Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a2a7bb63",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cuda\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import torch\n",
    "from torch import nn\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from sklearn.preprocessing import LabelEncoder, MultiLabelBinarizer, StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "import tensorflow as tf\n",
    "import tensorflow_hub as hub\n",
    "from transformers import CLIPProcessor, CLIPModel,BertTokenizer, BertModel\n",
    "from PIL import Image\n",
    "from tqdm import tqdm\n",
    "import random\n",
    "import os\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "print(f\"Using device: {device}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ca4f4e82",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['review_id', 'user_id', 'business_id', 'stars', 'useful_review', 'funny_review', 'cool_review', 'text', 'date', 'name', 'review_count', 'yelping_since', 'useful_user', 'funny_user', 'cool_user', 'elite', 'friends', 'fans', 'average_stars', 'compliment_hot', 'compliment_more', 'compliment_profile', 'compliment_cute', 'compliment_list', 'compliment_note', 'compliment_plain', 'compliment_cool', 'compliment_funny', 'compliment_writer', 'compliment_photos', 'name_business', 'address', 'city', 'state', 'postal_code', 'latitude', 'longitude', 'stars_business', 'review_count_business', 'is_open', 'attributes', 'categories', 'hours', 'user_index', 'business_index']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "business_df = pd.read_json('yelp_academic_dataset_business.json', lines=True,nrows=10000)\n",
    "user_df = pd.read_json('yelp_academic_dataset_user.json', lines=True,nrows=10000)\n",
    "review_df = pd.read_json('yelp_academic_dataset_review.json', lines=True,nrows=10000)\n",
    "photo_df = pd.read_json('photos.json', lines=True,nrows=10000)\n",
    "business_ids = set(business_df['business_id'])\n",
    "user_ids = set(user_df['user_id'])\n",
    "review_df = review_df[review_df['business_id'].isin(business_ids) & review_df['user_id'].isin(user_ids)]\n",
    "\n",
    "data = pd.merge(review_df, user_df, on='user_id', suffixes=('_review', '_user'))\n",
    "data = pd.merge(data, business_df, on='business_id', suffixes=('', '_business'))\n",
    "\n",
    "user_encoder = LabelEncoder()\n",
    "data['user_index'] = user_encoder.fit_transform(data['user_id'])\n",
    "business_encoder = LabelEncoder()\n",
    "data['business_index'] = business_encoder.fit_transform(data['business_id'])\n",
    "num_users = data['user_index'].nunique()\n",
    "num_businesses = data['business_index'].nunique()\n",
    "print(data.columns.tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d2482a28",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Missing columns: []\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at ./bert-base-uncased were not used when initializing BertModel: ['cls.predictions.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.dense.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.decoder.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias']\n",
      "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "100%|██████████████████| 592/592 [00:12<00:00, 46.32it/s]\n",
      "100%|█████████████| 10000/10000 [00:17<00:00, 569.73it/s]\n"
     ]
    }
   ],
   "source": [
    "user_features_list = ['review_count', 'useful_user', 'funny_user', 'cool_user', 'fans', 'average_stars']\n",
    "\n",
    "scaler = StandardScaler()\n",
    "user_features = scaler.fit_transform(data[user_features_list])\n",
    "def process_categories(x):\n",
    "    if isinstance(x, list):\n",
    "        return x\n",
    "    elif isinstance(x, str):\n",
    "        return x.split(', ')\n",
    "    else:\n",
    "        return []\n",
    "\n",
    "business_df['categories'] = business_df['categories'].fillna('').apply(process_categories)\n",
    "mlb = MultiLabelBinarizer()\n",
    "categories_encoded = mlb.fit_transform(business_df['categories'])\n",
    "categories_df = pd.DataFrame(categories_encoded, columns=mlb.classes_)\n",
    "\n",
    "business_df = business_df.reset_index(drop=True)\n",
    "categories_df = categories_df.reset_index(drop=True)\n",
    "\n",
    "business_df = pd.concat([business_df, categories_df], axis=1)\n",
    "\n",
    "business_columns_needed = ['business_id', 'latitude', 'longitude'] + list(mlb.classes_)\n",
    "business_df_reduced = business_df[business_columns_needed]\n",
    "data = pd.merge(data, business_df_reduced, on='business_id', how='left')\n",
    "business_id_to_index = dict(zip(business_encoder.classes_, business_encoder.transform(business_encoder.classes_)))\n",
    "\n",
    "business_index_to_categories = {}\n",
    "for idx, row in business_df.iterrows():\n",
    "    business_id = row['business_id']\n",
    "    categories = row['categories'] \n",
    "    business_index = business_id_to_index.get(business_id)\n",
    "    if business_index is not None:\n",
    "        business_index_to_categories[business_index] = categories\n",
    "\n",
    "category_to_business_indices = {}\n",
    "for business_index, categories in business_index_to_categories.items():\n",
    "    for category in categories:\n",
    "        if category not in category_to_business_indices:\n",
    "            category_to_business_indices[category] = set()\n",
    "        category_to_business_indices[category].add(business_index)\n",
    "business_features_list = ['latitude', 'longitude'] + list(mlb.classes_)\n",
    "\n",
    "missing_columns = [col for col in business_features_list if col not in data.columns]\n",
    "for col in missing_columns:\n",
    "    data[col] = 0\n",
    "business_features = scaler.fit_transform(data[business_features_list])\n",
    "missing_columns = [col for col in business_features_list if col not in data.columns]\n",
    "print(f\"Missing columns: {missing_columns}\")\n",
    "business_features = scaler.fit_transform(data[business_features_list])\n",
    "use_model = hub.load('https://tfhub.dev/google/universal-sentence-encoder/4')\n",
    "tokenizer = BertTokenizer.from_pretrained('./bert-base-uncased')\n",
    "bert_model = BertModel.from_pretrained('./bert-base-uncased')\n",
    "bert_model.to(device)\n",
    "bert_model.eval()\n",
    "\n",
    "def get_bert_embedding(text):\n",
    "    if not isinstance(text, str) or text.strip() == '':\n",
    "        return np.zeros(768)\n",
    "    try:\n",
    "        inputs = tokenizer(text, return_tensors='pt', truncation=True, max_length=512)\n",
    "        inputs = {k: v.to(device) for k, v in inputs.items()}\n",
    "        with torch.no_grad():\n",
    "            outputs = bert_model(**inputs)\n",
    "        embedding = outputs.last_hidden_state[:, 0, :].cpu().numpy()[0]\n",
    "        return embedding\n",
    "    except Exception as e:\n",
    "        print(f\"Error processing text: {text}\")\n",
    "        print(e)\n",
    "        return np.zeros(768)\n",
    "data = data.dropna(subset=['text'])\n",
    "tqdm.pandas()\n",
    "data['text_embedding'] = data['text'].progress_apply(get_bert_embedding)\n",
    "photo_df = photo_df[photo_df['business_id'].isin(business_df['business_id'])]\n",
    "business_photos = photo_df.groupby('business_id')['photo_id'].apply(list).reset_index()\n",
    "business_df = pd.merge(business_df, business_photos, on='business_id', how='left')\n",
    "\n",
    "clip_model = CLIPModel.from_pretrained(\"./clip-vit-base-patch32\")\n",
    "clip_processor = CLIPProcessor.from_pretrained(\"./clip-vit-base-patch32\")\n",
    "clip_model.to(device)\n",
    "clip_model.eval()\n",
    "\n",
    "def get_image_embedding(image_path):\n",
    "    try:\n",
    "        image = Image.open(image_path).convert(\"RGB\")\n",
    "        inputs = clip_processor(images=image, return_tensors=\"pt\").to(device)\n",
    "        with torch.no_grad():\n",
    "            outputs = clip_model.get_image_features(**inputs)\n",
    "        embedding = outputs.cpu().numpy()[0]\n",
    "        return embedding\n",
    "    except Exception as e:\n",
    "        return np.zeros(512)\n",
    "\n",
    "def get_business_image_embedding(photo_ids):\n",
    "    embeddings = []\n",
    "    if not isinstance(photo_ids, list):\n",
    "        return np.zeros(512)\n",
    "    for photo_id in photo_ids:\n",
    "        image_path = os.path.join('photos', f'{photo_id}.jpg') \n",
    "        embedding = get_image_embedding(image_path)\n",
    "        if np.count_nonzero(embedding) > 0:\n",
    "            embeddings.append(embedding)\n",
    "    if embeddings:\n",
    "        return np.mean(embeddings, axis=0)\n",
    "    else:\n",
    "        return np.zeros(512)\n",
    "\n",
    "tqdm.pandas()\n",
    "business_df['image_embedding'] = business_df['photo_id'].progress_apply(get_business_image_embedding)\n",
    "\n",
    "business_image_embeddings = business_df[['business_id', 'image_embedding']]\n",
    "data = pd.merge(data, business_image_embeddings, on='business_id', how='left')\n",
    "user_features = np.array(user_features)\n",
    "business_features = np.array(business_features)\n",
    "text_embeddings = np.stack(data['text_embedding'].values)\n",
    "image_embeddings = np.stack(data['image_embedding'].values)\n",
    "\n",
    "user_indices = data['user_index'].values\n",
    "business_indices = data['business_index'].values\n",
    "\n",
    "user_feature_dict = {}\n",
    "for idx, user_id in enumerate(data['user_index'].unique()):\n",
    "    user_feature_dict[user_id] = user_features[data['user_index'] == user_id][0]\n",
    "\n",
    "business_feature_dict = {}\n",
    "for idx, business_id in enumerate(data['business_index'].unique()):\n",
    "    business_feature_dict[business_id] = business_features[data['business_index'] == business_id][0]\n",
    "\n",
    "business_text_embedding_dict = {}\n",
    "for idx, business_id in enumerate(data['business_index'].unique()):\n",
    "    business_text_embedding_dict[business_id] = text_embeddings[data['business_index'] == business_id][0]\n",
    "\n",
    "business_image_embedding_dict = {}\n",
    "for idx, business_id in enumerate(data['business_index'].unique()):\n",
    "    business_image_embedding_dict[business_id] = image_embeddings[data['business_index'] == business_id][0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f4b576e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "class YelpDataset(Dataset):\n",
    "    def __init__(self, data, num_businesses, business_index_to_categories, category_to_business_indices, num_negatives=10):\n",
    "        self.data = data\n",
    "        self.num_businesses = num_businesses\n",
    "        self.num_negatives = num_negatives\n",
    "        self.business_index_to_categories = business_index_to_categories\n",
    "        self.category_to_business_indices = category_to_business_indices\n",
    "    \n",
    "        self.user_business_set = set(zip(data['user_index'], data['business_index']))\n",
    "        \n",
    "\n",
    "        self.user_positive_businesses = data.groupby('user_index')['business_index'].apply(set).to_dict()\n",
    "        \n",
    "\n",
    "        self.all_businesses = set(range(num_businesses))\n",
    "        \n",
    "        self.user_indices = data['user_index'].values\n",
    "        self.business_indices = data['business_index'].values\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.user_indices)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        user = self.user_indices[idx]\n",
    "        pos_business = self.business_indices[idx]\n",
    "\n",
    "        candidate_neg_businesses = self.all_businesses - self.user_positive_businesses[user]\n",
    "\n",
    "        neg_businesses = self.sample_hard_negatives(pos_business, candidate_neg_businesses, self.num_negatives)\n",
    "\n",
    "        return user, pos_business, neg_businesses\n",
    "\n",
    "    def sample_hard_negatives(self, pos_business, candidate_neg_businesses, num_negatives):\n",
    "        neg_businesses = random.sample(candidate_neg_businesses, num_negatives)\n",
    "        return neg_businesses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e09891f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data, val_data = train_test_split(data, test_size=0.2, random_state=42)\n",
    "\n",
    "train_dataset = YelpDataset(train_data, num_businesses, business_index_to_categories, category_to_business_indices, num_negatives=5)\n",
    "val_dataset = YelpDataset(val_data, num_businesses, business_index_to_categories, category_to_business_indices, num_negatives=5)\n",
    "def custom_collate(batch):\n",
    "    user_indices = torch.tensor([item[0] for item in batch], dtype=torch.long)\n",
    "    pos_business_indices = torch.tensor([item[1] for item in batch], dtype=torch.long)\n",
    "    neg_business_indices = [item[2] for item in batch] \n",
    "    neg_business_indices = torch.tensor(neg_business_indices, dtype=torch.long)\n",
    "    return user_indices, pos_business_indices, neg_business_indices\n",
    "train_loader = DataLoader(train_dataset, batch_size=256, shuffle=True, collate_fn=custom_collate)\n",
    "val_loader = DataLoader(val_dataset, batch_size=256, shuffle=False, collate_fn=custom_collate)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "73ea13ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "class TwoTowerModel(nn.Module):\n",
    "    def __init__(self, user_feature_dim, business_feature_dim, embedding_dim):\n",
    "        super(TwoTowerModel, self).__init__()\n",
    "        self.user_tower = nn.Sequential(\n",
    "            nn.Linear(user_feature_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, embedding_dim)\n",
    "        )\n",
    "        self.business_tower = nn.Sequential(\n",
    "            nn.Linear(business_feature_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, embedding_dim)\n",
    "        )\n",
    "\n",
    "    def forward(self, user_features, business_features):\n",
    "        user_vector = self.user_tower(user_features)\n",
    "        business_vector = self.business_tower(business_features)\n",
    "        user_vector = F.normalize(user_vector, p=2, dim=1)\n",
    "        business_vector = F.normalize(business_vector, p=2, dim=1)\n",
    "        return user_vector, business_vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "fc958af9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def contrastive_loss(user_vector, pos_business_vector, neg_business_vectors, temperature=0.07):\n",
    "    batch_size = user_vector.size(0)\n",
    "    user_vector = user_vector.unsqueeze(1) \n",
    "    pos_business_vector = pos_business_vector.unsqueeze(1)\n",
    "    all_business_vectors = torch.cat([pos_business_vector, neg_business_vectors], dim=1) \n",
    "    sim_matrix = torch.bmm(user_vector, all_business_vectors.permute(0, 2, 1)) \n",
    "    sim_matrix = sim_matrix.squeeze(1) / temperature  \n",
    "    \n",
    "    labels = torch.zeros(batch_size, dtype=torch.long).to(device)\n",
    "    loss = F.cross_entropy(sim_matrix, labels)\n",
    "    return loss\n",
    "def get_hard_negatives(user_idx, pos_business_idx, num_negatives):\n",
    "    user_positive_businesses = user_positive_businesses_dict[user_idx]\n",
    "    candidate_neg_businesses = all_businesses - user_positive_businesses\n",
    "    hard_negatives = random.sample(candidate_neg_businesses, num_negatives)\n",
    "    return hard_negatives\n",
    "def get_user_features(user_indices):\n",
    "    user_features_batch = []\n",
    "    for user_idx in user_indices.cpu().numpy():\n",
    "        user_features_batch.append(user_feature_dict[user_idx])\n",
    "    user_features_batch = torch.tensor(user_features_batch, dtype=torch.float).to(device)\n",
    "    return user_features_batch\n",
    "def get_business_features(business_indices):\n",
    "    business_features_batch = []\n",
    "    for business_idx in business_indices.cpu().numpy():\n",
    "        features = np.concatenate([\n",
    "            business_feature_dict[business_idx],\n",
    "            business_text_embedding_dict[business_idx],\n",
    "            business_image_embedding_dict[business_idx]\n",
    "        ])\n",
    "        business_features_batch.append(features)\n",
    "    business_features_batch = torch.tensor(business_features_batch, dtype=torch.float).to(device)\n",
    "    return business_features_batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "23226abb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_model(model, train_loader, optimizer, epoch):\n",
    "    model.train()\n",
    "    total_loss = 0\n",
    "    for batch_idx, batch in enumerate(tqdm(train_loader, desc=f\"Training Epoch {epoch+1}\")):\n",
    "        user_indices, pos_business_indices, neg_business_indices = batch\n",
    "        user_indices = user_indices.to(device)\n",
    "        pos_business_indices = pos_business_indices.to(device)\n",
    "        neg_business_indices = neg_business_indices.to(device)\n",
    "\n",
    "        user_features_batch = get_user_features(user_indices) \n",
    "\n",
    "        pos_business_features_batch = get_business_features(pos_business_indices) \n",
    "\n",
    "        # 获取负样本商家特征\n",
    "        batch_size, num_negatives = neg_business_indices.shape\n",
    "        neg_business_indices_flat = neg_business_indices.view(-1)\n",
    "        neg_business_features_batch = get_business_features(neg_business_indices_flat)  \n",
    "        neg_business_features_batch = neg_business_features_batch.view(batch_size, num_negatives, -1)  \n",
    "\n",
    "        optimizer.zero_grad()\n",
    "        user_vector = model.user_tower(user_features_batch) \n",
    "        pos_business_vector = model.business_tower(pos_business_features_batch)  \n",
    "        neg_business_vectors = []\n",
    "        for i in range(num_negatives):\n",
    "            neg_business_vector = model.business_tower(neg_business_features_batch[:, i, :]) \n",
    "            neg_business_vectors.append(neg_business_vector.unsqueeze(1))  \n",
    "        neg_business_vectors = torch.cat(neg_business_vectors, dim=1) \n",
    "\n",
    "        loss = contrastive_loss(user_vector, pos_business_vector, neg_business_vectors)\n",
    "\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        total_loss += loss.item()\n",
    "    avg_loss = total_loss / len(train_loader)\n",
    "    print(f\"Epoch {epoch+1}, Training Loss: {avg_loss:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5ce79fb5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_model(model, val_loader, epoch, k=10):\n",
    "    model.eval()\n",
    "    recalls = []\n",
    "    with torch.no_grad():\n",
    "        for batch in tqdm(val_loader, desc=f\"Evaluating Epoch {epoch+1}\"):\n",
    "            user_indices, pos_business_indices, _ = batch\n",
    "            user_features_batch = get_user_features(user_indices)\n",
    "            pos_business_indices = pos_business_indices.cpu().numpy()\n",
    "\n",
    "            user_vector = model.user_tower(user_features_batch)\n",
    "            user_vector = F.normalize(user_vector, p=2, dim=1)\n",
    "            business_vectors = []\n",
    "            for i in range(num_businesses):\n",
    "                features = get_business_features(torch.tensor([i]))\n",
    "                business_vector = model.business_tower(features)\n",
    "                business_vector = F.normalize(business_vector, p=2, dim=1)\n",
    "                business_vectors.append(business_vector)\n",
    "            business_vectors = torch.cat(business_vectors, dim=0)\n",
    "\n",
    "            scores = torch.matmul(user_vector, business_vectors.T)  \n",
    "\n",
    "            _, indices = torch.topk(scores, k, dim=1)\n",
    "\n",
    "            for i in range(len(user_indices)):\n",
    "                pos_business = pos_business_indices[i]\n",
    "                recommended_businesses = indices[i].cpu().numpy()\n",
    "                if pos_business in recommended_businesses:\n",
    "                    recalls.append(1)\n",
    "                else:\n",
    "                    recalls.append(0)\n",
    "    recall_at_k = sum(recalls) / len(recalls)\n",
    "    print(f\"Epoch {epoch+1}, Recall@{k}: {recall_at_k:.4f}\")\n",
    "    return recall_at_k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f90611e1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "business_features.shape: (592, 1061)\n",
      "business_feature_dim: 2341\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 1: 100%|████| 2/2 [00:00<00:00,  5.18it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1, Training Loss: 1.8725\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 1: 100%|██| 1/1 [00:00<00:00,  5.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1, Recall@10: 0.0000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 2: 100%|████| 2/2 [00:00<00:00,  4.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2, Training Loss: 1.7051\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 2: 100%|██| 1/1 [00:00<00:00,  5.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 3: 100%|████| 2/2 [00:00<00:00,  4.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3, Training Loss: 1.6361\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 3: 100%|██| 1/1 [00:00<00:00,  5.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 4: 100%|████| 2/2 [00:00<00:00,  4.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4, Training Loss: 1.5921\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 4: 100%|██| 1/1 [00:00<00:00,  5.31it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 5: 100%|████| 2/2 [00:00<00:00,  4.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5, Training Loss: 1.5342\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 5: 100%|██| 1/1 [00:00<00:00,  4.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 6: 100%|████| 2/2 [00:00<00:00,  4.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6, Training Loss: 1.5079\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 6: 100%|██| 1/1 [00:00<00:00,  4.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 7: 100%|████| 2/2 [00:00<00:00,  4.71it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7, Training Loss: 1.4415\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 7: 100%|██| 1/1 [00:00<00:00,  5.12it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 8: 100%|████| 2/2 [00:00<00:00,  5.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8, Training Loss: 1.3993\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 8: 100%|██| 1/1 [00:00<00:00,  5.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8, Recall@10: 0.0336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 9: 100%|████| 2/2 [00:00<00:00,  4.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9, Training Loss: 1.3861\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 9: 100%|██| 1/1 [00:00<00:00,  4.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 10: 100%|███| 2/2 [00:00<00:00,  4.47it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10, Training Loss: 1.3337\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 10: 100%|█| 1/1 [00:00<00:00,  5.31it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 11: 100%|███| 2/2 [00:00<00:00,  4.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11, Training Loss: 1.3176\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 11: 100%|█| 1/1 [00:00<00:00,  5.28it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 12: 100%|███| 2/2 [00:00<00:00,  4.60it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12, Training Loss: 1.2561\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 12: 100%|█| 1/1 [00:00<00:00,  4.97it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 13: 100%|███| 2/2 [00:00<00:00,  4.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13, Training Loss: 1.2648\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 13: 100%|█| 1/1 [00:00<00:00,  5.54it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 14: 100%|███| 2/2 [00:00<00:00,  4.72it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14, Training Loss: 1.2043\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 14: 100%|█| 1/1 [00:00<00:00,  5.09it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 15: 100%|███| 2/2 [00:00<00:00,  4.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15, Training Loss: 1.1588\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 15: 100%|█| 1/1 [00:00<00:00,  4.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 16: 100%|███| 2/2 [00:00<00:00,  4.38it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16, Training Loss: 1.1383\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 16: 100%|█| 1/1 [00:00<00:00,  5.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 17: 100%|███| 2/2 [00:00<00:00,  4.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17, Training Loss: 1.0979\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 17: 100%|█| 1/1 [00:00<00:00,  4.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 18: 100%|███| 2/2 [00:00<00:00,  4.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18, Training Loss: 1.1085\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 18: 100%|█| 1/1 [00:00<00:00,  5.23it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 19: 100%|███| 2/2 [00:00<00:00,  4.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19, Training Loss: 1.0590\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 19: 100%|█| 1/1 [00:00<00:00,  5.77it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 20: 100%|███| 2/2 [00:00<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20, Training Loss: 1.0415\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 20: 100%|█| 1/1 [00:00<00:00,  5.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 21: 100%|███| 2/2 [00:00<00:00,  4.97it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21, Training Loss: 1.0456\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 21: 100%|█| 1/1 [00:00<00:00,  4.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 22: 100%|███| 2/2 [00:00<00:00,  4.88it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22, Training Loss: 0.9712\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 22: 100%|█| 1/1 [00:00<00:00,  5.73it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 23: 100%|███| 2/2 [00:00<00:00,  5.06it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23, Training Loss: 0.9945\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 23: 100%|█| 1/1 [00:00<00:00,  5.39it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 24: 100%|███| 2/2 [00:00<00:00,  5.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24, Training Loss: 0.9243\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 24: 100%|█| 1/1 [00:00<00:00,  5.17it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 25: 100%|███| 2/2 [00:00<00:00,  5.13it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 25, Training Loss: 0.9414\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 25: 100%|█| 1/1 [00:00<00:00,  5.37it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 25, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 26: 100%|███| 2/2 [00:00<00:00,  5.00it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 26, Training Loss: 0.9009\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 26: 100%|█| 1/1 [00:00<00:00,  5.31it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 26, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 27: 100%|███| 2/2 [00:00<00:00,  5.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 27, Training Loss: 0.8706\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 27: 100%|█| 1/1 [00:00<00:00,  5.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 27, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 28: 100%|███| 2/2 [00:00<00:00,  5.09it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 28, Training Loss: 0.8661\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 28: 100%|█| 1/1 [00:00<00:00,  5.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 28, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 29: 100%|███| 2/2 [00:00<00:00,  5.11it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29, Training Loss: 0.8451\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 29: 100%|█| 1/1 [00:00<00:00,  5.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 30: 100%|███| 2/2 [00:00<00:00,  5.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 30, Training Loss: 0.8496\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 30: 100%|█| 1/1 [00:00<00:00,  4.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 30, Recall@10: 0.0000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 31: 100%|███| 2/2 [00:00<00:00,  5.00it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 31, Training Loss: 0.8083\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 31: 100%|█| 1/1 [00:00<00:00,  5.23it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 31, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 32: 100%|███| 2/2 [00:00<00:00,  5.16it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 32, Training Loss: 0.7874\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 32: 100%|█| 1/1 [00:00<00:00,  5.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 32, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 33: 100%|███| 2/2 [00:00<00:00,  5.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 33, Training Loss: 0.7491\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 33: 100%|█| 1/1 [00:00<00:00,  5.70it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 33, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 34: 100%|███| 2/2 [00:00<00:00,  5.11it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 34, Training Loss: 0.7943\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 34: 100%|█| 1/1 [00:00<00:00,  5.48it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 34, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 35: 100%|███| 2/2 [00:00<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 35, Training Loss: 0.7251\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 35: 100%|█| 1/1 [00:00<00:00,  5.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 35, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 36: 100%|███| 2/2 [00:00<00:00,  5.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 36, Training Loss: 0.7340\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 36: 100%|█| 1/1 [00:00<00:00,  5.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 36, Recall@10: 0.0084\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 37: 100%|███| 2/2 [00:00<00:00,  4.95it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 37, Training Loss: 0.7199\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 37: 100%|█| 1/1 [00:00<00:00,  5.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 37, Recall@10: 0.0168\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 38: 100%|███| 2/2 [00:00<00:00,  5.08it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 38, Training Loss: 0.7390\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 38: 100%|█| 1/1 [00:00<00:00,  5.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 38, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 39: 100%|███| 2/2 [00:00<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 39, Training Loss: 0.7058\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 39: 100%|█| 1/1 [00:00<00:00,  5.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 39, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 40: 100%|███| 2/2 [00:00<00:00,  5.27it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 40, Training Loss: 0.7746\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 40: 100%|█| 1/1 [00:00<00:00,  4.80it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 40, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 41: 100%|███| 2/2 [00:00<00:00,  5.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 41, Training Loss: 0.6722\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 41: 100%|█| 1/1 [00:00<00:00,  5.02it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 41, Recall@10: 0.0336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 42: 100%|███| 2/2 [00:00<00:00,  5.08it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 42, Training Loss: 0.6745\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 42: 100%|█| 1/1 [00:00<00:00,  5.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 42, Recall@10: 0.0420\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 43: 100%|███| 2/2 [00:00<00:00,  4.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 43, Training Loss: 0.7196\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 43: 100%|█| 1/1 [00:00<00:00,  5.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 43, Recall@10: 0.0336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 44: 100%|███| 2/2 [00:00<00:00,  5.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 44, Training Loss: 0.6674\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 44: 100%|█| 1/1 [00:00<00:00,  4.