03a_Advanced_BPR_Women_5core.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## procedure to load the data from the json format to rating matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from collections import defaultdict\n",
    "import os\n",
    "import struct\n",
    "import numpy as np\n",
    "import random\n",
    "\n",
    "# load data from amazon reviews dataset csv\n",
    "def load_data_hybrid(data_path, min_items=1, min_users=1, sampling= True, sample_size = 0.5):\n",
    "    user_ratings = defaultdict(set)\n",
    "    item_ratings = defaultdict(set)\n",
    "    max_u_id = -1\n",
    "    max_i_id = -1\n",
    "    user_count = 0\n",
    "    item_count = 0\n",
    "    reviews = 0\n",
    "    users = {}  # aid to id LUT\n",
    "    items = {}  # asid to id LUT\n",
    "    brands = {}\n",
    "    prices = {}\n",
    "    prod_desc = {}\n",
    "    random.seed(0)\n",
    "    with open(data_path, 'r') as f:\n",
    "        for line in f.readlines()[1:]:\n",
    "            if (sampling and random.random()>sample_size):\n",
    "                continue\n",
    "            reviews += 1\n",
    "            if (len(line.split(\",\"))==6):\n",
    "                auid, asid, _, brand, price, product_desc = line.split(\",\")\n",
    "            else:\n",
    "                auid, asid, _, brand, price = line.split(\",\")\n",
    "\n",
    "            u, i = None, None\n",
    "\n",
    "            if auid in users:\n",
    "                u = users[auid]\n",
    "            else:\n",
    "                user_count += 1  # new user so increment\n",
    "                users[auid] = user_count\n",
    "                u = user_count\n",
    "\n",
    "            if asid in items:\n",
    "                i = items[asid]\n",
    "            else:\n",
    "                item_count += 1  # new i so increment\n",
    "                items[asid] = item_count\n",
    "                i = item_count\n",
    "                brands[i] = brand\n",
    "                if (price=='' or price=='\\r\\n' or price=='\\n'):\n",
    "                    prices[i] = 0\n",
    "                else:\n",
    "                    prices[i] = float(price.rstrip())\n",
    "                if (len(line.split(\",\")) == 6):\n",
    "                    prod_desc[i] = [int(el) for el in list(product_desc)[:-2][1:]]\n",
    "                    if (len(prod_desc[i])==0):\n",
    "                        prod_desc[i] = list(np.zeros(4525))\n",
    "\n",
    "            user_ratings[u].add(i)\n",
    "            item_ratings[i].add(u)\n",
    "            max_u_id = max(u, max_u_id)\n",
    "            max_i_id = max(i, max_i_id)\n",
    "\n",
    "    print \"max_u_id: \", max_u_id\n",
    "    print \"max_i_id: \", max_i_id\n",
    "    print \"reviews : \", reviews\n",
    "\n",
    "    # filter out users w/ less than X reviews\n",
    "    num_u_id = 0\n",
    "    num_i_id = 0\n",
    "    num_reviews = 0\n",
    "    user_ratings_filtered = defaultdict(set)\n",
    "    for u, ids in user_ratings.iteritems():\n",
    "        if len(ids) > min_items:\n",
    "            # keep\n",
    "            user_ratings_filtered[u] = ids\n",
    "            num_u_id += 1\n",
    "            num_reviews += len(ids)\n",
    "    item_ratings_filtered = defaultdict(set)\n",
    "    for ids, u in item_ratings.iteritems():\n",
    "        if len(u) > min_users:\n",
    "            # keep\n",
    "            item_ratings_filtered[ids] = u\n",
    "            num_i_id += 1\n",
    "\n",
    "\n",
    "    print \"u_id: \", num_u_id\n",
    "    print \"i_id: \", num_i_id\n",
    "    print \"reviews : \", num_reviews\n",
    "    return max_u_id, max_i_id, users, items, user_ratings_filtered, item_ratings_filtered, brands, prices, prod_desc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Procedure to load image features of the products"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#load image features for the given asin collection into dictionary\n",
    "def load_image_features(path, items):\n",
    "    count=0\n",
    "    image_features = {}\n",
    "    f = open(path, 'rb')\n",
    "    while True:\n",
    "        asin = f.read(10)\n",
    "        if asin == '': break\n",
    "        features_bytes = f.read(16384) # 4 * 4096 = 16KB, fast read, don't unpack\n",
    "  \n",
    "        if asin in items: #only unpack 4096 bytes if w need it -- big speed up\n",
    "            features = (np.fromstring(features_bytes, dtype=np.float32)/58.388599)\n",
    "            iid=items[asin]\n",
    "            if len(features)==0:\n",
    "                image_features[iid] = np.zeros(4096)\n",
    "            else:\n",
    "                image_features[iid] = features\n",
    "    \n",
    "    return image_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Procedure to uniformly sample the (u,i,j) triplet from the raw data for the train set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import os\n",
    "import cPickle as pickle\n",
    "import numpy\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "def uniform_sample_batch(train_ratings, test_ratings, item_count, advanced_features):\n",
    "    neg_items = 2\n",
    "    for u in train_ratings.keys():\n",
    "        t = []\n",
    "        iv = []\n",
    "        jv = []\n",
    "        for i in train_ratings[u]:\n",
    "            if (u in test_ratings.keys()):\n",
    "                if (i != test_ratings[u]):  # make sure it's not in the test set\n",
    "                    for k in range(1,neg_items):\n",
    "                        j = random.randint(1, item_count)\n",
    "                        while j in train_ratings[u]:\n",
    "                            j = random.randint(1, item_count)\n",
    "                        # sometimes there will not be an image for given product\n",
    "                        try:\n",
    "                            advanced_features[i]\n",
    "                            advanced_features[j]\n",
    "                        except KeyError:\n",
    "                            continue\n",
    "                        iv.append(advanced_features[i])\n",
    "                        jv.append(advanced_features[j])\n",
    "                        t.append([u, i, j])\n",
    "            else:\n",
    "                for k in range(1,neg_items):\n",
    "                    j = random.randint(1, item_count)\n",
    "                    while j in train_ratings[u]:\n",
    "                        j = random.randint(1, item_count)\n",
    "                    # sometimes there will not be an image for given product\n",
    "                    try:\n",
    "                        advanced_features[i]\n",
    "                        advanced_features[j]\n",
    "                    except KeyError:\n",
    "                        continue\n",
    "                    iv.append(advanced_features[i])\n",
    "                    jv.append(advanced_features[j])\n",
    "                    t.append([u, i, j])\n",
    "\n",
    "        # block if queue is full\n",
    "        if len(iv)>1:\n",
    "            yield numpy.asarray(t), numpy.vstack(tuple(iv)), numpy.vstack(tuple(jv))\n",
    "        else:\n",
    "            continue"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Procedure to uniformly sample the (u,i,j) triplet from the raw data for the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def test_batch_generator_by_user(train_ratings, test_ratings, item_ratings, item_count, advanced_features, cold_start = False, cold_start_thresh = 5):\n",
    "    # using leave one cv\n",
    "    for u in random.sample(test_ratings.keys(), 4000):\n",
    "    #for u in test_ratings.keys():\n",
    "        i = test_ratings[u]\n",
    "        if (cold_start and len(item_ratings[i]) > cold_start_thresh-1):\n",
    "            continue\n",
    "        t = []\n",
    "        ilist = []\n",
    "        jlist = []\n",
    "        count = 0\n",
    "        for j in random.sample(range(item_count), 100):\n",
    "            # find item not in test[u] and train[u]\n",
    "            if j != test_ratings[u] and not (j in train_ratings[u]):\n",
    "                try:\n",
    "                    advanced_features[i]\n",
    "                    advanced_features[j]\n",
    "                except KeyError:\n",
    "                    continue\n",
    "\n",
    "                count += 1\n",
    "                t.append([u, i, j])\n",
    "                ilist.append(advanced_features[i])\n",
    "                jlist.append(advanced_features[j])\n",
    "\n",
    "        # print numpy.asarray(t).shape\n",
    "        # print numpy.vstack(tuple(ilist)).shape\n",
    "        # print numpy.vstack(tuple(jlist)).shape\n",
    "        if (len(ilist) == 0):\n",
    "            #print \"could not find neg item for user, count: \", count, u\n",
    "            continue\n",
    "        yield numpy.asarray(t), numpy.vstack(tuple(ilist)), numpy.vstack(tuple(jlist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Procedure to generate the (u,i) pair for the test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def generate_test(user_ratings):\n",
    "    '''\n",
    "    for each user, random select one rating into test set\n",
    "    '''\n",
    "    user_test = dict()\n",
    "    for u, i_list in user_ratings.items():\n",
    "        user_test[u] = random.sample(user_ratings[u], 1)[0]\n",
    "    return user_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bayesian Personalized Ranking (BPR) Model and it's extensions (VBPR, NVBPR, Hybrid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def abpr(user_count, item_count, advanced_features, bpr_extension= True, hidden_dim=10, hidden_img_dim=10,\n",
    "         l2_regulization=0.1,\n",
    "         bias_regulization=0.01,\n",
    "         embed_regulization = 0,\n",
    "         image_regulization =0.1,\n",
    "         visual_bias_regulization = 0.01):\n",
    "    \"\"\"\n",
    "    user_count: total number of users\n",
    "    item_count: total number of items\n",
    "    hidden_dim: hidden feature size of MF\n",
    "    hidden_img_dim: hidden visual/non-visual feature size of MF\n",
    "    \"\"\"\n",
    "    advanced_feat_dim = len(advanced_features[1])\n",
    "    u = tf.placeholder(tf.int32, [None])\n",
    "    i = tf.placeholder(tf.int32, [None])\n",
    "    j = tf.placeholder(tf.int32, [None])\n",
    "    iv = tf.placeholder(tf.float32, [None, advanced_feat_dim])\n",
    "    jv = tf.placeholder(tf.float32, [None, advanced_feat_dim])\n",
    "\n",
    "    # model parameters -- LEARN THESE\n",
    "    # latent factors\n",
    "    user_emb_w = tf.get_variable(\"user_emb_w\", [user_count + 1, hidden_dim],\n",
    "                                 initializer=tf.random_normal_initializer(0, 0.1))\n",
    "    item_emb_w = tf.get_variable(\"item_emb_w\", [item_count + 1, hidden_dim],\n",
    "                                 initializer=tf.random_normal_initializer(0, 0.1))\n",
    "\n",
    "    # UxD visual factors for users\n",
    "    user_img_w = tf.get_variable(\"user_img_w\", [user_count + 1, hidden_img_dim],\n",
    "                                 initializer=tf.random_normal_initializer(0, 0.1))\n",
    "    # this is E, the embedding matrix\n",
    "    img_emb_w = tf.get_variable(\"img_emb_w\", [hidden_img_dim, advanced_feat_dim],\n",
    "                                initializer=tf.random_normal_initializer(0, 0.1))\n",
    "\n",
    "    visual_bias = tf.get_variable(\"visual_bias\", [1, advanced_feat_dim], initializer=tf.random_normal_initializer(0, 0.1))\n",
    "\n",
    "    # biases\n",
    "    item_b = tf.get_variable(\"item_b\", [item_count + 1, 1], initializer=tf.constant_initializer(0.0))\n",
    "\n",
    "    # pull out the respective latent factor vectors for a given user u and items i & j\n",
    "    u_emb = tf.nn.embedding_lookup(user_emb_w, u)\n",
    "    i_emb = tf.nn.embedding_lookup(item_emb_w, i)\n",
    "    j_emb = tf.nn.embedding_lookup(item_emb_w, j)\n",
    "    # pull out the visual factor, 1 X D for user u\n",
    "    u_img = tf.nn.embedding_lookup(user_img_w, u)\n",
    "    # get the respective biases for items i & j\n",
    "    i_b = tf.nn.embedding_lookup(item_b, i)\n",
    "    j_b = tf.nn.embedding_lookup(item_b, j)\n",
    "\n",
    "\n",
    "    # MF predict: u_i > u_j\n",
    "    # MF predict: u_i > u_j\n",
    "    theta_i = tf.matmul(iv, img_emb_w, transpose_b=True)  # (f_i * E), eq. 3\n",
    "    theta_j = tf.matmul(jv, img_emb_w, transpose_b=True)  # (f_j * E), eq. 3\n",
    "    if bpr_extension:\n",
    "        xui = i_b + tf.reduce_sum(tf.multiply(u_emb, i_emb), 1, keep_dims=True) + tf.reduce_sum(tf.multiply(u_img, theta_i), 1, keep_dims=True) \\\n",
    "                                                                            + tf.reduce_sum(tf.multiply(visual_bias, iv), 1, keep_dims=True) \n",
    "        xuj = j_b + tf.reduce_sum(tf.multiply(u_emb, j_emb), 1, keep_dims=True) + tf.reduce_sum(tf.multiply(u_img, theta_j), 1, keep_dims=True) \\\n",
    "                                                                            + tf.reduce_sum(tf.multiply(visual_bias, jv), 1, keep_dims=True) \n",
    "    else:\n",
    "        xui = i_b + tf.reduce_sum(tf.multiply(u_emb, i_emb), 1, keep_dims=True)\n",
    "        xuj = j_b + tf.reduce_sum(tf.multiply(u_emb, j_emb), 1, keep_dims=True) \n",
    "    \n",
    "    xuij = xui - xuj\n",
    "\n",
    "    auc = tf.reduce_mean(tf.to_float(xuij > 0))\n",
    "\n",
    "    l2_norm = tf.add_n([\n",
    "        l2_regulization * tf.reduce_sum(tf.multiply(u_emb, u_emb)),\n",
    "        image_regulization * tf.reduce_sum(tf.multiply(u_img, u_img)),\n",
    "        l2_regulization * tf.reduce_sum(tf.multiply(i_emb, i_emb)),\n",
    "        l2_regulization * tf.reduce_sum(tf.multiply(j_emb, j_emb)),\n",
    "        embed_regulization * tf.reduce_sum(tf.multiply(img_emb_w, img_emb_w)),\n",
    "        bias_regulization * tf.reduce_sum(tf.multiply(i_b, i_b)),\n",
    "        bias_regulization * tf.reduce_sum(tf.multiply(j_b, j_b)),\n",
    "        visual_bias_regulization * tf.reduce_sum(tf.multiply(visual_bias, visual_bias))\n",
    "    ])\n",
    "\n",
    "    loss = l2_norm - tf.reduce_mean(tf.log(tf.sigmoid(xuij)))\n",
    "    train_op = tf.train.AdamOptimizer().minimize(loss)\n",
    "    return u, i, j, iv, jv, loss, auc, train_op"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Procedure to run the end-to-end model (from loading data to training the model and to test the model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def session_run(num_iter, bpr_extension, advanced_features):\n",
    "    ### Loading and parsing the review matrix for Women 5-core dataset\n",
    "    auc_train = []\n",
    "    auc_test = []\n",
    "    auc_test_cs = []\n",
    "    data_path = os.path.join('', 'reviews_Women_scraped_cpp_fv.csv')\n",
    "    user_count, item_count, users, items, user_ratings, item_ratings, brands, prices, prod_desc = load_data_hybrid(data_path, min_items=4, min_users=0, sampling= True, sample_size = 0.8)\n",
    "    user_ratings_test = generate_test(user_ratings)\n",
    "    with tf.Graph().as_default(), tf.Session() as session:\n",
    "        with tf.variable_scope('abpr'):\n",
    "            u, i, j, iv, jv, loss, auc, train_op = abpr(user_count, item_count, advanced_features, bpr_extension=bpr_extension)\n",
    "\n",
    "        session.run(tf.global_variables_initializer())\n",
    "        \n",
    "\n",
    "        for epoch in range(1, num_iter+1):\n",
    "            print \"epoch \", epoch\n",
    "            _loss_train = 0.0\n",
    "            user_count = 0\n",
    "            auc_train_values = []\n",
    "            for d, _iv, _jv in uniform_sample_batch(user_ratings, user_ratings_test, item_count, advanced_features):\n",
    "                user_count += 1\n",
    "                _loss, _auc, _ = session.run([loss, auc, train_op], feed_dict={u:d[:,0], i:d[:,1], j:d[:,2], iv:_iv, jv:_jv})\n",
    "                _loss_train += _loss\n",
    "                auc_train_values.append(_auc)\n",
    "            print \"train_loss:\", _loss_train/user_count, \"train auc: \", numpy.mean(auc_train_values)\n",
    "            auc_train.append(numpy.mean(auc_train_values))\n",
    "\n",
    "\n",
    "            auc_values = []\n",
    "            _loss_test = 0.0\n",
    "            user_count = 0\n",
    "            for d, _iv, _jv in test_batch_generator_by_user(user_ratings, user_ratings_test, item_ratings, item_count, advanced_features, cold_start = False):\n",
    "                user_count += 1\n",
    "                _loss, _auc = session.run([loss, auc], feed_dict={u: d[:, 0], i: d[:, 1], j: d[:, 2], iv: _iv, jv: _jv})\n",
    "                _loss_test += _loss\n",
    "                auc_values.append(_auc)\n",
    "            print \"test_loss: \", _loss_test / user_count, \"test auc: \", numpy.mean(auc_values)\n",
    "            auc_test.append(numpy.mean(auc_values))\n",
    "\n",
    "            auc_values_cs = []\n",
    "            _loss_test_cs = 0.0\n",
    "            user_count = 0\n",
    "            for d, _iv, _jv in test_batch_generator_by_user(user_ratings, user_ratings_test, item_ratings, item_count, advanced_features, cold_start = True, cold_start_thresh = 10):\n",
    "                user_count += 1\n",
    "                _loss, _auc = session.run([loss, auc], feed_dict={u: d[:, 0], i: d[:, 1], j: d[:, 2], iv: _iv, jv: _jv})\n",
    "                _loss_test_cs += _loss\n",
    "                auc_values_cs.append(_auc)\n",
    "            print \"cold start test_loss: \", _loss_test_cs / user_count, \"cold start auc: \", numpy.mean(auc_values_cs)\n",
    "            auc_test_cs.append(numpy.mean(auc_values_cs))\n",
    "        return auc_train, auc_test, auc_test_cs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the image related feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "images_path = \"image_features_Women.b\"\n",
    "f = load_image_features(images_path, items)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running the base BPR model with no advaced features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.827114112226 train auc:  0.618457\n",
      "test_loss:  2.11200136024 test auc:  0.646193\n",
      "cold start test_loss:  2.18598755597 cold start auc:  0.390812\n",
      "epoch  2\n",
      "train_loss: 0.716221601072 train auc:  0.688985\n",
      "test_loss:  1.57772952868 test auc:  0.6574\n",
      "cold start test_loss:  1.58615180764 cold start auc:  0.36806\n",
      "epoch  3\n",
      "train_loss: 0.664188998259 train auc:  0.704786\n",
      "test_loss:  1.33395079409 test auc:  0.651992\n",
      "cold start test_loss:  1.32647371687 cold start auc:  0.368631\n",
      "epoch  4\n",
      "train_loss: 0.633496807167 train auc:  0.714195\n",
      "test_loss:  1.25485928555 test auc:  0.658646\n",
      "cold start test_loss:  1.22450566984 cold start auc:  0.364785\n",
      "epoch  5\n",
      "train_loss: 0.615919861058 train auc:  0.717817\n",
      "test_loss:  1.26004702549 test auc:  0.662331\n",
      "cold start test_loss:  1.22469079573 cold start auc:  0.362866\n",
      "epoch  6\n",
      "train_loss: 0.606252320834 train auc:  0.72319\n",
      "test_loss:  1.29321172314 test auc:  0.657125\n",
      "cold start test_loss:  1.27929213921 cold start auc:  0.358484\n",
      "epoch  7\n",
      "train_loss: 0.602666742667 train auc:  0.723126\n",
      "test_loss:  1.36192563686 test auc:  0.659341\n",
      "cold start test_loss:  1.346357536 cold start auc:  0.360946\n",
      "epoch  8\n",
      "train_loss: 0.599115792664 train auc:  0.723429\n",
      "test_loss:  1.42134401187 test auc:  0.659316\n",
      "cold start test_loss:  1.43813013574 cold start auc:  0.355352\n",
      "epoch  9\n",
      "train_loss: 0.597863859788 train auc:  0.728371\n",
      "test_loss:  1.48997505234 test auc:  0.663087\n",
      "cold start test_loss:  1.53607236307 cold start auc:  0.350203\n",
      "epoch  10\n",
      "train_loss: 0.598406421818 train auc:  0.725273\n",
      "test_loss:  1.54042461496 test auc:  0.659658\n",
      "cold start test_loss:  1.61081698367 cold start auc:  0.356174\n",
      "epoch  11\n",
      "train_loss: 0.596609440998 train auc:  0.73092\n",
      "test_loss:  1.60474604891 test auc:  0.662175\n",
      "cold start test_loss:  1.69678634664 cold start auc:  0.355776\n",
      "epoch  12\n",
      "train_loss: 0.598198050921 train auc:  0.730326\n",
      "test_loss:  1.66302557128 test auc:  0.66132\n",
      "cold start test_loss:  1.76862638213 cold start auc:  0.357356\n",
      "epoch  13\n",
      "train_loss: 0.600930436176 train auc:  0.727629\n",
      "test_loss:  1.69905156237 test auc:  0.662664\n",
      "cold start test_loss:  1.83099064712 cold start auc:  0.359155\n",
      "epoch  14\n",
      "train_loss: 0.60056156064 train auc:  0.730025\n",
      "test_loss:  1.73755796064 test auc:  0.665658\n",
      "cold start test_loss:  1.88567370765 cold start auc:  0.361105\n",
      "epoch  15\n",
      "train_loss: 0.601023319391 train auc:  0.729983\n",
      "test_loss:  1.77395065736 test auc:  0.659247\n",
      "cold start test_loss:  1.95449255448 cold start auc:  0.356691\n",
      "epoch  16\n",
      "train_loss: 0.602641319598 train auc:  0.729249\n",
      "test_loss:  1.81945084688 test auc:  0.655083\n",
      "cold start test_loss:  2.00556843036 cold start auc:  0.351882\n",
      "epoch  17\n",
      "train_loss: 0.604492906529 train auc:  0.729672\n",
      "test_loss:  1.83358765558 test auc:  0.670065\n",
      "cold start test_loss:  2.05103999844 cold start auc:  0.360027\n",
      "epoch  18\n",
      "train_loss: 0.604183736239 train auc:  0.730849\n",
      "test_loss:  1.87333745032 test auc:  0.654126\n",
      "cold start test_loss:  2.10385432085 cold start auc:  0.349576\n",
      "epoch  19\n",
      "train_loss: 0.604254223052 train auc:  0.733686\n",
      "test_loss:  1.87907987338 test auc:  0.657937\n",
      "cold start test_loss:  2.15261160529 cold start auc:  0.346624\n",
      "epoch  20\n",
      "train_loss: 0.607218024464 train auc:  0.727586\n",
      "test_loss:  1.90447363079 test auc:  0.661402\n",
      "cold start test_loss:  2.16576752375 cold start auc:  0.354738\n"
     ]
    }
   ],
   "source": [
    "bpr_auc_train, bpr_auc_test, bpr_auc_test_cold_start = session_run(20, False, f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running the Advanced BPR model with image features only (VBPR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.823392766467 train auc:  0.608602\n",
      "test_loss:  2.29980899285 test auc:  0.566563\n",
      "cold start test_loss:  2.37684022853 cold start auc:  0.491743\n",
      "epoch  2\n",
      "train_loss: 0.702395898073 train auc:  0.688156\n",
      "test_loss:  1.95161553867 test auc:  0.61495\n",
      "cold start test_loss:  1.97803031039 cold start auc:  0.471283\n",
      "epoch  3\n",
      "train_loss: 0.618500380519 train auc:  0.744446\n",
      "test_loss:  1.88112606954 test auc:  0.646473\n",
      "cold start test_loss:  1.89078979355 cold start auc:  0.527067\n",
      "epoch  4\n",
      "train_loss: 0.54390594113 train auc:  0.7859\n",
      "test_loss:  1.89065556253 test auc:  0.677909\n",
      "cold start test_loss:  1.8938561881 cold start auc:  0.567171\n",
      "epoch  5\n",
      "train_loss: 0.487432890864 train auc:  0.817495\n",
      "test_loss:  1.96346691062 test auc:  0.699325\n",
      "cold start test_loss:  1.98524448772 cold start auc:  0.578255\n",
      "epoch  6\n",
      "train_loss: 0.447236079841 train auc:  0.84015\n",
      "test_loss:  1.99567016024 test auc:  0.700846\n",
      "cold start test_loss:  2.02691205174 cold start auc:  0.60693\n",
      "epoch  7\n",
      "train_loss: 0.411320794238 train auc:  0.856967\n",
      "test_loss:  2.05561378901 test auc:  0.708776\n",
      "cold start test_loss:  2.13188357223 cold start auc:  0.609214\n",
      "epoch  8\n",
      "train_loss: 0.389788630806 train auc:  0.867903\n",
      "test_loss:  2.09679462346 test auc:  0.704416\n",
      "cold start test_loss:  2.1981952109 cold start auc:  0.617988\n",
      "epoch  9\n",
      "train_loss: 0.369384804125 train auc:  0.878254\n",
      "test_loss:  2.11654807446 test auc:  0.714658\n",
      "cold start test_loss:  2.21347098619 cold start auc:  0.624407\n",
      "epoch  10\n",
      "train_loss: 0.353055513304 train auc:  0.885767\n",
      "test_loss:  2.12080226963 test auc:  0.713498\n",
      "cold start test_loss:  2.23677400164 cold start auc:  0.626854\n",
      "epoch  11\n",
      "train_loss: 0.340970547989 train auc:  0.89039\n",
      "test_loss:  2.17054685999 test auc:  0.705165\n",
      "cold start test_loss:  2.25014227106 cold start auc:  0.630787\n",
      "epoch  12\n",
      "train_loss: 0.33297884667 train auc:  0.896113\n",
      "test_loss:  2.1681059998 test auc:  0.709662\n",
      "cold start test_loss:  2.23469701826 cold start auc:  0.64597\n",
      "epoch  13\n",
      "train_loss: 0.325348135859 train auc:  0.897851\n",
      "test_loss:  2.15281752035 test auc:  0.711377\n",
      "cold start test_loss:  2.25666778193 cold start auc:  0.640927\n",
      "epoch  14\n",
      "train_loss: 0.318305466242 train auc:  0.901869\n",
      "test_loss:  2.14950181328 test auc:  0.714112\n",
      "cold start test_loss:  2.26386316083 cold start auc:  0.646777\n",
      "epoch  15\n",
      "train_loss: 0.313838546112 train auc:  0.901911\n",
      "test_loss:  2.18255184332 test auc:  0.701914\n",
      "cold start test_loss:  2.29757865498 cold start auc:  0.636804\n",
      "epoch  16\n",
      "train_loss: 0.304541557395 train auc:  0.905262\n",
      "test_loss:  2.16421047373 test auc:  0.707007\n",
      "cold start test_loss:  2.27675465113 cold start auc:  0.64647\n",
      "epoch  17\n",
      "train_loss: 0.302548850602 train auc:  0.907045\n",
      "test_loss:  2.13662023904 test auc:  0.708978\n",
      "cold start test_loss:  2.24060138834 cold start auc:  0.65442\n",
      "epoch  18\n",
      "train_loss: 0.303237930023 train auc:  0.906521\n",
      "test_loss:  2.1572859713 test auc:  0.706877\n",
      "cold start test_loss:  2.24492520263 cold start auc:  0.658649\n",
      "epoch  19\n",
      "train_loss: 0.297330140791 train auc:  0.909136\n",
      "test_loss:  2.11439812171 test auc:  0.71336\n",
      "cold start test_loss:  2.23600432254 cold start auc:  0.657889\n",
      "epoch  20\n",
      "train_loss: 0.297962028761 train auc:  0.907463\n",
      "test_loss:  2.11694240441 test auc:  0.712117\n",
      "cold start test_loss:  2.23361837074 cold start auc:  0.659215\n"
     ]
    }
   ],
   "source": [
    "vbpr_auc_train, vbpr_auc_test, vbpr_auc_test_cold_start = session_run(20, True, f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating the price related feature and running the advanced BPR model with price features only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.830345229913 train auc:  0.608377\n",
      "test_loss:  2.12580149287 test auc:  0.643243\n",
      "cold start test_loss:  2.21156557555 cold start auc:  0.399544\n",
      "epoch  2\n",
      "train_loss: 0.717594162303 train auc:  0.681105\n",
      "test_loss:  1.63544305508 test auc:  0.657618\n",
      "cold start test_loss:  1.64639584945 cold start auc:  0.381862\n",
      "epoch  3\n",
      "train_loss: 0.658378870899 train auc:  0.711155\n",
      "test_loss:  1.52127971487 test auc:  0.642807\n",
      "cold start test_loss:  1.51514492625 cold start auc:  0.413742\n",
      "epoch  4\n",
      "train_loss: 0.616656980813 train auc:  0.728427\n",
      "test_loss:  1.55501065354 test auc:  0.641158\n",
      "cold start test_loss:  1.54345673445 cold start auc:  0.420039\n",
      "epoch  5\n",
      "train_loss: 0.597163344246 train auc:  0.73653\n",
      "test_loss:  1.61997928062 test auc:  0.64573\n",
      "cold start test_loss:  1.62540994192 cold start auc:  0.40859\n",
      "epoch  6\n",
      "train_loss: 0.585913495053 train auc:  0.744732\n",
      "test_loss:  1.64917850374 test auc:  0.648131\n",
      "cold start test_loss:  1.67162102386 cold start auc:  0.409969\n",
      "epoch  7\n",
      "train_loss: 0.579690981463 train auc:  0.748055\n",
      "test_loss:  1.71836961035 test auc:  0.648072\n",
      "cold start test_loss:  1.73679611758 cold start auc:  0.407746\n",
      "epoch  8\n",
      "train_loss: 0.575104844562 train auc:  0.750977\n",
      "test_loss:  1.76914356723 test auc:  0.64626\n",
      "cold start test_loss:  1.82016517202 cold start auc:  0.405622\n",
      "epoch  9\n",
      "train_loss: 0.572656209996 train auc:  0.754229\n",
      "test_loss:  1.8072924131 test auc:  0.655018\n",
      "cold start test_loss:  1.90114117683 cold start auc:  0.392684\n",
      "epoch  10\n",
      "train_loss: 0.573381424316 train auc:  0.75435\n",
      "test_loss:  1.83544495292 test auc:  0.650087\n",
      "cold start test_loss:  1.94604937738 cold start auc:  0.403784\n",
      "epoch  11\n",
      "train_loss: 0.569798180521 train auc:  0.759928\n",
      "test_loss:  1.89536514664 test auc:  0.650134\n",
      "cold start test_loss:  2.0240577468 cold start auc:  0.392308\n",
      "epoch  12\n",
      "train_loss: 0.569697653612 train auc:  0.757877\n",
      "test_loss:  1.93108763218 test auc:  0.650571\n",
      "cold start test_loss:  2.07342423294 cold start auc:  0.400642\n",
      "epoch  13\n",
      "train_loss: 0.571606236856 train auc:  0.757557\n",
      "test_loss:  1.94926904513 test auc:  0.652171\n",
      "cold start test_loss:  2.10025328122 cold start auc:  0.400401\n",
      "epoch  14\n",
      "train_loss: 0.569480278402 train auc:  0.75905\n",
      "test_loss:  1.98687079751 test auc:  0.655687\n",
      "cold start test_loss:  2.1459411293 cold start auc:  0.397523\n",
      "epoch  15\n",
      "train_loss: 0.571528741226 train auc:  0.758992\n",
      "test_loss:  2.02325835031 test auc:  0.649185\n",
      "cold start test_loss:  2.1962360615 cold start auc:  0.395335\n",
      "epoch  16\n",
      "train_loss: 0.569701199193 train auc:  0.761593\n",
      "test_loss:  2.05001794392 test auc:  0.648475\n",
      "cold start test_loss:  2.24655105547 cold start auc:  0.395047\n",
      "epoch  17\n",
      "train_loss: 0.571258292426 train auc:  0.759958\n",
      "test_loss:  2.03816449432 test auc:  0.657459\n",
      "cold start test_loss:  2.26942960892 cold start auc:  0.401306\n",
      "epoch  18\n",
      "train_loss: 0.570519679792 train auc:  0.759024\n",
      "test_loss:  2.08502409883 test auc:  0.643143\n",
      "cold start test_loss:  2.2982456353 cold start auc:  0.396734\n",
      "epoch  19\n",
      "train_loss: 0.570869692047 train auc:  0.76186\n",
      "test_loss:  2.06177090712 test auc:  0.649449\n",
      "cold start test_loss:  2.34400024085 cold start auc:  0.393494\n",
      "epoch  20\n",
      "train_loss: 0.57330747051 train auc:  0.758968\n",
      "test_loss:  2.07533912574 test auc:  0.651735\n",
      "cold start test_loss:  2.32391580635 cold start auc:  0.395345\n"
     ]
    }
   ],
   "source": [
    "user_count, item_count, users, items, user_ratings, item_ratings, brands, prices, prod_desc = load_data_hybrid(data_path, min_items=4, min_users=0, sampling= True, sample_size = 0.8)\n",
    "prices_features= {}\n",
    "prices_all = list(set(prices.values()))\n",
    "price_quant_level = 10\n",
    "price_max = float(max(prices.values()))\n",
    "for key, value in prices.iteritems():\n",
    "    prices_vec = numpy.zeros(price_quant_level+1)\n",
    "    idx = int(numpy.ceil(float(value)/(price_max/price_quant_level)))\n",
    "    prices_vec[idx]=1\n",
    "    prices_features[key] = prices_vec\n",
    "    \n",
    "pbpr_auc_train, pbpr_auc_test, pbpr_auc_test_cold_start = session_run(20, True, prices_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating the brand related feature and running the advanced BPR model with brand features only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.828945811948 train auc:  0.629426\n",
      "test_loss:  2.16702553107 test auc:  0.613335\n",
      "cold start test_loss:  2.23144034299 cold start auc:  0.43102\n",
      "epoch  2\n",
      "train_loss: 0.715395983616 train auc:  0.700468\n",
      "test_loss:  1.68090840033 test auc:  0.648639\n",
      "cold start test_loss:  1.70759422159 cold start auc:  0.399297\n",
      "epoch  3\n",
      "train_loss: 0.656863541996 train auc:  0.73498\n",
      "test_loss:  1.54445612635 test auc:  0.65352\n",
      "cold start test_loss:  1.54178483286 cold start auc:  0.403256\n",
      "epoch  4\n",
      "train_loss: 0.610738417417 train auc:  0.766682\n",
      "test_loss:  1.58376873837 test auc:  0.676616\n",
      "cold start test_loss:  1.57269929902 cold start auc:  0.431419\n",
      "epoch  5\n",
      "train_loss: 0.564726885922 train auc:  0.799353\n",
      "test_loss:  1.73253963511 test auc:  0.683835\n",
      "cold start test_loss:  1.73032512681 cold start auc:  0.448864\n",
      "epoch  6\n",
      "train_loss: 0.521481174135 train auc:  0.822836\n",
      "test_loss:  1.88323237732 test auc:  0.685511\n",
      "cold start test_loss:  1.90499968752 cold start auc:  0.473586\n",
      "epoch  7\n",
      "train_loss: 0.491212544775 train auc:  0.839435\n",
      "test_loss:  2.03449822268 test auc:  0.688798\n",
      "cold start test_loss:  2.07573410393 cold start auc:  0.497234\n",
      "epoch  8\n",
      "train_loss: 0.464942677008 train auc:  0.852592\n",
      "test_loss:  2.15534925534 test auc:  0.694211\n",
      "cold start test_loss:  2.23771849284 cold start auc:  0.515727\n",
      "epoch  9\n",
      "train_loss: 0.446181460813 train auc:  0.863895\n",
      "test_loss:  2.2576760871 test auc:  0.69387\n",
      "cold start test_loss:  2.36684856942 cold start auc:  0.518312\n",
      "epoch  10\n",
      "train_loss: 0.43054278253 train auc:  0.871341\n",
      "test_loss:  2.3270966894 test auc:  0.695501\n",
      "cold start test_loss:  2.47391250744 cold start auc:  0.524001\n",
      "epoch  11\n",
      "train_loss: 0.417174696962 train auc:  0.878636\n",
      "test_loss:  2.40073475589 test auc:  0.691387\n",
      "cold start test_loss:  2.5295274592 cold start auc:  0.54377\n",
      "epoch  12\n",
      "train_loss: 0.405184369097 train auc:  0.885437\n",
      "test_loss:  2.43101473315 test auc:  0.697231\n",
      "cold start test_loss:  2.58109164706 cold start auc:  0.556785\n",
      "epoch  13\n",
      "train_loss: 0.396454586925 train auc:  0.889159\n",
      "test_loss:  2.462845283 test auc:  0.699197\n",
      "cold start test_loss:  2.6227258568 cold start auc:  0.558881\n",
      "epoch  14\n",
      "train_loss: 0.386724723528 train auc:  0.89385\n",
      "test_loss:  2.47459238118 test auc:  0.701154\n",
      "cold start test_loss:  2.63531027206 cold start auc:  0.568716\n",
      "epoch  15\n",
      "train_loss: 0.376735498207 train auc:  0.898592\n",
      "test_loss:  2.50520505588 test auc:  0.698322\n",
      "cold start test_loss:  2.66820638284 cold start auc:  0.575593\n",
      "epoch  16\n",
      "train_loss: 0.372760095079 train auc:  0.898836\n",
      "test_loss:  2.49831143343 test auc:  0.6964\n",
      "cold start test_loss:  2.6633863175 cold start auc:  0.585478\n",
      "epoch  17\n",
      "train_loss: 0.365917950572 train auc:  0.902062\n",
      "test_loss:  2.49526562895 test auc:  0.698233\n",
      "cold start test_loss:  2.64081807063 cold start auc:  0.590484\n",
      "epoch  18\n",
      "train_loss: 0.359710818752 train auc:  0.904993\n",
      "test_loss:  2.50816456977 test auc:  0.694808\n",
      "cold start test_loss:  2.65236314771 cold start auc:  0.580472\n",
      "epoch  19\n",
      "train_loss: 0.353596403123 train auc:  0.907152\n",
      "test_loss:  2.47133956568 test auc:  0.704992\n",
      "cold start test_loss:  2.68549305538 cold start auc:  0.595654\n",
      "epoch  20\n",
      "train_loss: 0.348513903753 train auc:  0.908284\n",
      "test_loss:  2.46319927384 test auc:  0.706781\n",
      "cold start test_loss:  2.65374010245 cold start auc:  0.597364\n"
     ]
    }
   ],
   "source": [
    "brands_features = {}\n",
    "brands_all = list(set(brands.values()))\n",
    "for key, value in brands.iteritems():\n",
    "    brands_vec = numpy.zeros(len(brands_all))\n",
    "    brands_vec[brands_all.index(value)] = 1\n",
    "    brands_features[key] = brands_vec\n",
    "bbpr_auc_train, bbpr_auc_test, bbpr_auc_test_cold_start = session_run(20, True, brands_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating the product description related feature and running the advanced BPR model with product description generated features only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.83543321795 train auc:  0.620745\n",
      "test_loss:  2.19886917356 test auc:  0.62803\n",
      "cold start test_loss:  2.26692693392 cold start auc:  0.427593\n",
      "epoch  2\n",
      "train_loss: 0.714837757014 train auc:  0.710505\n",
      "test_loss:  1.75880157854 test auc:  0.651866\n",
      "cold start test_loss:  1.76451525411 cold start auc:  0.409093\n",
      "epoch  3\n",
      "train_loss: 0.650783598705 train auc:  0.748742\n",
      "test_loss:  1.65077502303 test auc:  0.660971\n",
      "cold start test_loss:  1.64029363625 cold start auc:  0.417824\n",
      "epoch  4\n",
      "train_loss: 0.597674859173 train auc:  0.7802\n",
      "test_loss:  1.69008145259 test auc:  0.677076\n",
      "cold start test_loss:  1.68740993694 cold start auc:  0.450388\n",
      "epoch  5\n",
      "train_loss: 0.552858695424 train auc:  0.802611\n",
      "test_loss:  1.81496837537 test auc:  0.687145\n",
      "cold start test_loss:  1.80890016799 cold start auc:  0.460495\n",
      "epoch  6\n",
      "train_loss: 0.517357814726 train auc:  0.821524\n",
      "test_loss:  1.92363594067 test auc:  0.686706\n",
      "cold start test_loss:  1.94498795327 cold start auc:  0.473624\n",
      "epoch  7\n",
      "train_loss: 0.493488375345 train auc:  0.830661\n",
      "test_loss:  2.05050629593 test auc:  0.681434\n",
      "cold start test_loss:  2.06590923401 cold start auc:  0.489509\n",
      "epoch  8\n",
      "train_loss: 0.474673238006 train auc:  0.84104\n",
      "test_loss:  2.13476869279 test auc:  0.684973\n",
      "cold start test_loss:  2.20548897286 cold start auc:  0.504376\n",
      "epoch  9\n",
      "train_loss: 0.46148573286 train auc:  0.848541\n",
      "test_loss:  2.20639912207 test auc:  0.688497\n",
      "cold start test_loss:  2.30667321376 cold start auc:  0.505321\n",
      "epoch  10\n",
      "train_loss: 0.450909334261 train auc:  0.853125\n",
      "test_loss:  2.26375875528 test auc:  0.687253\n",
      "cold start test_loss:  2.36835723323 cold start auc:  0.511198\n",
      "epoch  11\n",
      "train_loss: 0.441277835037 train auc:  0.858343\n",
      "test_loss:  2.32953745936 test auc:  0.67957\n",
      "cold start test_loss:  2.42190488233 cold start auc:  0.513776\n",
      "epoch  12\n",
      "train_loss: 0.433344150828 train auc:  0.864638\n",
      "test_loss:  2.34099139412 test auc:  0.686669\n",
      "cold start test_loss:  2.46158760492 cold start auc:  0.530411\n",
      "epoch  13\n",
      "train_loss: 0.426366403449 train auc:  0.866789\n",
      "test_loss:  2.38964398749 test auc:  0.684552\n",
      "cold start test_loss:  2.4911217605 cold start auc:  0.521455\n",
      "epoch  14\n",
      "train_loss: 0.417582885707 train auc:  0.87151\n",
      "test_loss:  2.38122321716 test auc:  0.681454\n",
      "cold start test_loss:  2.53546641983 cold start auc:  0.524765\n",
      "epoch  15\n",
      "train_loss: 0.412877210367 train auc:  0.873171\n",
      "test_loss:  2.40298136735 test auc:  0.676103\n",
      "cold start test_loss:  2.53527147788 cold start auc:  0.532399\n",
      "epoch  16\n",
      "train_loss: 0.408717017443 train auc:  0.876224\n",
      "test_loss:  2.4146478489 test auc:  0.67634\n",
      "cold start test_loss:  2.55045339653 cold start auc:  0.531429\n",
      "epoch  17\n",
      "train_loss: 0.404122171589 train auc:  0.877775\n",
      "test_loss:  2.4276621341 test auc:  0.682986\n",
      "cold start test_loss:  2.56426695399 cold start auc:  0.5344\n",
      "epoch  18\n",
      "train_loss: 0.39976793373 train auc:  0.878837\n",
      "test_loss:  2.44591135737 test auc:  0.673232\n",
      "cold start test_loss:  2.55503479997 cold start auc:  0.542741\n",
      "epoch  19\n",
      "train_loss: 0.395419172558 train auc:  0.881987\n",
      "test_loss:  2.41270125492 test auc:  0.67902\n",
      "cold start test_loss:  2.59491265066 cold start auc:  0.542913\n",
      "epoch  20\n",
      "train_loss: 0.39293829675 train auc:  0.88037\n",
      "test_loss:  2.40064252943 test auc:  0.686352\n",
      "cold start test_loss:  2.5393953618 cold start auc:  0.54706\n"
     ]
    }
   ],
   "source": [
    "d = prod_desc\n",
    "pdbpr_auc_train, pdbpr_auc_test, pdbpr_auc_test_cold_start = session_run(20, True, prod_desc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running the advanced BPR model with the hybrid of all non-visual features:  price+brand+product description features (NVBPR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.83397676466 train auc:  0.622576\n",
      "test_loss:  2.22193963921 test auc:  0.604957\n",
      "cold start test_loss:  2.33043541687 cold start auc:  0.452484\n",
      "epoch  2\n",
      "train_loss: 0.710886764481 train auc:  0.711418\n",
      "test_loss:  1.82348409002 test auc:  0.645431\n",
      "cold start test_loss:  1.85779562236 cold start auc:  0.424151\n",
      "epoch  3\n",
      "train_loss: 0.639795182133 train auc:  0.760532\n",
      "test_loss:  1.74163559884 test auc:  0.652937\n",
      "cold start test_loss:  1.74590533734 cold start auc:  0.449618\n",
      "epoch  4\n",
      "train_loss: 0.576318179312 train auc:  0.798491\n",
      "test_loss:  1.79236164476 test auc:  0.672374\n",
      "cold start test_loss:  1.81105523367 cold start auc:  0.48998\n",
      "epoch  5\n",
      "train_loss: 0.519506688895 train auc:  0.826072\n",
      "test_loss:  1.91901169142 test auc:  0.682017\n",
      "cold start test_loss:  1.93822872405 cold start auc:  0.50498\n",
      "epoch  6\n",
      "train_loss: 0.473635795914 train auc:  0.84911\n",
      "test_loss:  2.01707543823 test auc:  0.688914\n",
      "cold start test_loss:  2.06613169253 cold start auc:  0.529564\n",
      "epoch  7\n",
      "train_loss: 0.440570210668 train auc:  0.862515\n",
      "test_loss:  2.126228939 test auc:  0.692339\n",
      "cold start test_loss:  2.19050437603 cold start auc:  0.549065\n",
      "epoch  8\n",
      "train_loss: 0.416174966337 train auc:  0.874393\n",
      "test_loss:  2.19795107463 test auc:  0.696119\n",
      "cold start test_loss:  2.28513009318 cold start auc:  0.572975\n",
      "epoch  9\n",
      "train_loss: 0.394847160668 train auc:  0.883882\n",
      "test_loss:  2.24508696081 test auc:  0.701196\n",
      "cold start test_loss:  2.37945238778 cold start auc:  0.571661\n",
      "epoch  10\n",
      "train_loss: 0.381329290153 train auc:  0.889326\n",
      "test_loss:  2.28354868215 test auc:  0.697615\n",
      "cold start test_loss:  2.43178129655 cold start auc:  0.572633\n",
      "epoch  11\n",
      "train_loss: 0.366973654825 train auc:  0.89586\n",
      "test_loss:  2.31643079013 test auc:  0.694668\n",
      "cold start test_loss:  2.41005835421 cold start auc:  0.597973\n",
      "epoch  12\n",
      "train_loss: 0.357296112632 train auc:  0.900439\n",
      "test_loss:  2.316346962 test auc:  0.706308\n",
      "cold start test_loss:  2.44823059353 cold start auc:  0.605512\n",
      "epoch  13\n",
      "train_loss: 0.346122655697 train auc:  0.904596\n",
      "test_loss:  2.33133046082 test auc:  0.700032\n",
      "cold start test_loss:  2.45614079563 cold start auc:  0.610245\n",
      "epoch  14\n",
      "train_loss: 0.338229382038 train auc:  0.907193\n",
      "test_loss:  2.30247477514 test auc:  0.707062\n",
      "cold start test_loss:  2.4527010665 cold start auc:  0.615064\n",
      "epoch  15\n",
      "train_loss: 0.333342450651 train auc:  0.908218\n",
      "test_loss:  2.31758799887 test auc:  0.697459\n",
      "cold start test_loss:  2.44449408872 cold start auc:  0.610716\n",
      "epoch  16\n",
      "train_loss: 0.326940171925 train auc:  0.909823\n",
      "test_loss:  2.29276039663 test auc:  0.701391\n",
      "cold start test_loss:  2.4413215001 cold start auc:  0.617111\n",
      "epoch  17\n",
      "train_loss: 0.321469591711 train auc:  0.911395\n",
      "test_loss:  2.28424881859 test auc:  0.706169\n",
      "cold start test_loss:  2.45155977133 cold start auc:  0.616776\n",
      "epoch  18\n",
      "train_loss: 0.317456397493 train auc:  0.912998\n",
      "test_loss:  2.29351247275 test auc:  0.700211\n",
      "cold start test_loss:  2.41766861733 cold start auc:  0.626929\n",
      "epoch  19\n",
      "train_loss: 0.311267913443 train auc:  0.916611\n",
      "test_loss:  2.2671799234 test auc:  0.70685\n",
      "cold start test_loss:  2.42171019048 cold start auc:  0.633143\n",
      "epoch  20\n",
      "train_loss: 0.30828365987 train auc:  0.914922\n",
      "test_loss:  2.2353721038 test auc:  0.710456\n",
      "cold start test_loss:  2.38930951095 cold start auc:  0.632687\n"
     ]
    }
   ],
   "source": [
    "a = prices_features\n",
    "b = brands_features\n",
    "c = dict([(k, numpy.append(a[k],b[k])) for k in set(b) & set(a)])\n",
    "\n",
    "e = dict([(k, numpy.append(c[k],d[k])) for k in set(c) & set(d)])\n",
    "\n",
    "hybrid_non_visual_features = e\n",
    "nvbpr_auc_train, nvbpr_auc_test, nvbpr_auc_test_cold_start = session_run(20, True, hybrid_non_visual_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running the base advanced BPR model with the hybrid of all visual and non-visual features: image+price+brand+product description features (HBPR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_u_id:  34502\n",
      "max_i_id:  14779\n",
      "reviews :  144992\n",
      "u_id:  12373\n",
      "i_id:  14779\n",
      "reviews :  84556\n",
      "epoch  1\n",
      "train_loss: 0.829893482303 train auc:  0.619778\n",
      "test_loss:  2.33956347852 test auc:  0.569678\n",
      "cold start test_loss:  2.43441153348 cold start auc:  0.490776\n",
      "epoch  2\n",
      "train_loss: 0.698156786005 train auc:  0.704174\n",
      "test_loss:  1.99335180886 test auc:  0.612473\n",
      "cold start test_loss:  2.03011705907 cold start auc:  0.475689\n",
      "epoch  3\n",
      "train_loss: 0.610190412864 train auc:  0.763694\n",
      "test_loss:  1.91133403422 test auc:  0.643724\n",
      "cold start test_loss:  1.9417069061 cold start auc:  0.529449\n",
      "epoch  4\n",
      "train_loss: 0.529283218722 train auc:  0.807396\n",
      "test_loss:  1.92709438539 test auc:  0.677772\n",
      "cold start test_loss:  1.95432954565 cold start auc:  0.563544\n",
      "epoch  5\n",
      "train_loss: 0.468332351518 train auc:  0.836723\n",
      "test_loss:  1.98867381583 test auc:  0.698338\n",
      "cold start test_loss:  2.05932136993 cold start auc:  0.577253\n",
      "epoch  6\n",
      "train_loss: 0.423339424614 train auc:  0.858576\n",
      "test_loss:  2.01198713336 test auc:  0.712614\n",
      "cold start test_loss:  2.08656726117 cold start auc:  0.604001\n",
      "epoch  7\n",
      "train_loss: 0.391119925752 train auc:  0.873222\n",
      "test_loss:  2.07521461041 test auc:  0.713274\n",
      "cold start test_loss:  2.17659895713 cold start auc:  0.606775\n",
      "epoch  8\n",
      "train_loss: 0.36448158014 train auc:  0.887179\n",
      "test_loss:  2.10211738438 test auc:  0.713368\n",
      "cold start test_loss:  2.19300658052 cold start auc:  0.634654\n",
      "epoch  9\n",
      "train_loss: 0.346180694226 train auc:  0.894628\n",
      "test_loss:  2.11637612787 test auc:  0.722852\n",
      "cold start test_loss:  2.24212158885 cold start auc:  0.62983\n",
      "epoch  10\n",
      "train_loss: 0.325715505902 train auc:  0.904154\n",
      "test_loss:  2.10604626337 test auc:  0.726264\n",
      "cold start test_loss:  2.21537662459 cold start auc:  0.649031\n",
      "epoch  11\n",
      "train_loss: 0.317042112802 train auc:  0.905842\n",
      "test_loss:  2.13180550333 test auc:  0.718934\n",
      "cold start test_loss:  2.20899537651 cold start auc:  0.65767\n",
      "epoch  12\n",
      "train_loss: 0.306950139149 train auc:  0.910354\n",
      "test_loss:  2.10094269388 test auc:  0.72983\n",
      "cold start test_loss:  2.18587061339 cold start auc:  0.664678\n",
      "epoch  13\n",
      "train_loss: 0.300836113497 train auc:  0.912985\n",
      "test_loss:  2.11897465226 test auc:  0.719622\n",
      "cold start test_loss:  2.2279492625 cold start auc:  0.655649\n",
      "epoch  14\n",
      "train_loss: 0.291611238075 train auc:  0.91656\n",
      "test_loss:  2.07720089883 test auc:  0.728708\n",
      "cold start test_loss:  2.22616332646 cold start auc:  0.656486\n",
      "epoch  15\n",
      "train_loss: 0.285729438744 train auc:  0.918311\n",
      "test_loss:  2.1026226003 test auc:  0.718718\n",
      "cold start test_loss:  2.22058438674 cold start auc:  0.662704\n",
      "epoch  16\n",
      "train_loss: 0.279663205251 train auc:  0.919489\n",
      "test_loss:  2.0900232525 test auc:  0.717654\n",
      "cold start test_loss:  2.21271761963 cold start auc:  0.658881\n",
      "epoch  17\n",
      "train_loss: 0.275443080488 train auc:  0.922171\n",
      "test_loss:  2.06801556712 test auc:  0.722215\n",
      "cold start test_loss:  2.20318794309 cold start auc:  0.663844\n",
      "epoch  18\n",
      "train_loss: 0.274797516407 train auc:  0.920633\n",
      "test_loss:  2.06136039773 test auc:  0.721764\n",
      "cold start test_loss:  2.16748625722 cold start auc:  0.67168\n",
      "epoch  19\n",
      "train_loss: 0.267424475459 train auc:  0.923157\n",
      "test_loss:  2.01533820221 test auc:  0.729359\n",
      "cold start test_loss:  2.17595162087 cold start auc:  0.667085\n",
      "epoch  20\n",
      "train_loss: 0.266154963257 train auc:  0.923375\n",
      "test_loss:  1.99704607107 test auc:  0.734109\n",
      "cold start test_loss:  2.11993383997 cold start auc:  0.679747\n"
     ]
    }
   ],
   "source": [
    "hybrid_all_features = dict([(k, numpy.append(e[k],f[k])) for k in set(e) & set(f)])\n",
    "hbpr_auc_train, hbpr_auc_test, hbpr_auc_test_cold_start = session_run(20, True, hybrid_all_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## comparing Test AUC vs. number of iterations for different models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x117fedf50>"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJ8AAAJQCAYAAAAg48UqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1171cc590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,10))\n",
    "plt.plot(bpr_auc_test, label = \"BPR\")\n",
    "plt.plot(pbpr_auc_test, label = \"Price-BPR\")\n",
    "plt.plot(bbpr_auc_test, label = \"Brand-BPR\")\n",
    "plt.plot(pdbpr_auc_test, label = \"Product Description-BPR\")\n",
    "plt.plot(nvbpr_auc_test, label = \"NonVisual-BPR\")\n",
    "plt.plot(vbpr_auc_test, label = \"Visual-BPR\")\n",
    "plt.plot(hbpr_auc_test, label = \"Hybrid-BPR\")\n",
    "plt.legend()\n",
    "plt.ylabel(\"Test AUC\")\n",
    "plt.xlabel(\"Number of Iterations\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## comparing Cold Start Test AUC vs. number of iterations for different models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1138563d0>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJkAAAJQCAYAAAAt/bVtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRERERERGTSGTiIiIiIiIiIiM2v8HZv5XxsdnTb4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114a18990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,10))\n",
    "plt.plot(bpr_auc_test_cold_start, label = \"BPR\")\n",
    "plt.plot(pbpr_auc_test_cold_start, label = \"Price-BPR\")\n",
    "plt.plot(bbpr_auc_test_cold_start, label = \"Brand-BPR\")\n",
    "plt.plot(pdbpr_auc_test_cold_start, label = \"Product Description-BPR\")\n",
    "plt.plot(nvbpr_auc_test_cold_start, label = \"NonVisual-BPR\")\n",
    "plt.plot(vbpr_auc_test_cold_start, label = \"Visual-BPR\")\n",
    "plt.plot(hbpr_auc_test_cold_start, label = \"Hybrid-BPR\")\n",
    "plt.legend()\n",
    "plt.ylabel(\"Cold Start Test AUC\")\n",
    "plt.xlabel(\"Number of Iterations\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Writing the results to a CSV file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import csv\n",
    "rows = zip(bpr_auc_test, pbpr_auc_test, bbpr_auc_test, pdbpr_auc_test, nvbpr_auc_test, vbpr_auc_test, \n",
    "           hbpr_auc_test, bpr_auc_test_cold_start, pbpr_auc_test_cold_start, bbpr_auc_test_cold_start,\n",
    "          pdbpr_auc_test_cold_start, nvbpr_auc_test_cold_start, vbpr_auc_test_cold_start, hbpr_auc_test_cold_start)\n",
    "\n",
    "with open(\"AUC_5core_Women.csv\", \"wb\") as f:\n",
    "    writer = csv.writer(f)\n",
    "    writer.writerow([\"bpr_auc_test\", \"pbpr_auc_test\", \"bbpr_auc_test\", \"pdbpr_auc_test\", \"nvbpr_auc_test\", \n",
    "                     \"vbpr_auc_test\", \"hbpr_auc_test\", \"bpr_auc_test_cold_start\", \"pbpr_auc_test_cold_start\",\n",
    "                    \"bbpr_auc_test_cold_start\", \"pdbpr_auc_test_cold_start\", \"nvbpr_auc_test_cold_start\",\n",
    "                    \"vbpr_auc_test_cold_start\", \"hbpr_auc_test_cold_start\"])\n",
    "    for row in rows:\n",
    "        writer.writerow(row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}