Transformer based E-Commerce-Recommendation(Foundational Model).ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":8257098,"sourceType":"datasetVersion","datasetId":4900332},{"sourceId":8274967,"sourceType":"datasetVersion","datasetId":4913624},{"sourceId":8275876,"sourceType":"datasetVersion","datasetId":4914241},{"sourceId":8275884,"sourceType":"datasetVersion","datasetId":4914245},{"sourceId":8275890,"sourceType":"datasetVersion","datasetId":4914251},{"sourceId":8275891,"sourceType":"datasetVersion","datasetId":4914252},{"sourceId":8276687,"sourceType":"datasetVersion","datasetId":4914777},{"sourceId":8276700,"sourceType":"datasetVersion","datasetId":4914789},{"sourceId":8276706,"sourceType":"datasetVersion","datasetId":4914794},{"sourceId":8276709,"sourceType":"datasetVersion","datasetId":4914796},{"sourceId":8299703,"sourceType":"datasetVersion","datasetId":4930626},{"sourceId":8299708,"sourceType":"datasetVersion","datasetId":4930631},{"sourceId":8299720,"sourceType":"datasetVersion","datasetId":4930640},{"sourceId":8299727,"sourceType":"datasetVersion","datasetId":4930645}],"dockerImageVersionId":30699,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-05-03T15:04:19.070050Z","iopub.execute_input":"2024-05-03T15:04:19.070705Z","iopub.status.idle":"2024-05-03T15:04:20.238272Z","shell.execute_reply.started":"2024-05-03T15:04:19.070677Z","shell.execute_reply":"2024-05-03T15:04:20.237279Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stdout","text":"/kaggle/input/multilabel-3000-10epochs/multi-label-binarizer.pkl\n/kaggle/input/multilabelbinarizer-3000-15epochs-cross/multi-label-binarizer.pkl\n/kaggle/input/encodedcsv-3000-15epochs-cross/encodedCSV.csv\n/kaggle/input/datasetcsv54k-12epochs-cross/encodedCSV.csv\n/kaggle/input/cleanedproductdata81k8batch/newData81k.csv\n/kaggle/input/model-3000-15epochs-cross/config.json\n/kaggle/input/model-3000-15epochs-cross/training_args.bin\n/kaggle/input/model-3000-15epochs-cross/model.safetensors\n/kaggle/input/model-3000-10epochs/config.json\n/kaggle/input/model-3000-10epochs/training_args.bin\n/kaggle/input/model-3000-10epochs/model.safetensors\n/kaggle/input/tokenizer-3000-10epochs/tokenizer_config.json\n/kaggle/input/tokenizer-3000-10epochs/special_tokens_map.json\n/kaggle/input/tokenizer-3000-10epochs/vocab.txt\n/kaggle/input/tokenizer-54k-12epoch-cross/tokenizer_config.json\n/kaggle/input/tokenizer-54k-12epoch-cross/special_tokens_map.json\n/kaggle/input/tokenizer-54k-12epoch-cross/vocab.txt\n/kaggle/input/multilabel-binarizer-54k-12epochs-cross/multi-label-binarizer.pkl\n/kaggle/input/encodedcsv-3000-10epoch/encodedCSV.csv\n/kaggle/input/tokenizer-3000-15epochs-cross/tokenizer_config.json\n/kaggle/input/tokenizer-3000-15epochs-cross/special_tokens_map.json\n/kaggle/input/tokenizer-3000-15epochs-cross/vocab.txt\n/kaggle/input/model-54k-12epochs-cross/config.json\n/kaggle/input/model-54k-12epochs-cross/training_args.bin\n/kaggle/input/model-54k-12epochs-cross/model.safetensors\n/kaggle/input/train-mod-loop-n/train_module.py\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# **Jai Hanuman Dada ki Jai**","metadata":{}},{"cell_type":"code","source":"# df_new=pd.read_csv('/kaggle/input/preprocesseddata1lac/Testtfidftransformer1Lac.csv')\ndf=pd.read_csv('/kaggle/input/cleanedproductdata81k8batch/newData81k.csv')","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:59:28.967080Z","iopub.execute_input":"2024-05-02T20:59:28.967650Z","iopub.status.idle":"2024-05-02T20:59:31.184358Z","shell.execute_reply.started":"2024-05-02T20:59:28.967615Z","shell.execute_reply":"2024-05-02T20:59:31.183426Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Create an empty DataFrame to store the sampled data\nsampled_data = pd.DataFrame()\n\n# Group the DataFrame by 'main_category'\ngrouped = df.groupby('main_category')\n\n# Iterate over each group and sample 2000 rows\nfor group_name, group_data in grouped:\n    sampled_data = pd.concat([sampled_data, group_data.sample(n=2000, random_state=42)])\n    \ndf = sampled_data","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:59:32.294660Z","iopub.execute_input":"2024-05-02T20:59:32.295600Z","iopub.status.idle":"2024-05-02T20:59:32.447244Z","shell.execute_reply.started":"2024-05-02T20:59:32.295561Z","shell.execute_reply":"2024-05-02T20:59:32.446299Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df = df.sample(54000)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:59:34.709532Z","iopub.execute_input":"2024-05-02T20:59:34.710045Z","iopub.status.idle":"2024-05-02T20:59:34.726396Z","shell.execute_reply.started":"2024-05-02T20:59:34.710000Z","shell.execute_reply":"2024-05-02T20:59:34.725433Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import re\nfrom nltk.tokenize import word_tokenize\nfrom nltk.corpus import stopwords\nfrom nltk.stem import WordNetLemmatizer\nfrom nltk.stem import PorterStemmer\n\n#Download NLTK resources\nimport nltk\nimport subprocess\nnltk.download('punkt')\nnltk.download('stopwords')\n\ntry:\n    nltk.data.find('wordnet.zip')\nexcept:\n    nltk.download('wordnet', download_dir='/kaggle/working/')\n    command = \"unzip /kaggle/working/corpora/wordnet.zip -d /kaggle/working/corpora\"\n    subprocess.run(command.split())\n    nltk.data.path.append('/kaggle/working/')\n    \n# Function for text preprocessing\ndef preprocess_text(text):\n    # Remove HTML tags\n    text = re.sub(r'<.*?>', '', text)\n    \n    # Remove special characters and numbers\n    text = re.sub(r'[^a-zA-Z]', ' ', text)\n    \n    # Convert to lowercase\n    text = text.lower()\n    \n    # Tokenization\n    tokens = word_tokenize(text)\n    \n    # Remove stop words\n    stop_words = set(stopwords.words('english'))\n    tokens = [word for word in tokens if word not in stop_words]\n    \n    # Lemmatization\n    lemmatizer = WordNetLemmatizer()\n    tokens = [lemmatizer.lemmatize(word) for word in tokens]\n    \n    # Join tokens back into a string\n    preprocessed_text = ' '.join(tokens)\n    \n    return preprocessed_text","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:59:37.700553Z","iopub.execute_input":"2024-05-02T20:59:37.701554Z","iopub.status.idle":"2024-05-02T20:59:39.741162Z","shell.execute_reply.started":"2024-05-02T20:59:37.701509Z","shell.execute_reply":"2024-05-02T20:59:39.740086Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df['CummProdInfo'] = df['CummProdInfo'].apply(preprocess_text)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:59:41.276818Z","iopub.execute_input":"2024-05-02T20:59:41.278040Z","iopub.status.idle":"2024-05-02T21:01:39.749623Z","shell.execute_reply.started":"2024-05-02T20:59:41.278003Z","shell.execute_reply":"2024-05-02T21:01:39.748275Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.preprocessing import LabelEncoder\nlabel_encoder = LabelEncoder()\ndf['label'] = label_encoder.fit_transform(df['title'])","metadata":{"execution":{"iopub.status.busy":"2024-05-02T21:02:01.662536Z","iopub.execute_input":"2024-05-02T21:02:01.663237Z","iopub.status.idle":"2024-05-02T21:02:01.879464Z","shell.execute_reply.started":"2024-05-02T21:02:01.663203Z","shell.execute_reply":"2024-05-02T21:02:01.878528Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Applying TFIDF to find 50 similar items to each item.","metadata":{}},{"cell_type":"code","source":"# from sklearn.feature_extraction.text import TfidfVectorizer\n# from sklearn.metrics.pairwise import cosine_similarity\n\n# # TF-IDF Vectorization\n# tfidf_vectorizer = TfidfVectorizer()\n# tfidf_matrix = tfidf_vectorizer.fit_transform(df['CummProdInfo'])\n\n# # Calculate cosine similarity between each row and all other rows\n# cosine_similarities = cosine_similarity(tfidf_matrix)\n\n# # Find the indices of the top 50 most similar items for each row\n# top_50_indices = cosine_similarities.argsort(axis=1)[:, :-51:-1]\n\n# # Get the label values from the original DataFrame that match the top 50 most similar items for each row\n# top_50_labels = [[df['label'].iloc[i] for i in indices] for indices in top_50_indices]\n\n# # Add the list of labels as values of each row into a new column called label_list\n# df['label_list'] = top_50_labels\n\n# # Display the DataFrame\n# df\n\n\nfrom sklearn.feature_extraction.text import TfidfVectorizer\nfrom sklearn.metrics.pairwise import cosine_similarity\nimport heapq\nimport numpy as np\n\n# TF-IDF Vectorization\ntfidf_vectorizer = TfidfVectorizer()\ntfidf_matrix = tfidf_vectorizer.fit_transform(df['CummProdInfo'])\n\n# Define a function to calculate cosine similarity for a chunk of data\ndef calculate_cosine_similarity(chunk):\n    return cosine_similarity(chunk, tfidf_matrix)\n\n# Initialize an empty dictionary to store the top 50 most similar items for each row\ntop_50_labels = {}\n\n# Process the data in chunks\nchunk_size = 1000\nnum_chunks = int(np.ceil(tfidf_matrix.shape[0] / chunk_size))\nfor chunk_index in range(num_chunks):\n    start_index = chunk_index * chunk_size\n    end_index = min((chunk_index + 1) * chunk_size, tfidf_matrix.shape[0])\n    chunk = tfidf_matrix[start_index:end_index]\n\n    # Calculate cosine similarity for the chunk\n    cosine_similarities_chunk = calculate_cosine_similarity(chunk)\n    \n    # Find the indices of the top 50 most similar items for each row in the chunk\n    for i in range(len(cosine_similarities_chunk)):\n        top_50 = heapq.nlargest(50, range(len(cosine_similarities_chunk[i])), key=lambda k: cosine_similarities_chunk[i][k])\n        top_50_labels[start_index + i] = [df['label'].iloc[k] for k in top_50]\n\n# Add the list of labels as values of each row into a new column called label_list\ndf['label_list'] = [top_50_labels.get(i, []) for i in range(len(df))]\n\n# Display the DataFrame\ndf\n\n","metadata":{"execution":{"iopub.status.busy":"2024-05-02T21:04:41.530605Z","iopub.execute_input":"2024-05-02T21:04:41.531076Z","iopub.status.idle":"2024-05-02T21:25:11.703628Z","shell.execute_reply.started":"2024-05-02T21:04:41.531039Z","shell.execute_reply":"2024-05-02T21:25:11.701579Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"TFIDF done\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df.info()","metadata":{"execution":{"iopub.status.busy":"2024-04-30T18:52:33.393862Z","iopub.execute_input":"2024-04-30T18:52:33.394226Z","iopub.status.idle":"2024-04-30T18:52:33.411760Z","shell.execute_reply.started":"2024-04-30T18:52:33.394200Z","shell.execute_reply":"2024-04-30T18:52:33.410845Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Get the unique titles in the 'title' column\nunique_title = df['main_category'].unique()\nprint(f\"The unique titles in the 'tile' column are: {unique_title}\")\n\n# Get the count of each tile in the 'title' column\ncategory_counts = df['main_category'].value_counts()\nprint(\"\\nThe count of each title is:\")\nprint(category_counts)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:52:30.294833Z","iopub.execute_input":"2024-05-02T19:52:30.295202Z","iopub.status.idle":"2024-05-02T19:52:30.307002Z","shell.execute_reply.started":"2024-05-02T19:52:30.295171Z","shell.execute_reply":"2024-05-02T19:52:30.305886Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df['CummProdInfo'].str.len().plot.hist(bins=40)","metadata":{"execution":{"iopub.status.busy":"2024-04-29T23:26:46.603466Z","iopub.execute_input":"2024-04-29T23:26:46.603931Z","iopub.status.idle":"2024-04-29T23:26:46.637204Z","shell.execute_reply.started":"2024-04-29T23:26:46.603882Z","shell.execute_reply":"2024-04-29T23:26:46.636002Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.preprocessing import MultiLabelBinarizer\nmultilabel = MultiLabelBinarizer(sparse_output=True)\n\nlabelTemp = df['label_list'].tolist()\n\nlabels = multilabel.fit_transform(labelTemp)\nlabels = labels.astype('float32')\n\nlabels = labels.toarray()\n\ntexts = df['CummProdInfo'].tolist()","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:53:43.159465Z","iopub.execute_input":"2024-05-02T19:53:43.160372Z","iopub.status.idle":"2024-05-02T19:53:43.413144Z","shell.execute_reply.started":"2024-05-02T19:53:43.160336Z","shell.execute_reply":"2024-05-02T19:53:43.412223Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"labels","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:53:46.433340Z","iopub.execute_input":"2024-05-02T19:53:46.433980Z","iopub.status.idle":"2024-05-02T19:53:46.440722Z","shell.execute_reply.started":"2024-05-02T19:53:46.433939Z","shell.execute_reply":"2024-05-02T19:53:46.439465Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# labels\n# np.set_printoptions(threshold=np.inf)","metadata":{"execution":{"iopub.status.busy":"2024-04-30T18:53:10.242716Z","iopub.execute_input":"2024-04-30T18:53:10.243110Z","iopub.status.idle":"2024-04-30T18:53:10.248160Z","shell.execute_reply.started":"2024-04-30T18:53:10.243079Z","shell.execute_reply":"2024-04-30T18:53:10.247059Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os\nimport torch\nfrom transformers import DistilBertTokenizer, AutoTokenizer\nfrom transformers import DistilBertForSequenceClassification, AutoModelForSequenceClassification\nfrom sklearn.model_selection import train_test_split\nfrom torch.utils.data import DataLoader, Dataset\n\nos.environ[\"WANDB_DISABLED\"] = \"true\"","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:53:49.145743Z","iopub.execute_input":"2024-05-02T19:53:49.146653Z","iopub.status.idle":"2024-05-02T19:53:56.698224Z","shell.execute_reply.started":"2024-05-02T19:53:49.146617Z","shell.execute_reply":"2024-05-02T19:53:56.697239Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Split data into train and validation sets\ntrain_inputs, val_inputs, train_labels, val_labels = train_test_split(texts, labels, test_size=0.15, random_state=42)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:54:47.557746Z","iopub.execute_input":"2024-05-02T19:54:47.558545Z","iopub.status.idle":"2024-05-02T19:54:47.599306Z","shell.execute_reply.started":"2024-05-02T19:54:47.558512Z","shell.execute_reply":"2024-05-02T19:54:47.598385Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tokenizer = DistilBertTokenizer.from_pretrained(\"distilbert-base-uncased\")\nmodel = DistilBertForSequenceClassification.from_pretrained(\"distilbert-base-uncased\", num_labels = len(labels[0]),\n                                                           problem_type=\"multi_label_classification\")","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:54:50.507901Z","iopub.execute_input":"2024-05-02T19:54:50.508794Z","iopub.status.idle":"2024-05-02T19:54:53.725324Z","shell.execute_reply.started":"2024-05-02T19:54:50.508753Z","shell.execute_reply":"2024-05-02T19:54:53.724370Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class ECommerceFoundDataset(Dataset):\n    def __init__(self, texts, labels, tokenizer, max_token_len=512):\n        self.tokenizer = tokenizer\n        self.texts = texts\n        self.labels = labels\n        self.max_token_len = max_token_len\n\n    def __len__(self):\n        return len(self.texts)\n\n    def __getitem__(self, index):\n        prodInfo = str(self.texts[index])\n        label = torch.tensor(self.labels[index])\n\n        inputs = self.tokenizer(\n            prodInfo,\n            add_special_tokens=True,\n            padding=\"max_length\",\n            truncation=True,\n            max_length=self.max_token_len,\n            return_attention_mask=True,\n            return_tensors='pt'\n        )\n\n        return {\n            'input_ids': inputs['input_ids'].flatten(),\n            'attention_mask': inputs['attention_mask'].flatten(),\n            'labels': label\n        }","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:55:07.174307Z","iopub.execute_input":"2024-05-02T19:55:07.174960Z","iopub.status.idle":"2024-05-02T19:55:07.182359Z","shell.execute_reply.started":"2024-05-02T19:55:07.174928Z","shell.execute_reply":"2024-05-02T19:55:07.181415Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_dataset = ECommerceFoundDataset(train_inputs, train_labels, tokenizer)\nval_dataset = ECommerceFoundDataset(val_inputs, val_labels, tokenizer)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:56:00.665204Z","iopub.execute_input":"2024-05-02T19:56:00.665844Z","iopub.status.idle":"2024-05-02T19:56:00.670942Z","shell.execute_reply.started":"2024-05-02T19:56:00.665807Z","shell.execute_reply":"2024-05-02T19:56:00.669713Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#Evaluation Metrics\nfrom sklearn.metrics import roc_auc_score, f1_score, hamming_loss\nfrom transformers import EvalPrediction\n\ndef compute_metrics_multilabel(predictions, labels, threshold=0.5):\n    sigmoid = torch.nn.Sigmoid()\n    probability = sigmoid(torch.Tensor(predictions))\n    y_pred = np.zeros(probability.shape)\n    y_pred[np.where(probability>=threshold)] = 1\n    y_true_value = labels\n    f1_scoreV = f1_score(y_true_value, y_pred, average='macro')\n    \n    # Check if there is only one class present in y_true\n    if len(np.unique(y_true_value)) == 1:\n        # Handle the edge case where there is only one class\n        roc_auc = None\n    else:\n        # Calculate ROC AUC only if there are multiple classes\n        try:\n            roc_auc = roc_auc_score(y_true_value, y_pred, average='macro')\n        except ValueError:\n            # Handle the case where roc_auc_score throws an error due to single class\n            roc_auc = None\n        \n    hamming = hamming_loss(y_true_value, y_pred)\n    \n    metrics_value = {\n        \"f1_score\":f1_scoreV,\n        \"hamming_loss\":hamming,\n        \"roc_auc\": roc_auc\n    }\n    \n    return metrics_value\n\ndef compute_metrics(p:EvalPrediction):\n    preds = p.predictions[0] if isinstance(p.predictions, tuple) else p.predictions\n    result = compute_metrics_multilabel(predictions=preds,labels=p.label_ids)\n    \n    return result","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:56:05.503453Z","iopub.execute_input":"2024-05-02T19:56:05.504153Z","iopub.status.idle":"2024-05-02T19:56:05.516977Z","shell.execute_reply.started":"2024-05-02T19:56:05.504118Z","shell.execute_reply":"2024-05-02T19:56:05.516194Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from transformers import DataCollatorWithPadding\n\ndata_collator = DataCollatorWithPadding(tokenizer=tokenizer)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:56:08.171765Z","iopub.execute_input":"2024-05-02T19:56:08.172906Z","iopub.status.idle":"2024-05-02T19:56:19.106451Z","shell.execute_reply.started":"2024-05-02T19:56:08.172862Z","shell.execute_reply":"2024-05-02T19:56:19.105284Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from transformers import TrainingArguments, Trainer\n\nargs = TrainingArguments(\n    per_device_train_batch_size=8,\n    per_device_eval_batch_size=8,\n    learning_rate=5e-5,\n    weight_decay=0.01,\n    evaluation_strategy=\"epoch\",\n    save_strategy=\"epoch\",\n    output_dir='./results',\n    num_train_epochs=12,\n    save_steps=1000,\n    save_total_limit=2\n)\n\ntrainer = Trainer(model= model,\n                 args=args,\n                 train_dataset=train_dataset,\n                 eval_dataset=val_dataset,\n                 data_collator=data_collator,\n                 compute_metrics=compute_metrics)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:56:40.161483Z","iopub.execute_input":"2024-05-02T19:56:40.161839Z","iopub.status.idle":"2024-05-02T19:56:41.067270Z","shell.execute_reply.started":"2024-05-02T19:56:40.161811Z","shell.execute_reply":"2024-05-02T19:56:41.066292Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"trainer.train()","metadata":{"execution":{"iopub.status.busy":"2024-05-02T19:56:45.514922Z","iopub.execute_input":"2024-05-02T19:56:45.515723Z","iopub.status.idle":"2024-05-02T20:28:10.866262Z","shell.execute_reply.started":"2024-05-02T19:56:45.515691Z","shell.execute_reply":"2024-05-02T20:28:10.865333Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"trainer.evaluate()","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:28:20.043864Z","iopub.execute_input":"2024-05-02T20:28:20.044782Z","iopub.status.idle":"2024-05-02T20:28:32.500971Z","shell.execute_reply.started":"2024-05-02T20:28:20.044747Z","shell.execute_reply":"2024-05-02T20:28:32.500181Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Validation","metadata":{}},{"cell_type":"code","source":"# text = \"I want a soft pillow\" #Good\n# text = \"baby\"  #Good\n# text = \"music\"  #Good\n# text = \"pillow\"  #Good\n# text = \"diaper\"  #Bad\n# text = \"i want beauty items\" #Good\n# text = \"teeth whitening item\" #Bad\n# text = \"I want shoes\" #Good\n# text = \"I want to buy an acoustic guitar\" #Good\n# text = \"I want to buy a camera\" #Good\n# text = \"I want to buy pen camera\" #Good\n# text = \"Suggest me baby products, handmade and camera\" #Good\n# text = \"Suggest me baby products and guitar\"#Good\ntext = \"I want to buy a diaper\" #Good\n\n\n\ntext = preprocess_text(text)\ntext\n\nencoding = tokenizer(text, return_tensors='pt')\nencoding.to(model.device)\n\noutputs = model(**encoding)\n\n\nimport torch\nsigmoid = torch.nn.Sigmoid()\nprobs = sigmoid(outputs.logits[0].cpu())\n\n# Convert probabilities to numpy array\nprobs_np = probs.detach().numpy()\n\n# Thresholding to get predictions\npreds = np.zeros(probs_np.shape)\n\npreds[np.where(probs_np >= 0)] = 1\n\n# Inverse transform to get labels\nvalues = multilabel.inverse_transform(preds.reshape(1, -1))\n\nvalues = np.array(values[0])\n\nindices = np.where(preds == 1)\n\n# Extract the corresponding probability values for these indices\nlabel_probabilities = probs_np[indices]\n\n# Sort label_probabilities array in descending order and retain indices\nsorted_indices = np.argsort(label_probabilities)[::-1]\nsorted_label_probabilities = label_probabilities[sorted_indices]\n\n# Extract values from the list 'values' based on the sorted indices\nsorted_values = values[sorted_indices]\n\nprint(sorted_label_probabilities)\nprint(sorted_values)\n\n\n# Reset index of the DataFrame\ndf_reset_index = df.reset_index(drop=True)\n\n# Create a categorical variable with custom ordering\ncustom_order = pd.Categorical(df_reset_index['label'], categories=sorted_values, ordered=True)\n\n# Sort the DataFrame by the custom ordering\ndf_sorted = df_reset_index.loc[custom_order.argsort()]\n\n# Select rows where label is in sorted_values\nselected_rows = df_sorted[df_sorted['label'].isin(sorted_values)]\n\nselected_rows[:40]","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:37:30.959326Z","iopub.execute_input":"2024-05-02T20:37:30.959967Z","iopub.status.idle":"2024-05-02T20:37:31.036277Z","shell.execute_reply.started":"2024-05-02T20:37:30.959936Z","shell.execute_reply":"2024-05-02T20:37:31.035393Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"trainer.save_model(\"distilbert-finetuned-ecommerce\")\ntokenizer.save_pretrained(\"/kaggle/working/tokenizerRS\")\ndf.to_csv(\"encodedCSV.csv\",index=False)\nimport pickle\nwith open(\"multi-label-binarizer.pkl\",\"wb\") as f:\n    pickle.dump(multilabel,f)","metadata":{"execution":{"iopub.status.busy":"2024-05-02T20:32:01.954103Z","iopub.execute_input":"2024-05-02T20:32:01.954802Z","iopub.status.idle":"2024-05-02T20:32:02.863616Z","shell.execute_reply.started":"2024-05-02T20:32:01.954759Z","shell.execute_reply":"2024-05-02T20:32:02.862764Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Loading Stored model, tokenizer, labelled_csv, multilabelBinarizer","metadata":{}},{"cell_type":"markdown","source":"# Testing","metadata":{}},{"cell_type":"code","source":"import os\nimport torch\nfrom transformers import DistilBertTokenizer, DistilBertForSequenceClassification\nimport pickle\n\n\n# Load the tokenizer\ntokenizer1 = DistilBertTokenizer.from_pretrained(\"/kaggle/input/tokenizer-54k-12epoch-cross\")\n\n# Load the model\nmodel1 = DistilBertForSequenceClassification.from_pretrained(\"/kaggle/input/model-54k-12epochs-cross\")\n\n#load the label encoded csv\ndf1 = pd.read_csv('/kaggle/input/datasetcsv54k-12epochs-cross/encodedCSV.csv')\n\n# Load the multi-label binarizer\nwith open(\"/kaggle/input/multilabel-binarizer-54k-12epochs-cross/multi-label-binarizer.pkl\", \"rb\") as f:\n    multilabel1 = pickle.load(f)","metadata":{"execution":{"iopub.status.busy":"2024-05-03T14:21:22.585593Z","iopub.execute_input":"2024-05-03T14:21:22.586635Z","iopub.status.idle":"2024-05-03T14:21:36.599350Z","shell.execute_reply.started":"2024-05-03T14:21:22.586592Z","shell.execute_reply":"2024-05-03T14:21:36.598451Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"code","source":"import re\nfrom nltk.tokenize import word_tokenize\nfrom nltk.corpus import stopwords\nfrom nltk.stem import WordNetLemmatizer\nfrom nltk.stem import PorterStemmer\n\n#Download NLTK resources\nimport nltk\nimport subprocess\nnltk.download('punkt')\nnltk.download('stopwords')\n\ntry:\n    nltk.data.find('wordnet.zip')\nexcept:\n    nltk.download('wordnet', download_dir='/kaggle/working/')\n    command = \"unzip /kaggle/working/corpora/wordnet.zip -d /kaggle/working/corpora\"\n    subprocess.run(command.split())\n    nltk.data.path.append('/kaggle/working/')\n    \n# Function for text preprocessing\ndef preprocess_text(text):\n    # Remove HTML tags\n    text = re.sub(r'<.*?>', '', text)\n    \n    # Remove special characters and numbers\n    text = re.sub(r'[^a-zA-Z]', ' ', text)\n    \n    # Convert to lowercase\n    text = text.lower()\n    \n    # Tokenization\n    tokens = word_tokenize(text)\n    \n    # Remove stop words\n    stop_words = set(stopwords.words('english'))\n    tokens = [word for word in tokens if word not in stop_words]\n    \n    # Lemmatization\n    lemmatizer = WordNetLemmatizer()\n    tokens = [lemmatizer.lemmatize(word) for word in tokens]\n    \n    # Join tokens back into a string\n    preprocessed_text = ' '.join(tokens)\n    \n    return preprocessed_text\n\n\n","metadata":{"execution":{"iopub.status.busy":"2024-05-03T14:21:42.924743Z","iopub.execute_input":"2024-05-03T14:21:42.925125Z","iopub.status.idle":"2024-05-03T14:21:44.313029Z","shell.execute_reply.started":"2024-05-03T14:21:42.925092Z","shell.execute_reply":"2024-05-03T14:21:44.311885Z"},"trusted":true},"execution_count":3,"outputs":[{"name":"stdout","text":"[nltk_data] Downloading package punkt to /usr/share/nltk_data...\n[nltk_data]   Package punkt is already up-to-date!\n[nltk_data] Downloading package stopwords to /usr/share/nltk_data...\n[nltk_data]   Package stopwords is already up-to-date!\n[nltk_data] Downloading package wordnet to /kaggle/working/...\nArchive:  /kaggle/working/corpora/wordnet.zip\n   creating: /kaggle/working/corpora/wordnet/\n  inflating: /kaggle/working/corpora/wordnet/lexnames  \n  inflating: /kaggle/working/corpora/wordnet/data.verb  \n  inflating: /kaggle/working/corpora/wordnet/index.adv  \n  inflating: /kaggle/working/corpora/wordnet/adv.exc  \n  inflating: /kaggle/working/corpora/wordnet/index.verb  \n  inflating: /kaggle/working/corpora/wordnet/cntlist.rev  \n  inflating: /kaggle/working/corpora/wordnet/data.adj  \n  inflating: /kaggle/working/corpora/wordnet/index.adj  \n  inflating: /kaggle/working/corpora/wordnet/LICENSE  \n  inflating: /kaggle/working/corpora/wordnet/citation.bib  \n  inflating: /kaggle/working/corpora/wordnet/noun.exc  \n  inflating: /kaggle/working/corpora/wordnet/verb.exc  \n  inflating: /kaggle/working/corpora/wordnet/README  \n  inflating: /kaggle/working/corpora/wordnet/index.sense  \n  inflating: /kaggle/working/corpora/wordnet/data.noun  \n  inflating: /kaggle/working/corpora/wordnet/data.adv  \n  inflating: /kaggle/working/corpora/wordnet/index.noun  \n  inflating: /kaggle/working/corpora/wordnet/adj.exc  \n","output_type":"stream"}]},{"cell_type":"code","source":"# text = \"I want a soft pillow\" #Good\n# text = \"baby\"  #Good\n# text = \"suggest me good music\"  #Good\n# text = \"pillow\"  #Good\n# text = \"diaper\"  #Bad\n# text = \"i want beauty items\" #Good\n# text = \"teeth whitening item\" #Bad\n# text = \"grocery items\" #Good\n# text = \"suggest me some artwork\" #Good\n# text = \"I want shoes\" #Good\n# text = \"I need tea\" #Good\n# text = \"suggest me baby guitar\"#Good\n# text = \"I want cameras and amazon fashion\"#Good\n# text = \"I want sweater\" #Even though sweater not present in the dataset , it belongs to the \n                        #category amazon fashion\n# text = \"I want mobile phone\" #good\n# text = \"I want fridge\" #good(here referigerator is there...still able to find fridge )\n# text = \"Suggest me chocolates\" #good\ntext = \"I have an birthday party tommorow. Suggest me some products\"#Good\n# text=\"I want coffee and tea\" #Good\n# text = \"Want a very fast computational device \" #it is understanding that I need a fast computational device \n# text = \"Want a bottle and shoes\"  #understanding both of the concept\n# text = \"Want a gas and cooker\"\n\ntext = preprocess_text(text)\ntext\n\nencoding = tokenizer1(text, return_tensors='pt')\nencoding.to(model1.device)\n\noutputs = model1(**encoding)\n\n\n\nimport torch\nsigmoid = torch.nn.Sigmoid()\nprobs = sigmoid(outputs.logits[0].cpu())\n\n# Convert probabilities to numpy array\nprobs_np = probs.detach().numpy()\n\n# Thresholding to get predictions\npreds = np.zeros(probs_np.shape)\n\npreds[np.where(probs_np >= 0)] = 1\n\n# Inverse transform to get labels\nvalues = multilabel1.inverse_transform(preds.reshape(1, -1))\n\nvalues = np.array(values[0])\n\nindices = np.where(preds == 1)\n\n# Extract the corresponding probability values for these indices\nlabel_probabilities = probs_np[indices]\n\n# Sort label_probabilities array in descending order and retain indices\nsorted_indices = np.argsort(label_probabilities)[::-1]\nsorted_label_probabilities = label_probabilities[sorted_indices]\n\n# Extract values from the list 'values' based on the sorted indices\nsorted_values = values[sorted_indices]\n\n# print(sorted_label_probabilities)\n# print(sorted_values)\n\n\n\n\n# Reset index of the DataFrame\ndf_reset_index = df1.reset_index(drop=True)\n\n# Create a categorical variable with custom ordering\ncustom_order = pd.Categorical(df_reset_index['label'], categories=sorted_values, ordered=True)\n\n# Sort the DataFrame by the custom ordering\ndf_sorted = df_reset_index.loc[custom_order.argsort()]\n\n# Select rows where label is in sorted_values\nselected_rows = df_sorted[df_sorted['label'].isin(sorted_values)]\n\nselected_rows[:60]","metadata":{"execution":{"iopub.status.busy":"2024-05-03T14:22:18.754312Z","iopub.execute_input":"2024-05-03T14:22:18.755262Z","iopub.status.idle":"2024-05-03T14:22:18.878834Z","shell.execute_reply.started":"2024-05-03T14:22:18.755203Z","shell.execute_reply":"2024-05-03T14:22:18.877820Z"},"trusted":true},"execution_count":5,"outputs":[{"execution_count":5,"output_type":"execute_result","data":{"text/plain":"                 main_category  \\\n23958           Camera & Photo   \n28789             Toys & Games   \n18252             Toys & Games   \n41123                 Handmade   \n36844           Camera & Photo   \n35670           Camera & Photo   \n4434              Toys & Games   \n9279            Camera & Photo   \n34012             Toys & Games   \n24684             Toys & Games   \n16846             Toys & Games   \n9034              Toys & Games   \n953               Toys & Games   \n52105           Camera & Photo   \n42675   Health & Personal Care   \n42665             Toys & Games   \n26054             Toys & Games   \n44586             Toys & Games   \n22290           Camera & Photo   \n30820             Toys & Games   \n584            Office Products   \n49645           Camera & Photo   \n27243             Toys & Games   \n37723             Toys & Games   \n35082              Amazon Home   \n24477             Toys & Games   \n772             Camera & Photo   \n12404   Health & Personal Care   \n16747           Camera & Photo   \n19617             Toys & Games   \n35248             Toys & Games   \n12015           Camera & Photo   \n23243             Toys & Games   \n15083          Office Products   \n21042           Camera & Photo   \n16100             Toys & Games   \n42688      Musical Instruments   \n53945              Amazon Home   \n26411             Toys & Games   \n38050           Camera & Photo   \n25353           Camera & Photo   \n19960             Toys & Games   \n48221             Toys & Games   \n16672             Toys & Games   \n11693           Camera & Photo   \n32538             Toys & Games   \n22420           Camera & Photo   \n42392             Toys & Games   \n35042           Camera & Photo   \n43837   Health & Personal Care   \n35180             Toys & Games   \n52131             Toys & Games   \n17315   Health & Personal Care   \n2539              Toys & Games   \n8501              Toys & Games   \n500    Industrial & Scientific   \n11528             Toys & Games   \n4029            Camera & Photo   \n3743              Toys & Games   \n34351   Health & Personal Care   \n\n                                                   title  \\\n23958  Gatherfun Birthday Party Supplies Happy 25th B...   \n28789  Crseniny Happy Birthday Cake Topper,Gold Glitt...   \n18252  Level 9 Game Birthday Cake Topper - Video Game...   \n41123  42nd Birthday Party Decorations for 42nd Birth...   \n36844  Pimvimcim 17th Birthday Banner Decorations for...   \n35670  Mocsicka 3rd Donut Birthday Backdrop Three is ...   \n4434   Party Decorations Supplies cartoon game us The...   \n9279   Large Spiderman Happy Birthday Banner | Superh...   \n34012  213pcs Video Gamer Birthday Party Decorations ...   \n24684  Dinosaur Party - Happy Birthday Banner | Volca...   \n16846  Esweny Dog Birthday Party Supplies Decorations...   \n9034   Luxiocio We Will Miss You Balloon Banner Decor...   \n953    Blue Orchards Brick Table Covers (Pack of 2) -...   \n52105  Trgowaul 75th Birthday Decorations for Women 7...   \n42675  Gold 60th Birthday Decorations for Women, 60 B...   \n42665  Cute Party Pups Table Decorating Kit Birthday ...   \n26054  9 Pcs Cat Party Centerpieces for Tables Cat Bi...   \n44586  Butterfly Birthday Party Decorations Kit Encan...   \n22290  CSFOTO 8x8ft Happy Birthday Backdrop Black Gol...   \n30820  21st Birthday Decorations for Her Blue, Yoart ...   \n584    FALAMON 90th Birthday Decorations for Women or...   \n49645  OurWarm 5 x 7ft City Photography Backdrop and ...   \n27243  148 pcs Sport Balloons Arch Garland Football P...   \n37723  Happy 10th Birthday Cake Topper- Gold and Blac...   \n35082  COMBO MUPPE BABIE 8FT 10PC TRIANGLE BANNER PAR...   \n24477  9 Pieces Blue Silver 60th Birthday Decorations...   \n772    Happy Birthday Backdrop Banner Blue and Silver...   \n12404  2-Pack 12.6\" Large Happy Birthday Gift Bags fo...   \n16747  5665 Glow Neon Happy 49th Birthday Backdrop Ba...   \n19617  40th Birthday Decorations Supplies, Extra Larg...   \n35248  100pcs Gold Balloons Garland Different Size - ...   \n12015  Mocsicka Sweet 16 Birthday Backdrop for Girls ...   \n23243  KatchOn, Black and Gold Happy Birthday Banner ...   \n15083  QUPT 20 Video Game Birthday Party Invitations ...   \n21042  Happy 13th Birthday Door Banner Backdrop Decor...   \n16100  MRSpecialties Unicorn and Rainbow Balloons and...   \n42688  Black Panther Birthday Party Decorations 50PCS...   \n53945  Gatherfun Dog Paw Party Supplies Disposable Ta...   \n26411  Gepulin 106Pcs Baby Yoda Birthday Decorations ...   \n38050  5×3 FT Summer Beach Theme Backdrop Banner, Lua...   \n25353  Happy Sweet 16th Birthday Banner Backdrop Deco...   \n19960  3 Pieces Happy 80th Birthday Party Table Decor...   \n48221  Deloklte Football Happy 10th Birthday Cake Top...   \n16672  ALLHEARTDESIRES New Year Christmas Party Decor...   \n11693  Chemical Science Party Happy Birthday Photogra...   \n32538  1 PCS Happy Birthday Mom Cake Topper with Glit...   \n22420  Mexican Fiesta Theme Birthday Photography Back...   \n42392  DJMUMU 12Pcs Video Game Birthday Party Favors ...   \n35042  Avezano Glow Neon Birthday Background Neon Lig...   \n43837  Pink Gold Birthday Decorations Pom Poms Flower...   \n35180  Video Game Cake Topper, Glittery Happy 6th Bir...   \n52131  50 PCS Wizard Party Chocolate Decoration Golde...   \n17315  LASKYER Happy 10th Birthday Yard Sign Banner -...   \n2539     POP parties Fireman Party Happy Birthday Banner   \n8501   GOGOSO BlueBall Pit Balls for Toddlers 1-3, Pa...   \n500    Summer Pool Toys for Kids Birthday Gifts Party...   \n11528  Ufocusmi Sweet 16 Cake Toppers, Glitter Rose G...   \n4029   INPHER Black and Gold Birthday Decorations Bac...   \n3743   ASTER 1st Dinosaur Birthday Party Decorations ...   \n34351  Unicorn Party Supplies, Rainbow Plates (10 x 5...   \n\n                                            CummProdInfo  average_rating  \\\n23958  product gatherfun birthday party supply happy ...             4.8   \n28789  product crseniny happy birthday cake topper go...             4.7   \n18252  product level game birthday cake topper video ...             4.7   \n41123  product nd birthday party decoration nd birthd...             4.7   \n36844  product pimvimcim th birthday banner decoratio...             4.4   \n35670  product mocsicka rd donut birthday backdrop th...             4.7   \n4434   product party decoration supply cartoon game u...             4.3   \n9279   product large spiderman happy birthday banner ...             4.6   \n34012  product pc video gamer birthday party decorati...             4.4   \n24684  product dinosaur party happy birthday banner v...             4.5   \n16846  product esweny dog birthday party supply decor...             4.2   \n9034   product luxiocio miss balloon banner decoratio...             4.5   \n953    product blue orchard brick table cover pack x ...             4.7   \n52105  product trgowaul th birthday decoration woman ...             4.7   \n42675  product gold th birthday decoration woman birt...             4.7   \n42665  product cute party pup table decorating kit bi...             4.6   \n26054  product pc cat party centerpiece table cat bir...             4.4   \n44586  product butterfly birthday party decoration ki...             4.6   \n22290  product csfoto x ft happy birthday backdrop bl...             4.6   \n30820  product st birthday decoration blue yoart birt...             4.4   \n584    product falamon th birthday decoration woman m...             4.8   \n49645  product ourwarm x ft city photography backdrop...             4.8   \n27243  product pc sport balloon arch garland football...             4.5   \n37723  product happy th birthday cake topper gold bla...             4.5   \n35082  product combo muppe babie ft pc triangle banne...             3.0   \n24477  product piece blue silver th birthday decorati...             4.5   \n772    product happy birthday backdrop banner blue si...             5.0   \n12404  product pack large happy birthday gift bag bir...             4.7   \n16747  product glow neon happy th birthday backdrop b...             5.0   \n19617  product th birthday decoration supply extra la...             4.4   \n35248  product pc gold balloon garland different size...             4.1   \n12015  product mocsicka sweet birthday backdrop girl ...             4.7   \n23243  product katchon black gold happy birthday bann...             4.9   \n15083  product qupt video game birthday party invitat...             4.5   \n21042  product happy th birthday door banner backdrop...             4.7   \n16100  product mrspecialties unicorn rainbow balloon ...             3.6   \n42688  product black panther birthday party decoratio...             4.6   \n53945  product gatherfun dog paw party supply disposa...             4.5   \n26411  product gepulin pc baby yoda birthday decorati...             4.3   \n38050  product ft summer beach theme backdrop banner ...             4.7   \n25353  product happy sweet th birthday banner backdro...             4.6   \n19960  product piece happy th birthday party table de...             4.9   \n48221  product deloklte football happy th birthday ca...             4.7   \n16672  product allheartdesires new year christmas par...             4.3   \n11693  product chemical science party happy birthday ...             4.4   \n32538  product pc happy birthday mom cake topper glit...             4.6   \n22420  product mexican fiesta theme birthday photogra...             4.7   \n42392  product djmumu pc video game birthday party fa...             4.5   \n35042  product avezano glow neon birthday background ...             4.6   \n43837  product pink gold birthday decoration pom pom ...             4.6   \n35180  product video game cake topper glittery happy ...             4.4   \n52131  product pc wizard party chocolate decoration g...             4.7   \n17315  product laskyer happy th birthday yard sign ba...             4.7   \n2539   product pop party fireman party happy birthday...             4.6   \n8501   product gogoso blueball pit ball toddler pack ...             4.1   \n500    product summer pool toy kid birthday gift part...             4.7   \n11528  product ufocusmi sweet cake topper glitter ros...             4.3   \n4029   product inpher black gold birthday decoration ...             4.3   \n3743   product aster st dinosaur birthday party decor...             4.7   \n34351  product unicorn party supply rainbow plate x p...             4.6   \n\n       label                                         label_list  \n23958  19766  [19766, 9813, 47353, 21961, 36272, 21951, 2717...  \n28789  12640  [12640, 21949, 29749, 3942, 35381, 4020, 48075...  \n18252  27622  [27622, 12640, 11275, 1399, 35381, 194, 29749,...  \n41123   2102  [2102, 16926, 20437, 19766, 12640, 47353, 3101...  \n36844  36272  [36272, 47353, 2095, 19766, 21951, 25572, 2198...  \n35670  31013  [31013, 31017, 6076, 9813, 19766, 31018, 21989...  \n4434   35381  [35381, 1719, 9964, 12640, 8053, 1381, 9435, 9...  \n9279   27179  [27179, 19766, 9813, 31013, 21961, 47353, 3444...  \n34012   1259  [1259, 28508, 35381, 1381, 677, 19768, 47354, ...  \n24684  14322  [14322, 3579, 21439, 31012, 34861, 19766, 1432...  \n16846  16579  [16579, 448, 27186, 750, 19766, 1265, 21668, 3...  \n9034   28508  [28508, 1259, 677, 2069, 35381, 27622, 1265, 1...  \n953     8257  [8257, 27622, 12843, 1259, 12640, 2102, 19768,...  \n52105  47353  [47353, 36272, 19766, 9813, 21951, 25572, 2196...  \n42675  20437  [20437, 7955, 26972, 19766, 12640, 44599, 9813...  \n42665  12843  [12843, 8257, 27622, 2790, 1556, 2102, 19768, ...  \n26054   2790  [2790, 10632, 27622, 10634, 9691, 38714, 8257,...  \n44586   9282  [9282, 35381, 9435, 28754, 12640, 1719, 27622,...  \n22290   9813  [9813, 19766, 47353, 27179, 25572, 31013, 2196...  \n30820   1265  [1265, 27186, 750, 20698, 677, 2069, 21668, 44...  \n584    16926  [16926, 19766, 12640, 20437, 2102, 47353, 2793...  \n49645  34446  [34446, 44448, 27179, 10953, 9811, 7759, 27622...  \n27243    677  [677, 1265, 2069, 27186, 9386, 750, 448, 38402...  \n37723  21949  [21949, 20453, 12640, 21953, 2793, 49106, 4807...  \n35082   9691  [9691, 27622, 2790, 1259, 2432, 28508, 13232, ...  \n24477   2793  [2793, 21949, 1556, 47353, 16926, 19766, 36272...  \n772    21961  [21961, 19766, 3184, 9813, 47353, 25572, 31017...  \n12404   1058  [1058, 37149, 20437, 19766, 36302, 9813, 25572...  \n16747   2371  [2371, 6074, 9811, 2095, 21962, 36272, 19766, ...  \n19617   2095  [2095, 36272, 19766, 47353, 12538, 26469, 1040...  \n35248    448  [448, 21668, 27186, 750, 47039, 789, 39823, 27...  \n12015  31017  [31017, 31013, 21989, 21961, 47353, 19766, 981...  \n23243  25572  [25572, 9813, 19766, 47353, 36272, 21961, 2342...  \n15083  37841  [37841, 7163, 7819, 27622, 2102, 20437, 19766,...  \n21042  21951  [21951, 21989, 19766, 47353, 36272, 9813, 2196...  \n16100  29025  [29025, 27186, 48255, 750, 2069, 448, 18817, 2...  \n42688   8053  [8053, 35381, 12640, 9964, 2469, 3942, 29749, ...  \n53945  19768  [19768, 19766, 1259, 16579, 3656, 677, 2069, 1...  \n26411  20033  [20033, 35381, 1381, 1719, 18818, 9282, 448, 1...  \n38050   2432  [2432, 2095, 10405, 44636, 31013, 2371, 27179,...  \n25353  21989  [21989, 21951, 31017, 19766, 31013, 36272, 473...  \n19960   1556  [1556, 2793, 21949, 12640, 7821, 26469, 20437,...  \n48221  13917  [13917, 12640, 21949, 35381, 49106, 2793, 8053...  \n16672   3293  [3293, 27622, 2790, 2778, 28508, 19388, 12640,...  \n11693  10953  [10953, 30440, 9811, 21962, 9813, 7759, 34446,...  \n32538    194  [194, 12640, 191, 35381, 27622, 8053, 9964, 39...  \n22420  30440  [30440, 10953, 9813, 21962, 9811, 12640, 7759,...  \n42392  13232  [13232, 49106, 44263, 1259, 9691, 752, 8259, 3...  \n35042   6074  [6074, 9813, 19766, 2371, 31013, 27179, 47353,...  \n43837  36302  [36302, 25572, 20437, 47353, 19766, 9813, 3627...  \n35180  49106  [49106, 21949, 27621, 3942, 35381, 29749, 1264...  \n52131   2289  [2289, 27622, 35381, 28508, 4051, 1644, 1259, ...  \n17315  26469  [26469, 36272, 47353, 21961, 7817, 2095, 21949...  \n2539   34861  [34861, 25572, 14322, 2095, 10405, 44636, 2432...  \n8501   19388  [19388, 27622, 28508, 3293, 1259, 9691, 34446,...  \n500    44263  [44263, 13232, 36799, 1412, 20437, 53522, 1896...  \n11528  48075  [48075, 48768, 3277, 29749, 12640, 21949, 3406...  \n4029   23427  [23427, 19766, 448, 25572, 21961, 47353, 9813,...  \n3743    3579  [3579, 14322, 31012, 21439, 448, 35381, 51589,...  \n34351  48260  [48260, 48259, 21964, 17465, 27622, 3669, 4343...  ","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>main_category</th>\n      <th>title</th>\n      <th>CummProdInfo</th>\n      <th>average_rating</th>\n      <th>label</th>\n      <th>label_list</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>23958</th>\n      <td>Camera &amp; Photo</td>\n      <td>Gatherfun Birthday Party Supplies Happy 25th B...</td>\n      <td>product gatherfun birthday party supply happy ...</td>\n      <td>4.8</td>\n      <td>19766</td>\n      <td>[19766, 9813, 47353, 21961, 36272, 21951, 2717...</td>\n    </tr>\n    <tr>\n      <th>28789</th>\n      <td>Toys &amp; Games</td>\n      <td>Crseniny Happy Birthday Cake Topper,Gold Glitt...</td>\n      <td>product crseniny happy birthday cake topper go...</td>\n      <td>4.7</td>\n      <td>12640</td>\n      <td>[12640, 21949, 29749, 3942, 35381, 4020, 48075...</td>\n    </tr>\n    <tr>\n      <th>18252</th>\n      <td>Toys &amp; Games</td>\n      <td>Level 9 Game Birthday Cake Topper - Video Game...</td>\n      <td>product level game birthday cake topper video ...</td>\n      <td>4.7</td>\n      <td>27622</td>\n      <td>[27622, 12640, 11275, 1399, 35381, 194, 29749,...</td>\n    </tr>\n    <tr>\n      <th>41123</th>\n      <td>Handmade</td>\n      <td>42nd Birthday Party Decorations for 42nd Birth...</td>\n      <td>product nd birthday party decoration nd birthd...</td>\n      <td>4.7</td>\n      <td>2102</td>\n      <td>[2102, 16926, 20437, 19766, 12640, 47353, 3101...</td>\n    </tr>\n    <tr>\n      <th>36844</th>\n      <td>Camera &amp; Photo</td>\n      <td>Pimvimcim 17th Birthday Banner Decorations for...</td>\n      <td>product pimvimcim th birthday banner decoratio...</td>\n      <td>4.4</td>\n      <td>36272</td>\n      <td>[36272, 47353, 2095, 19766, 21951, 25572, 2198...</td>\n    </tr>\n    <tr>\n      <th>35670</th>\n      <td>Camera &amp; Photo</td>\n      <td>Mocsicka 3rd Donut Birthday Backdrop Three is ...</td>\n      <td>product mocsicka rd donut birthday backdrop th...</td>\n      <td>4.7</td>\n      <td>31013</td>\n      <td>[31013, 31017, 6076, 9813, 19766, 31018, 21989...</td>\n    </tr>\n    <tr>\n      <th>4434</th>\n      <td>Toys &amp; Games</td>\n      <td>Party Decorations Supplies cartoon game us The...</td>\n      <td>product party decoration supply cartoon game u...</td>\n      <td>4.3</td>\n      <td>35381</td>\n      <td>[35381, 1719, 9964, 12640, 8053, 1381, 9435, 9...</td>\n    </tr>\n    <tr>\n      <th>9279</th>\n      <td>Camera &amp; Photo</td>\n      <td>Large Spiderman Happy Birthday Banner | Superh...</td>\n      <td>product large spiderman happy birthday banner ...</td>\n      <td>4.6</td>\n      <td>27179</td>\n      <td>[27179, 19766, 9813, 31013, 21961, 47353, 3444...</td>\n    </tr>\n    <tr>\n      <th>34012</th>\n      <td>Toys &amp; Games</td>\n      <td>213pcs Video Gamer Birthday Party Decorations ...</td>\n      <td>product pc video gamer birthday party decorati...</td>\n      <td>4.4</td>\n      <td>1259</td>\n      <td>[1259, 28508, 35381, 1381, 677, 19768, 47354, ...</td>\n    </tr>\n    <tr>\n      <th>24684</th>\n      <td>Toys &amp; Games</td>\n      <td>Dinosaur Party - Happy Birthday Banner | Volca...</td>\n      <td>product dinosaur party happy birthday banner v...</td>\n      <td>4.5</td>\n      <td>14322</td>\n      <td>[14322, 3579, 21439, 31012, 34861, 19766, 1432...</td>\n    </tr>\n    <tr>\n      <th>16846</th>\n      <td>Toys &amp; Games</td>\n      <td>Esweny Dog Birthday Party Supplies Decorations...</td>\n      <td>product esweny dog birthday party supply decor...</td>\n      <td>4.2</td>\n      <td>16579</td>\n      <td>[16579, 448, 27186, 750, 19766, 1265, 21668, 3...</td>\n    </tr>\n    <tr>\n      <th>9034</th>\n      <td>Toys &amp; Games</td>\n      <td>Luxiocio We Will Miss You Balloon Banner Decor...</td>\n      <td>product luxiocio miss balloon banner decoratio...</td>\n      <td>4.5</td>\n      <td>28508</td>\n      <td>[28508, 1259, 677, 2069, 35381, 27622, 1265, 1...</td>\n    </tr>\n    <tr>\n      <th>953</th>\n      <td>Toys &amp; Games</td>\n      <td>Blue Orchards Brick Table Covers (Pack of 2) -...</td>\n      <td>product blue orchard brick table cover pack x ...</td>\n      <td>4.7</td>\n      <td>8257</td>\n      <td>[8257, 27622, 12843, 1259, 12640, 2102, 19768,...</td>\n    </tr>\n    <tr>\n      <th>52105</th>\n      <td>Camera &amp; Photo</td>\n      <td>Trgowaul 75th Birthday Decorations for Women 7...</td>\n      <td>product trgowaul th birthday decoration woman ...</td>\n      <td>4.7</td>\n      <td>47353</td>\n      <td>[47353, 36272, 19766, 9813, 21951, 25572, 2196...</td>\n    </tr>\n    <tr>\n      <th>42675</th>\n      <td>Health &amp; Personal Care</td>\n      <td>Gold 60th Birthday Decorations for Women, 60 B...</td>\n      <td>product gold th birthday decoration woman birt...</td>\n      <td>4.7</td>\n      <td>20437</td>\n      <td>[20437, 7955, 26972, 19766, 12640, 44599, 9813...</td>\n    </tr>\n    <tr>\n      <th>42665</th>\n      <td>Toys &amp; Games</td>\n      <td>Cute Party Pups Table Decorating Kit Birthday ...</td>\n      <td>product cute party pup table decorating kit bi...</td>\n      <td>4.6</td>\n      <td>12843</td>\n      <td>[12843, 8257, 27622, 2790, 1556, 2102, 19768, ...</td>\n    </tr>\n    <tr>\n      <th>26054</th>\n      <td>Toys &amp; Games</td>\n      <td>9 Pcs Cat Party Centerpieces for Tables Cat Bi...</td>\n      <td>product pc cat party centerpiece table cat bir...</td>\n      <td>4.4</td>\n      <td>2790</td>\n      <td>[2790, 10632, 27622, 10634, 9691, 38714, 8257,...</td>\n    </tr>\n    <tr>\n      <th>44586</th>\n      <td>Toys &amp; Games</td>\n      <td>Butterfly Birthday Party Decorations Kit Encan...</td>\n      <td>product butterfly birthday party decoration ki...</td>\n      <td>4.6</td>\n      <td>9282</td>\n      <td>[9282, 35381, 9435, 28754, 12640, 1719, 27622,...</td>\n    </tr>\n    <tr>\n      <th>22290</th>\n      <td>Camera &amp; Photo</td>\n      <td>CSFOTO 8x8ft Happy Birthday Backdrop Black Gol...</td>\n      <td>product csfoto x ft happy birthday backdrop bl...</td>\n      <td>4.6</td>\n      <td>9813</td>\n      <td>[9813, 19766, 47353, 27179, 25572, 31013, 2196...</td>\n    </tr>\n    <tr>\n      <th>30820</th>\n      <td>Toys &amp; Games</td>\n      <td>21st Birthday Decorations for Her Blue, Yoart ...</td>\n      <td>product st birthday decoration blue yoart birt...</td>\n      <td>4.4</td>\n      <td>1265</td>\n      <td>[1265, 27186, 750, 20698, 677, 2069, 21668, 44...</td>\n    </tr>\n    <tr>\n      <th>584</th>\n      <td>Office Products</td>\n      <td>FALAMON 90th Birthday Decorations for Women or...</td>\n      <td>product falamon th birthday decoration woman m...</td>\n      <td>4.8</td>\n      <td>16926</td>\n      <td>[16926, 19766, 12640, 20437, 2102, 47353, 2793...</td>\n    </tr>\n    <tr>\n      <th>49645</th>\n      <td>Camera &amp; Photo</td>\n      <td>OurWarm 5 x 7ft City Photography Backdrop and ...</td>\n      <td>product ourwarm x ft city photography backdrop...</td>\n      <td>4.8</td>\n      <td>34446</td>\n      <td>[34446, 44448, 27179, 10953, 9811, 7759, 27622...</td>\n    </tr>\n    <tr>\n      <th>27243</th>\n      <td>Toys &amp; Games</td>\n      <td>148 pcs Sport Balloons Arch Garland Football P...</td>\n      <td>product pc sport balloon arch garland football...</td>\n      <td>4.5</td>\n      <td>677</td>\n      <td>[677, 1265, 2069, 27186, 9386, 750, 448, 38402...</td>\n    </tr>\n    <tr>\n      <th>37723</th>\n      <td>Toys &amp; Games</td>\n      <td>Happy 10th Birthday Cake Topper- Gold and Blac...</td>\n      <td>product happy th birthday cake topper gold bla...</td>\n      <td>4.5</td>\n      <td>21949</td>\n      <td>[21949, 20453, 12640, 21953, 2793, 49106, 4807...</td>\n    </tr>\n    <tr>\n      <th>35082</th>\n      <td>Amazon Home</td>\n      <td>COMBO MUPPE BABIE 8FT 10PC TRIANGLE BANNER PAR...</td>\n      <td>product combo muppe babie ft pc triangle banne...</td>\n      <td>3.0</td>\n      <td>9691</td>\n      <td>[9691, 27622, 2790, 1259, 2432, 28508, 13232, ...</td>\n    </tr>\n    <tr>\n      <th>24477</th>\n      <td>Toys &amp; Games</td>\n      <td>9 Pieces Blue Silver 60th Birthday Decorations...</td>\n      <td>product piece blue silver th birthday decorati...</td>\n      <td>4.5</td>\n      <td>2793</td>\n      <td>[2793, 21949, 1556, 47353, 16926, 19766, 36272...</td>\n    </tr>\n    <tr>\n      <th>772</th>\n      <td>Camera &amp; Photo</td>\n      <td>Happy Birthday Backdrop Banner Blue and Silver...</td>\n      <td>product happy birthday backdrop banner blue si...</td>\n      <td>5.0</td>\n      <td>21961</td>\n      <td>[21961, 19766, 3184, 9813, 47353, 25572, 31017...</td>\n    </tr>\n    <tr>\n      <th>12404</th>\n      <td>Health &amp; Personal Care</td>\n      <td>2-Pack 12.6\" Large Happy Birthday Gift Bags fo...</td>\n      <td>product pack large happy birthday gift bag bir...</td>\n      <td>4.7</td>\n      <td>1058</td>\n      <td>[1058, 37149, 20437, 19766, 36302, 9813, 25572...</td>\n    </tr>\n    <tr>\n      <th>16747</th>\n      <td>Camera &amp; Photo</td>\n      <td>5665 Glow Neon Happy 49th Birthday Backdrop Ba...</td>\n      <td>product glow neon happy th birthday backdrop b...</td>\n      <td>5.0</td>\n      <td>2371</td>\n      <td>[2371, 6074, 9811, 2095, 21962, 36272, 19766, ...</td>\n    </tr>\n    <tr>\n      <th>19617</th>\n      <td>Toys &amp; Games</td>\n      <td>40th Birthday Decorations Supplies, Extra Larg...</td>\n      <td>product th birthday decoration supply extra la...</td>\n      <td>4.4</td>\n      <td>2095</td>\n      <td>[2095, 36272, 19766, 47353, 12538, 26469, 1040...</td>\n    </tr>\n    <tr>\n      <th>35248</th>\n      <td>Toys &amp; Games</td>\n      <td>100pcs Gold Balloons Garland Different Size - ...</td>\n      <td>product pc gold balloon garland different size...</td>\n      <td>4.1</td>\n      <td>448</td>\n      <td>[448, 21668, 27186, 750, 47039, 789, 39823, 27...</td>\n    </tr>\n    <tr>\n      <th>12015</th>\n      <td>Camera &amp; Photo</td>\n      <td>Mocsicka Sweet 16 Birthday Backdrop for Girls ...</td>\n      <td>product mocsicka sweet birthday backdrop girl ...</td>\n      <td>4.7</td>\n      <td>31017</td>\n      <td>[31017, 31013, 21989, 21961, 47353, 19766, 981...</td>\n    </tr>\n    <tr>\n      <th>23243</th>\n      <td>Toys &amp; Games</td>\n      <td>KatchOn, Black and Gold Happy Birthday Banner ...</td>\n      <td>product katchon black gold happy birthday bann...</td>\n      <td>4.9</td>\n      <td>25572</td>\n      <td>[25572, 9813, 19766, 47353, 36272, 21961, 2342...</td>\n    </tr>\n    <tr>\n      <th>15083</th>\n      <td>Office Products</td>\n      <td>QUPT 20 Video Game Birthday Party Invitations ...</td>\n      <td>product qupt video game birthday party invitat...</td>\n      <td>4.5</td>\n      <td>37841</td>\n      <td>[37841, 7163, 7819, 27622, 2102, 20437, 19766,...</td>\n    </tr>\n    <tr>\n      <th>21042</th>\n      <td>Camera &amp; Photo</td>\n      <td>Happy 13th Birthday Door Banner Backdrop Decor...</td>\n      <td>product happy th birthday door banner backdrop...</td>\n      <td>4.7</td>\n      <td>21951</td>\n      <td>[21951, 21989, 19766, 47353, 36272, 9813, 2196...</td>\n    </tr>\n    <tr>\n      <th>16100</th>\n      <td>Toys &amp; Games</td>\n      <td>MRSpecialties Unicorn and Rainbow Balloons and...</td>\n      <td>product mrspecialties unicorn rainbow balloon ...</td>\n      <td>3.6</td>\n      <td>29025</td>\n      <td>[29025, 27186, 48255, 750, 2069, 448, 18817, 2...</td>\n    </tr>\n    <tr>\n      <th>42688</th>\n      <td>Musical Instruments</td>\n      <td>Black Panther Birthday Party Decorations 50PCS...</td>\n      <td>product black panther birthday party decoratio...</td>\n      <td>4.6</td>\n      <td>8053</td>\n      <td>[8053, 35381, 12640, 9964, 2469, 3942, 29749, ...</td>\n    </tr>\n    <tr>\n      <th>53945</th>\n      <td>Amazon Home</td>\n      <td>Gatherfun Dog Paw Party Supplies Disposable Ta...</td>\n      <td>product gatherfun dog paw party supply disposa...</td>\n      <td>4.5</td>\n      <td>19768</td>\n      <td>[19768, 19766, 1259, 16579, 3656, 677, 2069, 1...</td>\n    </tr>\n    <tr>\n      <th>26411</th>\n      <td>Toys &amp; Games</td>\n      <td>Gepulin 106Pcs Baby Yoda Birthday Decorations ...</td>\n      <td>product gepulin pc baby yoda birthday decorati...</td>\n      <td>4.3</td>\n      <td>20033</td>\n      <td>[20033, 35381, 1381, 1719, 18818, 9282, 448, 1...</td>\n    </tr>\n    <tr>\n      <th>38050</th>\n      <td>Camera &amp; Photo</td>\n      <td>5×3 FT Summer Beach Theme Backdrop Banner, Lua...</td>\n      <td>product ft summer beach theme backdrop banner ...</td>\n      <td>4.7</td>\n      <td>2432</td>\n      <td>[2432, 2095, 10405, 44636, 31013, 2371, 27179,...</td>\n    </tr>\n    <tr>\n      <th>25353</th>\n      <td>Camera &amp; Photo</td>\n      <td>Happy Sweet 16th Birthday Banner Backdrop Deco...</td>\n      <td>product happy sweet th birthday banner backdro...</td>\n      <td>4.6</td>\n      <td>21989</td>\n      <td>[21989, 21951, 31017, 19766, 31013, 36272, 473...</td>\n    </tr>\n    <tr>\n      <th>19960</th>\n      <td>Toys &amp; Games</td>\n      <td>3 Pieces Happy 80th Birthday Party Table Decor...</td>\n      <td>product piece happy th birthday party table de...</td>\n      <td>4.9</td>\n      <td>1556</td>\n      <td>[1556, 2793, 21949, 12640, 7821, 26469, 20437,...</td>\n    </tr>\n    <tr>\n      <th>48221</th>\n      <td>Toys &amp; Games</td>\n      <td>Deloklte Football Happy 10th Birthday Cake Top...</td>\n      <td>product deloklte football happy th birthday ca...</td>\n      <td>4.7</td>\n      <td>13917</td>\n      <td>[13917, 12640, 21949, 35381, 49106, 2793, 8053...</td>\n    </tr>\n    <tr>\n      <th>16672</th>\n      <td>Toys &amp; Games</td>\n      <td>ALLHEARTDESIRES New Year Christmas Party Decor...</td>\n      <td>product allheartdesires new year christmas par...</td>\n      <td>4.3</td>\n      <td>3293</td>\n      <td>[3293, 27622, 2790, 2778, 28508, 19388, 12640,...</td>\n    </tr>\n    <tr>\n      <th>11693</th>\n      <td>Camera &amp; Photo</td>\n      <td>Chemical Science Party Happy Birthday Photogra...</td>\n      <td>product chemical science party happy birthday ...</td>\n      <td>4.4</td>\n      <td>10953</td>\n      <td>[10953, 30440, 9811, 21962, 9813, 7759, 34446,...</td>\n    </tr>\n    <tr>\n      <th>32538</th>\n      <td>Toys &amp; Games</td>\n      <td>1 PCS Happy Birthday Mom Cake Topper with Glit...</td>\n      <td>product pc happy birthday mom cake topper glit...</td>\n      <td>4.6</td>\n      <td>194</td>\n      <td>[194, 12640, 191, 35381, 27622, 8053, 9964, 39...</td>\n    </tr>\n    <tr>\n      <th>22420</th>\n      <td>Camera &amp; Photo</td>\n      <td>Mexican Fiesta Theme Birthday Photography Back...</td>\n      <td>product mexican fiesta theme birthday photogra...</td>\n      <td>4.7</td>\n      <td>30440</td>\n      <td>[30440, 10953, 9813, 21962, 9811, 12640, 7759,...</td>\n    </tr>\n    <tr>\n      <th>42392</th>\n      <td>Toys &amp; Games</td>\n      <td>DJMUMU 12Pcs Video Game Birthday Party Favors ...</td>\n      <td>product djmumu pc video game birthday party fa...</td>\n      <td>4.5</td>\n      <td>13232</td>\n      <td>[13232, 49106, 44263, 1259, 9691, 752, 8259, 3...</td>\n    </tr>\n    <tr>\n      <th>35042</th>\n      <td>Camera &amp; Photo</td>\n      <td>Avezano Glow Neon Birthday Background Neon Lig...</td>\n      <td>product avezano glow neon birthday background ...</td>\n      <td>4.6</td>\n      <td>6074</td>\n      <td>[6074, 9813, 19766, 2371, 31013, 27179, 47353,...</td>\n    </tr>\n    <tr>\n      <th>43837</th>\n      <td>Health &amp; Personal Care</td>\n      <td>Pink Gold Birthday Decorations Pom Poms Flower...</td>\n      <td>product pink gold birthday decoration pom pom ...</td>\n      <td>4.6</td>\n      <td>36302</td>\n      <td>[36302, 25572, 20437, 47353, 19766, 9813, 3627...</td>\n    </tr>\n    <tr>\n      <th>35180</th>\n      <td>Toys &amp; Games</td>\n      <td>Video Game Cake Topper, Glittery Happy 6th Bir...</td>\n      <td>product video game cake topper glittery happy ...</td>\n      <td>4.4</td>\n      <td>49106</td>\n      <td>[49106, 21949, 27621, 3942, 35381, 29749, 1264...</td>\n    </tr>\n    <tr>\n      <th>52131</th>\n      <td>Toys &amp; Games</td>\n      <td>50 PCS Wizard Party Chocolate Decoration Golde...</td>\n      <td>product pc wizard party chocolate decoration g...</td>\n      <td>4.7</td>\n      <td>2289</td>\n      <td>[2289, 27622, 35381, 28508, 4051, 1644, 1259, ...</td>\n    </tr>\n    <tr>\n      <th>17315</th>\n      <td>Health &amp; Personal Care</td>\n      <td>LASKYER Happy 10th Birthday Yard Sign Banner -...</td>\n      <td>product laskyer happy th birthday yard sign ba...</td>\n      <td>4.7</td>\n      <td>26469</td>\n      <td>[26469, 36272, 47353, 21961, 7817, 2095, 21949...</td>\n    </tr>\n    <tr>\n      <th>2539</th>\n      <td>Toys &amp; Games</td>\n      <td>POP parties Fireman Party Happy Birthday Banner</td>\n      <td>product pop party fireman party happy birthday...</td>\n      <td>4.6</td>\n      <td>34861</td>\n      <td>[34861, 25572, 14322, 2095, 10405, 44636, 2432...</td>\n    </tr>\n    <tr>\n      <th>8501</th>\n      <td>Toys &amp; Games</td>\n      <td>GOGOSO BlueBall Pit Balls for Toddlers 1-3, Pa...</td>\n      <td>product gogoso blueball pit ball toddler pack ...</td>\n      <td>4.1</td>\n      <td>19388</td>\n      <td>[19388, 27622, 28508, 3293, 1259, 9691, 34446,...</td>\n    </tr>\n    <tr>\n      <th>500</th>\n      <td>Industrial &amp; Scientific</td>\n      <td>Summer Pool Toys for Kids Birthday Gifts Party...</td>\n      <td>product summer pool toy kid birthday gift part...</td>\n      <td>4.7</td>\n      <td>44263</td>\n      <td>[44263, 13232, 36799, 1412, 20437, 53522, 1896...</td>\n    </tr>\n    <tr>\n      <th>11528</th>\n      <td>Toys &amp; Games</td>\n      <td>Ufocusmi Sweet 16 Cake Toppers, Glitter Rose G...</td>\n      <td>product ufocusmi sweet cake topper glitter ros...</td>\n      <td>4.3</td>\n      <td>48075</td>\n      <td>[48075, 48768, 3277, 29749, 12640, 21949, 3406...</td>\n    </tr>\n    <tr>\n      <th>4029</th>\n      <td>Camera &amp; Photo</td>\n      <td>INPHER Black and Gold Birthday Decorations Bac...</td>\n      <td>product inpher black gold birthday decoration ...</td>\n      <td>4.3</td>\n      <td>23427</td>\n      <td>[23427, 19766, 448, 25572, 21961, 47353, 9813,...</td>\n    </tr>\n    <tr>\n      <th>3743</th>\n      <td>Toys &amp; Games</td>\n      <td>ASTER 1st Dinosaur Birthday Party Decorations ...</td>\n      <td>product aster st dinosaur birthday party decor...</td>\n      <td>4.7</td>\n      <td>3579</td>\n      <td>[3579, 14322, 31012, 21439, 448, 35381, 51589,...</td>\n    </tr>\n    <tr>\n      <th>34351</th>\n      <td>Health &amp; Personal Care</td>\n      <td>Unicorn Party Supplies, Rainbow Plates (10 x 5...</td>\n      <td>product unicorn party supply rainbow plate x p...</td>\n      <td>4.6</td>\n      <td>48260</td>\n      <td>[48260, 48259, 21964, 17465, 27622, 3669, 4343...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}