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 44, Recall@10: 0.0420\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 45: 100%|███| 2/2 [00:00<00:00,  5.12it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 45, Training Loss: 0.6826\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 45: 100%|█| 1/1 [00:00<00:00,  5.39it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 45, Recall@10: 0.0420\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 46: 100%|███| 2/2 [00:00<00:00,  4.99it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 46, Training Loss: 0.6669\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 46: 100%|█| 1/1 [00:00<00:00,  5.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 46, Recall@10: 0.0336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 47: 100%|███| 2/2 [00:00<00:00,  5.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 47, Training Loss: 0.6421\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 47: 100%|█| 1/1 [00:00<00:00,  5.25it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 47, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 48: 100%|███| 2/2 [00:00<00:00,  5.24it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 48, Training Loss: 0.6284\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 48: 100%|█| 1/1 [00:00<00:00,  5.67it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 48, Recall@10: 0.0336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 49: 100%|███| 2/2 [00:00<00:00,  5.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 49, Training Loss: 0.5998\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 49: 100%|█| 1/1 [00:00<00:00,  4.85it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 49, Recall@10: 0.0336\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Training Epoch 50: 100%|███| 2/2 [00:00<00:00,  5.16it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 50, Training Loss: 0.6205\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating Epoch 50: 100%|█| 1/1 [00:00<00:00,  5.17it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 50, Recall@10: 0.0252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import torch.nn.functional as F\n",
    "warnings.filterwarnings(\"ignore\", category=UserWarning)\n",
    "embedding_dim = 128 \n",
    "user_feature_dim = user_features.shape[1]\n",
    "business_feature_dim = business_features.shape[1] + 768 + 512\n",
    "print(\"business_features.shape:\", business_features.shape)\n",
    "print(\"business_feature_dim:\", business_feature_dim)\n",
    "model = TwoTowerModel(user_feature_dim, business_feature_dim, embedding_dim)\n",
    "model.to(device)\n",
    "optimizer = torch.optim.AdamW(model.parameters(), lr=1e-4, weight_decay=1e-5)\n",
    "scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=2, gamma=0.3)\n",
    "num_epochs = 50\n",
    "train_losses = []\n",
    "val_roc_aucs = []\n",
    "results = []\n",
    "for epoch in range(num_epochs):\n",
    "    avg_loss = train_model(model, train_loader, optimizer, epoch)\n",
    "    roc_auc = evaluate_model(model, val_loader, epoch)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "75dbd6fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "sample_size = 20\n",
    "sample_data = data.sample(n=sample_size, random_state=42).reset_index(drop=True)\n",
    "sample_embeddings = np.stack(sample_data['text_embedding'].values)\n",
    "\n",
    "# Compute cosine similarity matrix\n",
    "similarity_matrix = cosine_similarity(sample_embeddings)\n",
    "\n",
    "# Plot the heatmap\n",
    "plt.figure(figsize=(12, 10))\n",
    "sns.heatmap(similarity_matrix, annot=False, cmap='Blues')\n",
    "plt.title('USE Transformer Encoder Heatmap')\n",
    "plt.xlabel('Review Index')\n",
    "plt.ylabel('Review Index')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "007f34e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "from PIL import Image\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import matplotlib.pyplot as plt\n",
    "photo_df = pd.read_json('photos.json', lines=True,nrows=5000)\n",
    "labels_to_use = ['food', 'drink', 'menu', 'inside', 'outside']\n",
    "photo_df = photo_df[photo_df['label'].isin(labels_to_use)]\n",
    "label_encoder = LabelEncoder()\n",
    "photo_df['label_encoded'] = label_encoder.fit_transform(photo_df['label'])\n",
    "train_photos, test_photos = train_test_split(photo_df, test_size=0.2, random_state=42, stratify=photo_df['label'])\n",
    "def get_image_embedding(image_path):\n",
    "    try:\n",
    "        image = Image.open(image_path).convert(\"RGB\")\n",
    "        inputs = clip_processor(images=image, return_tensors=\"pt\").to(device)\n",
    "        with torch.no_grad():\n",
    "            outputs = clip_model.get_image_features(**inputs)\n",
    "        embedding = outputs.cpu().numpy()[0]\n",
    "        return embedding\n",
    "    except Exception:\n",
    "        return np.zeros(512)\n",
    "train_embeddings = []\n",
    "for photo_id in train_photos['photo_id']:\n",
    "    image_path = os.path.join('photos', f'{photo_id}.jpg')\n",
    "    embedding = get_image_embedding(image_path)\n",
    "    train_embeddings.append(embedding)\n",
    "train_embeddings = np.array(train_embeddings)\n",
    "test_embeddings = []\n",
    "for photo_id in test_photos['photo_id']:\n",
    "    image_path = os.path.join('photos', f'{photo_id}.jpg')\n",
    "    embedding = get_image_embedding(image_path)\n",
    "    test_embeddings.append(embedding)\n",
    "test_embeddings = np.array(test_embeddings)\n",
    "clf = LogisticRegression(max_iter=1000)\n",
    "clf.fit(train_embeddings, train_photos['label_encoded'])\n",
    "preds = clf.predict(test_embeddings)\n",
    "true_labels = test_photos['label_encoded']\n",
    "cm = confusion_matrix(true_labels, preds, normalize='true')\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=label_encoder.classes_)\n",
    "fig, ax = plt.subplots(figsize=(8, 8))\n",
    "disp.plot(cmap='Blues', ax=ax, values_format='.2f')\n",
    "plt.title('5-way Precision Confusion Matrix')\n",
    "plt.xlabel('Predicted Label')\n",
    "plt.ylabel('True Label')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ae0c0a6b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['inside' 'outside' 'drink' 'food' 'menu']\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\python37\\lib\\site-packages\\ipykernel_launcher.py:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  import sys\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "unique_labels = photo_df['label'].unique()\n",
    "print(unique_labels)\n",
    "home_service_categories = ['inside', 'outside']\n",
    "photo_df_home = photo_df[photo_df['label'].isin(home_service_categories)]\n",
    "\n",
    "label_encoder_home = LabelEncoder()\n",
    "photo_df_home['label_encoded'] = label_encoder_home.fit_transform(photo_df_home['label'])\n",
    "\n",
    "train_photos_home, test_photos_home = train_test_split(photo_df_home, test_size=0.2, random_state=42, stratify=photo_df_home['label'])\n",
    "\n",
    "train_embeddings_home = []\n",
    "for photo_id in train_photos_home['photo_id']:\n",
    "    image_path = os.path.join('photos', f'{photo_id}.jpg')\n",
    "    embedding = get_image_embedding(image_path)\n",
    "    train_embeddings_home.append(embedding)\n",
    "train_embeddings_home = np.array(train_embeddings_home)\n",
    "\n",
    "test_embeddings_home = []\n",
    "for photo_id in test_photos_home['photo_id']:\n",
    "    image_path = os.path.join('photos', f'{photo_id}.jpg')\n",
    "    embedding = get_image_embedding(image_path)\n",
    "    test_embeddings_home.append(embedding)\n",
    "test_embeddings_home = np.array(test_embeddings_home)\n",
    "\n",
    "clf_home = LogisticRegression(max_iter=1000)\n",
    "clf_home.fit(train_embeddings_home, train_photos_home['label_encoded'])\n",
    "\n",
    "preds_home = clf_home.predict(test_embeddings_home)\n",
    "true_labels_home = test_photos_home['label_encoded']\n",
    "\n",
    "cm_home = confusion_matrix(true_labels_home, preds_home, normalize='true')\n",
    "disp_home = ConfusionMatrixDisplay(confusion_matrix=cm_home, display_labels=label_encoder_home.classes_)\n",
    "fig, ax = plt.subplots(figsize=(8, 8))\n",
    "disp_home.plot(cmap='Blues', ax=ax, values_format='.2f')\n",
    "plt.title('Home Services Contractors Precision Confusion Matrix')\n",
    "plt.xlabel('Predicted Label')\n",
    "plt.ylabel('True Label')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ab1a02cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████| 592/592 [00:11<00:00, 50.84it/s]\n"
     ]
    }
   ],
   "source": [
    "from tqdm import tqdm\n",
    "tqdm.pandas()\n",
    "\n",
    "def get_text_embedding(text):\n",
    "    if not isinstance(text, str) or text.strip() == '':\n",
    "        return np.zeros(768)\n",
    "    try:\n",
    "        inputs = tokenizer(text, return_tensors='pt', truncation=True, max_length=512)\n",
    "        inputs = {k: v.to(device) for k, v in inputs.items()}\n",
    "        with torch.no_grad():\n",
    "            outputs = bert_model(**inputs)\n",
    "        embedding = outputs.last_hidden_state[:, 0, :].cpu().numpy()[0]\n",
    "        return embedding\n",
    "    except Exception as e:\n",
    "        print(f\"Error processing text: {text}\")\n",
    "        print(e)\n",
    "        return np.zeros(768)\n",
    "\n",
    "data = data.dropna(subset=['text'])\n",
    "data['text_embedding'] = data['text'].progress_apply(get_text_embedding)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0ed7749d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "\n",
    "sample_data = data.head(100)\n",
    "embeddings = np.stack(sample_data['text_embedding'].values)\n",
    "\n",
    "similarity_matrix = cosine_similarity(embeddings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "2e90dd0e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 35780 (\\N{CJK UNIFIED IDEOGRAPH-8BC4}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 35770 (\\N{CJK UNIFIED IDEOGRAPH-8BBA}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 30456 (\\N{CJK UNIFIED IDEOGRAPH-76F8}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 20284 (\\N{CJK UNIFIED IDEOGRAPH-4F3C}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 24615 (\\N{CJK UNIFIED IDEOGRAPH-6027}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 28909 (\\N{CJK UNIFIED IDEOGRAPH-70ED}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 21147 (\\N{CJK UNIFIED IDEOGRAPH-529B}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 22270 (\\N{CJK UNIFIED IDEOGRAPH-56FE}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 32034 (\\N{CJK UNIFIED IDEOGRAPH-7D22}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n",
      "d:\\python37\\lib\\site-packages\\IPython\\core\\pylabtools.py:151: UserWarning: Glyph 24341 (\\N{CJK UNIFIED IDEOGRAPH-5F15}) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.figure(figsize=(10, 8))\n",
    "sns.heatmap(similarity_matrix, cmap='viridis')\n",
    "plt.title('Yelp 评论相似性热力图')\n",
    "plt.xlabel('评论索引')\n",
    "plt.ylabel('评论索引')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "823e44c5",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.7.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}