An espresso note wraps up the close. (Score: 0.5553) \n", | |
| "\n", | |
| "5 Fresh apple and white floral notes meld effortlessly into an elegant vanilla note on the nose and palate of this dry, full-bodied Chard. Nicely concentrated, yet brisk with acidity, it's an excellent food pairing companion. (Score: 0.5496) \n", | |
| "\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "for i, (score, idx) in enumerate(zip(ft_top_results[0], ft_top_results[1])):\n", | |
| " print(i + 1, wine_df_sample[idx], \"(Score: {:.4f})\".format(score), \"\\n\")" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "pycharm": { | |
| "name": "#%%\n" | |
| } | |
| }, | |
| "outputs": [], | |
| "source": [] | |
| } | |
| ], | |
| "metadata": { | |
| "kernelspec": { | |
| "display_name": "Python 3", | |
| "language": "python", | |
| "name": "python3" | |
| }, | |
| "language_info": { | |
| "codemirror_mode": { | |
| "name": "ipython", | |
| "version": 3 | |
| }, | |
| "file_extension": ".py", | |
| "mimetype": "text/x-python", | |
| "name": "python", | |
| "nbconvert_exporter": "python", | |
| "pygments_lexer": "ipython3", | |
| "version": "3.9.13" | |
| } | |
| }, | |
| "nbformat": 4, | |
| "nbformat_minor": 1 | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment