{
 "cells": [
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SASRec & SSEPT\n",
    "\n",
    "### Sequential Recommendation Using Transformer\n",
    "\n",
    "![image.png](attachment:image.png)\n",
    "\n",
    "This is a class of sequential recommendation that uses Transformer \\[2\\] for encoding the users preference represented in terms of a sequence of items purchased/viewed before. Instead of using CNN (Caser \\[3\\]) or RNN (GRU \\[4\\], SLI-Rec \\[5\\] etc.) the approach relies on Transformer based encoder that generates a new representation of the item sequence. Two variants of this Transformer based approaches are included here, \n",
    "\n",
    "- Self-Attentive Sequential Recommendation (or SASRec [1]) that is based on vanilla Transformer and models only the item sequence and\n",
    "- Stochastic Shared Embedding based Personalized Transformer or SSE-PT [6], that also models the users along with the items. \n",
    "\n",
    "This notebook provides an example of necessary steps to train and test either a SASRec or a SSE-PT model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "System version: 3.11.14 (main, Jan 14 2026, 19:35:32) [Clang 21.1.4 ]\n",
      "PyTorch version: 2.10.0+cu128\n",
      "CUDA available: True\n",
      "CUDA version: 12.8\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import sys\n",
    "import pandas as pd \n",
    "import torch\n",
    "\n",
    "from recommenders.utils.timer import Timer\n",
    "from recommenders.datasets.amazon_reviews import get_review_data\n",
    "from recommenders.datasets.split_utils import filter_k_core\n",
    "from recommenders.models.sasrec.model import SASREC\n",
    "from recommenders.models.sasrec.ssept import SSEPT\n",
    "from recommenders.models.sasrec.sampler import WarpSampler\n",
    "from recommenders.models.sasrec.util import SASRecDataSet\n",
    "from recommenders.utils.notebook_utils import store_metadata\n",
    "\n",
    "print(f\"System version: {sys.version}\")\n",
    "print(f\"PyTorch version: {torch.__version__}\")\n",
    "print(f\"CUDA available: {torch.cuda.is_available()}\")\n",
    "if torch.cuda.is_available():\n",
    "    print(f\"CUDA version: {torch.version.cuda}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Input Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "tags": [
     "parameters"
    ]
   },
   "outputs": [],
   "source": [
    "num_epochs = 5\n",
    "batch_size = 128\n",
    "seed = 100  # Set None for non-deterministic result\n",
    "\n",
    "# data_dir = os.path.join(\"tests\", \"recsys_data\", \"RecSys\", \"SASRec-tf2\", \"data\")\n",
    "data_dir = os.path.join(\"..\", \"..\", \"tests\", \"resources\", \"deeprec\", \"sasrec\")\n",
    "\n",
    "# Amazon Electronics Data\n",
    "dataset = \"reviews_Electronics_5\"\n",
    "\n",
    "lr = 0.001             # learning rate\n",
    "maxlen = 50            # maximum sequence length for each user\n",
    "num_blocks = 2         # number of transformer blocks\n",
    "hidden_units = 100     # number of units in the attention calculation\n",
    "num_heads = 1          # number of attention heads\n",
    "dropout_rate = 0.1     # dropout rate\n",
    "l2_emb = 0.0           # L2 regularization coefficient\n",
    "num_neg_test = 100     # number of negative examples per positive example\n",
    "model_name = \"sasrec\"  # \"sasrec\" or \"ssept\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "reviews_name = dataset + '.json'\n",
    "outfile = dataset + '.txt'\n",
    "\n",
    "reviews_file = os.path.join(data_dir, reviews_name)\n",
    "if not os.path.exists(reviews_file):\n",
    "    reviews_output = get_review_data(reviews_file)\n",
    "else:\n",
    "    reviews_output = os.path.join(data_dir, dataset+\".json_output\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.exists(os.path.join(data_dir, outfile)):\n",
    "    df = pd.read_csv(reviews_output, sep=\"\\t\", names=[\"userID\", \"itemID\", \"time\"])\n",
    "    df = filter_k_core(df, 10)  # filter for users & items with less than 10 interactions\n",
    "    \n",
    "    user_set, item_set = set(df['userID'].unique()), set(df['itemID'].unique())\n",
    "    user_map = dict()\n",
    "    item_map = dict()\n",
    "    for u, user in enumerate(user_set):\n",
    "        user_map[user] = u+1\n",
    "    for i, item in enumerate(item_set):\n",
    "        item_map[item] = i+1\n",
    "    \n",
    "    df[\"userID\"] = df[\"userID\"].apply(lambda x: user_map[x])\n",
    "    df[\"itemID\"] = df[\"itemID\"].apply(lambda x: item_map[x])\n",
    "    df = df.sort_values(by=[\"userID\", \"time\"])\n",
    "    df.drop(columns=[\"time\"], inplace=True)\n",
    "    df.to_csv(os.path.join(data_dir, outfile), sep=\"\\t\", header=False, index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SASRec requires sequence input and sequence target. Targets are for both positive and negative examples. Inputs to the model are \n",
    "\n",
    "* user's item history as input to the transformer\n",
    "* user's item history shifted (by 1) as target to the transformer (positive examples)\n",
    "* a sequence of items that are not equal to the positive examples (negative examples)\n",
    "\n",
    "From each user's history three samples are created. If there are $N_u$ items for user-$u$ then $N_u-2$ items are used in training and the last two items are used for validation and testing, respectively. \n",
    "\n",
    "## Dataset Format \n",
    "\n",
    "- The input files should have the following format:\n",
    "    - each row has user-id and item-id converted into integers (starting from 1)\n",
    "    - the rows are sorted by user-id and time of interaction\n",
    "    - for every user the last item is used for testing and the last but one is used for validation\n",
    "    - for example, for user `30449` the sorted inputs are:\n",
    "        - `30449 2771`\n",
    "        - `30449 61842`\n",
    "        - `30449 60293`\n",
    "        - `30449 30047`\n",
    "        - `30449 63296`\n",
    "        - `30449 22042`\n",
    "        - `30449 6717`\n",
    "        - `30449 75780`\n",
    "      \n",
    "      then the train inputs are \n",
    "        - [`2771`, `61842`, `60293`, `30047`, `63296`] (input sequence)\n",
    "        - [`61842`, `60293`, `30047`, `63296`, `22042`] (target sequence for positive examples)\n",
    "        - [`1001`, `50490`, `33312`, `19294`, `45342`] (sample negative examples)\n",
    "\n",
    "      and the validation inputs are \n",
    "        - [`2771`, `61842`, `60293`, `30047`, `63296`, `22042`] (input sequence)\n",
    "        - [`61842`, `60293`, `30047`, `63296`, `22042`, `6717`] (target sequence for positive examples)\n",
    "        - [`4401`, `60351`, `22176`, `23456`, `45342`, '1193`] (sample negative examples)\n",
    "        \n",
    "      and the test inputs are \n",
    "        - [`2771`, `61842`, `60293`, `30047`, `63296`, `22042`, `6717`] (input sequence)\n",
    "        - [`61842`, `60293`, `30047`, `63296`, `22042`, `6717`, `75780`] (target sequence for positive examples)\n",
    "        - [`4401`, `60351`, `22176`, `23456`, `45342`, '1193`, `54231`] (sample negative examples)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../../tests/resources/deeprec/sasrec/reviews_Electronics_5.txt\n",
      "Split complete:\n",
      "  Total users: 36262\n",
      "  Total items: 35074\n",
      "  Users with train/valid/test: 36262\n",
      "  Users with train only: 0\n",
      "  Valid size per user: 1\n",
      "  Test size per user: 1\n",
      "{'num_users': 36262, 'num_items': 35074, 'users_with_splits': 36262, 'users_train_only': 0, 'valid_size': 1, 'test_size': 1, 'min_interactions': 3}\n",
      "average sequence length: 15.01\n"
     ]
    }
   ],
   "source": [
    "inp_file = os.path.join(data_dir, dataset + \".txt\")\n",
    "print(inp_file)\n",
    "\n",
    "# initiate a dataset class \n",
    "data = SASRecDataSet(filename=inp_file, col_sep=\"\\t\")\n",
    "\n",
    "# create train, validation and test splits using leave-one-out strategy\n",
    "# - valid_size: number of items per user for validation\n",
    "# - test_size: number of items per user for testing  \n",
    "# - min_interactions: minimum interactions for a user to have valid/test splits\n",
    "# - verbose: print split statistics\n",
    "split_stats = data.split(valid_size=1, test_size=1, min_interactions=3, verbose=True)\n",
    "print(split_stats)\n",
    "\n",
    "# some statistics\n",
    "num_steps = int(len(data.user_train) / batch_size)\n",
    "cc = 0.0\n",
    "for u in data.user_train:\n",
    "    cc += len(data.user_train[u])\n",
    "print('average sequence length: %.2f' % (cc / len(data.user_train)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Creation\n",
    "\n",
    "Model parameters are\n",
    "\n",
    "    - number of items\n",
    "    - maximum sequence length of the user interaction history\n",
    "    - number of Transformer blocks\n",
    "    - embedding dimension for item embedding\n",
    "    - dimension of the attention\n",
    "    - number of attention heads\n",
    "    - dropout rate\n",
    "    - dimension of the convolution layers, list\n",
    "    - L_2-regularization coefficient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "if model_name == 'sasrec':\n",
    "    model = SASREC(item_num=data.itemnum,\n",
    "                   seq_max_len=maxlen,\n",
    "                   num_blocks=num_blocks,\n",
    "                   embedding_dim=hidden_units,\n",
    "                   attention_dim=hidden_units,\n",
    "                   attention_num_heads=num_heads,\n",
    "                   dropout_rate=dropout_rate,\n",
    "                   conv_dims = [100, 100],\n",
    "                   l2_reg=l2_emb,\n",
    "                   num_neg_test=num_neg_test\n",
    "    )\n",
    "elif model_name == \"ssept\":\n",
    "    model = SSEPT(item_num=data.itemnum,\n",
    "                  user_num=data.usernum,\n",
    "                  seq_max_len=maxlen,\n",
    "                  num_blocks=num_blocks,\n",
    "                  # embedding_dim=hidden_units,  # optional\n",
    "                  user_embedding_dim=10,\n",
    "                  item_embedding_dim=hidden_units,\n",
    "                  attention_dim=hidden_units,\n",
    "                  attention_num_heads=num_heads,\n",
    "                  dropout_rate=dropout_rate,\n",
    "                  conv_dims = [110, 110],\n",
    "                  l2_reg=l2_emb,\n",
    "                  num_neg_test=num_neg_test\n",
    "    )\n",
    "else:\n",
    "    print(f\"Model-{model_name} not found\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sampler\n",
    "\n",
    "    - the sampler creates negative samples from the training data for each batch\n",
    "    - this is done by looking at the original user interaction history and creating items that are not present at all\n",
    "    - the sampler generates a sequence of negative items of the same length as the original history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "sampler = WarpSampler(data.user_train, data.usernum, data.itemnum, batch_size=batch_size, maxlen=maxlen, n_workers=3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Training\n",
    "\n",
    "    - the loss function is defined over all the negative and positive logits\n",
    "    - a mask has to be applied to indicate the non-zero items present in the output\n",
    "    - we also add the regularization loss here\n",
    "    \n",
    "    - having a train-step signature function can speed up the training process"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1: loss=1.1672\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2: loss=1.0167\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3: loss=0.8861\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4: loss=0.7827\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                      "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5: loss=0.6970, val_NDCG@10=0.3659, val_HR@10=0.5700, time=12.9s\n",
      "Training complete. Total time: 12.9s\n",
      "\n",
      "Training time: 0.22 mins\n",
      "Final training loss: 0.6970\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r"
     ]
    }
   ],
   "source": [
    "# Train the model\n",
    "# - val_epoch: evaluate on validation set every N epochs (0 to disable)\n",
    "# - eval_batch_size: batch size for evaluation (higher = faster)\n",
    "with Timer() as train_time:\n",
    "    history = model.train_model(\n",
    "        data, \n",
    "        sampler, \n",
    "        num_epochs=num_epochs, \n",
    "        batch_size=batch_size, \n",
    "        learning_rate=lr, \n",
    "        val_epoch=5,  # validate every 5 epochs\n",
    "        eval_batch_size=256,\n",
    "        verbose=True\n",
    "    )\n",
    "\n",
    "print(f'\\nTraining time: {train_time.interval/60.0:.2f} mins')\n",
    "print(f'Final training loss: {history[\"loss\"][-1]:.4f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating on test set...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                      "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluation time: 0.47s\n",
      "\n",
      "Test Results:\n",
      "  NDCG@10: 0.342169\n",
      "  HR@10:   0.538600\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r"
     ]
    }
   ],
   "source": [
    "# Evaluate on test set\n",
    "# - seed: for reproducibility when sampling users (if > 10000 users)\n",
    "# - eval_batch_size: batch size for evaluation\n",
    "print(\"Evaluating on test set...\")\n",
    "with Timer() as eval_time:\n",
    "    test_metrics = model.evaluate(data, seed=seed, eval_batch_size=256)\n",
    "\n",
    "print(f\"Evaluation time: {eval_time.interval:.2f}s\")\n",
    "print(f\"\\nTest Results:\")\n",
    "print(f\"  NDCG@10: {test_metrics[0]:.6f}\")\n",
    "print(f\"  HR@10:   {test_metrics[1]:.6f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/notebook_utils.json+json": {
       "data": 0.34216850327502274,
       "encoder": "json",
       "name": "ndcg@10"
      }
     },
     "metadata": {
      "notebook_utils": {
       "data": true,
       "display": false,
       "name": "ndcg@10"
      }
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "application/notebook_utils.json+json": {
       "data": 0.5386,
       "encoder": "json",
       "name": "Hit@10"
      }
     },
     "metadata": {
      "notebook_utils": {
       "data": true,
       "display": false,
       "name": "Hit@10"
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Record results for tests - ignore this cell\n",
    "store_metadata(\"ndcg@10\", test_metrics[0])\n",
    "store_metadata(\"Hit@10\", test_metrics[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reference\n",
    "\\[1\\] Wang-Cheng Kang, Julian McAuley: Self-Attentive Sequential Recommendation, arXiv preprint arXiv:1808.09781 (2018) <br>\n",
    "\n",
    "\\[2\\] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In Advances in Neural Information Processing Systems. 5998–6008 <br>\n",
    "\n",
    "\\[3\\] Jiaxi Tang and Ke Wang. 2018. Personalized top-n sequential recommendation via convolutional sequence embedding. In Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining. ACM, 565–573.\n",
    "\n",
    "\\[4\\] Kyunghyun Cho, Bart van Merrienboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation. arXiv preprint arXiv:1406.1078. 2014.\n",
    "\n",
    "\\[5\\] Zeping Yu, Jianxun Lian, Ahmad Mahmoody, Gongshen Liu, Xing Xie. Adaptive User Modeling with Long and Short-Term Preferences for Personailzed Recommendation. In Proceedings of the 28th International Joint Conferences on Artificial Intelligence, IJCAI’19, Pages 4213-4219. AAAI Press, 2019.\n",
    "\n",
    "\\[6\\] Liwei Wu, Shuqing Li, Cho-Jui Hsieh, James Sharpnack. SSE-PT: Sequential Recommendation Via Personalized Transformer. In Fourteenth ACM Conference on Recommender Systems, RecSys'20:, Pages 328–337, 2020."
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Tags",
  "kernelspec": {
   "display_name": "recommenders",
   "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": "3.11.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}