ScaledHybridSVD_AMZ.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "env: MKL_NUM_THREADS=16\n",
      "env: OMP_NUM_THREADS=16\n"
     ]
    }
   ],
   "source": [
    "%env MKL_NUM_THREADS=16\n",
    "%env OMP_NUM_THREADS=16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from ipypb import track\n",
    "\n",
    "from polara.evaluation import evaluation_engine as ee\n",
    "from polara.evaluation.pipelines import (find_optimal_config,\n",
    "                                         set_config)\n",
    "\n",
    "from hybrids import ScaledHybridSVD, ScaledHybridSVDColdStart\n",
    "from data_preprocessing import (get_amazon_data,\n",
    "                                get_similarity_data,\n",
    "                                prepare_data_model,\n",
    "                                prepare_cold_start_data_model)\n",
    "from utils import (report_results, save_results,\n",
    "                   apply_config, read_config, print_data_stats,\n",
    "                   save_training_time, save_cv_training_time)\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from polara.recommender import defaults\n",
    "defaults.memory_hard_limit = 15 # allowed memory usage during recommendations generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "seed = 42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment_name = 'scaledhybridsvd'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Experiment setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_labels = ['AMZe', 'AMZvg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "ranks_grid = [1, 5, 10, 15, 20, 30, 50, 60, 75, 100, 125, 150, 200, 250, 300,\n",
    "              350, 400, 500, 750, 1000, 1250, 1500, 1750, 2000, 2500, 3000]\n",
    "svd_ranks = {'AMZe': ranks_grid,\n",
    "             'AMZvg': [r for r in ranks_grid if r <= 500]\n",
    "            }\n",
    "\n",
    "hybridsvd_weights = [0.1, 0.3, 0.5, 0.7, 0.9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "topk_values = [1, 3, 10, 20, 30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_metric = 'mrr'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict = dict.fromkeys(data_labels)\n",
    "meta_dict = dict.fromkeys(data_labels)\n",
    "similarities = dict.fromkeys(data_labels)\n",
    "sim_indices = dict.fromkeys(data_labels)\n",
    "feature_idx = dict.fromkeys(data_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_data = [data_dict, similarities, sim_indices, meta_dict]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Amazon Electronics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "lbl = 'AMZe'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict[lbl], meta_dict[lbl] = get_amazon_data('../datasets/amazon/ratings_Electronics.csv',\n",
    "                                                 meta_path='../datasets/amazon/meta/meta_Electronics.json.gz',\n",
    "                                                 implicit=True,\n",
    "                                                 pcore=5,\n",
    "                                                 filter_no_meta=True,\n",
    "                                                 flat_categories=True) # take only bottom level categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "similarities[lbl], sim_indices[lbl], feature_idx[lbl] = get_similarity_data(meta_dict[lbl])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(meta_dict[lbl].applymap(len).sum(axis=1)==0).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Amazon Video Games"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "lbl = 'AMZvg'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict[lbl], meta_dict[lbl] = get_amazon_data('../datasets/amazon/ratings_Video_Games.csv',\n",
    "                                                 meta_path='../datasets/amazon/meta/meta_Video_Games.json.gz',\n",
    "                                                 implicit=True,\n",
    "                                                 pcore=5,\n",
    "                                                 filter_data={'categories': ['Games']}, # filter uniformative category\n",
    "                                                 filter_no_meta=True,\n",
    "                                                 flat_categories=True) # take only bottom level categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "similarities[lbl], sim_indices[lbl], feature_idx[lbl] = get_similarity_data(meta_dict[lbl])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(meta_dict[lbl].applymap(len).sum(axis=1)==0).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AMZe\n",
      "{'userid': 124895, 'asin': 44843}\n",
      "density 0.019153791836615672\n",
      "similarity matrix density 1.1054998336712965\n",
      "AMZvg\n",
      "{'userid': 14251, 'asin': 6858}\n",
      "density 0.13281340440589384\n",
      "similarity matrix density 9.081814734274188\n"
     ]
    }
   ],
   "source": [
    "print_data_stats(data_labels, all_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Standard experiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaling_params = {lbl: read_config('scaledsvd', lbl)['PureSVDs']['col_scaling'] for lbl in data_labels}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_recommender_models(data_label, data_models, config):\n",
    "    data_model = data_models[data_label]\n",
    "    svd = ScaledHybridSVD(data_model)\n",
    "    svd.col_scaling = scaling_params[data_label]\n",
    "    models = [svd]\n",
    "    apply_config(models, config, data_label)\n",
    "    return models\n",
    "\n",
    "\n",
    "def fine_tune_hybridsvd(model, ranks, weights, label, record_time=False):\n",
    "    model.col_scaling = scaling_params[label]\n",
    "    param_grid = [(w, r) for w in weights for r in reversed(list(sorted(ranks)))]\n",
    "    param_names = ('features_weight', 'rank')\n",
    "    best_svd_config, svd_scores = find_optimal_config(model, param_grid, param_names,\n",
    "                                                      target_metric,\n",
    "                                                      return_scores=True,\n",
    "                                                      force_build=False,\n",
    "                                                      iterator=lambda x: track(x, label=label))\n",
    "    model_config = {model.method: dict(zip(param_names, best_svd_config))}\n",
    "    model_scores = {model.method: svd_scores}\n",
    "    try:\n",
    "        if record_time:\n",
    "            max_rank = max(ranks)\n",
    "            save_training_time(f'{experiment_name}_rank_{max_rank}', model, svd_scores.xs(max_rank, level='rank').index, label)\n",
    "    finally:\n",
    "        return model_config, model_scores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = {}\n",
    "scores = {}\n",
    "data_models = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:0; max-width:15ex; vertical-align:middle; text-align:right\"></span>\n",
       "<progress style=\"width:60ex\" max=\"2\" value=\"2\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">2/2</span>\n",
       "<span class=\"Time-label\">[12:21:47<07:23, 22253.36s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 2/2 [12:21:47<07:23, 22253.36s/it]\u001b[B"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZe</span>\n",
       "<progress style=\"width:45ex\" max=\"130\" value=\"130\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">130/130</span>\n",
       "<span class=\"Time-label\">[12:14:11<00:12, 338.85s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "           AMZe [█████████████████████████████████████████████] 130/130 [12:14:11<00:12, 338.85s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZvg</span>\n",
       "<progress style=\"width:45ex\" max=\"90\" value=\"90\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">90/90</span>\n",
       "<span class=\"Time-label\">[07:21<00:01, 4.90s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "          AMZvg [█████████████████████████████████████████████] 90/90 [07:21<00:01, 4.90s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for label in track(data_labels):\n",
    "    data_models[label] = prepare_data_model(label, *all_data, seed)\n",
    "    config[label], scores[label] = fine_tune_hybridsvd(ScaledHybridSVD(data_models[label]),\n",
    "                                                       svd_ranks[label],\n",
    "                                                       hybridsvd_weights,\n",
    "                                                       label, record_time=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "report_results('rank', scores);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'AMZe': {'HybridSVDs': {'features_weight': 0.1, 'rank': 3000}},\n",
       " 'AMZvg': {'HybridSVDs': {'features_weight': 0.3, 'rank': 300}}}"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "config"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### saving data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_results(experiment_name, config=config, tuning=scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:0; max-width:15ex; vertical-align:middle; text-align:right\"></span>\n",
       "<progress style=\"width:60ex\" max=\"2\" value=\"2\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">2/2</span>\n",
       "<span class=\"Time-label\">[10:54:05<03:34, 19622.52s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 2/2 [10:54:05<03:34, 19622.52s/it]\u001b[B"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZe</span>\n",
       "<progress style=\"width:45ex\" max=\"5\" value=\"5\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">5/5</span>\n",
       "<span class=\"Time-label\">[10:50:31<02:01:51, 7806.18s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "           AMZe [█████████████████████████████████████████████] 5/5 [10:50:31<02:01:51, 7806.18s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZvg</span>\n",
       "<progress style=\"width:45ex\" max=\"5\" value=\"5\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">5/5</span>\n",
       "<span class=\"Time-label\">[03:34<00:44, 42.81s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "          AMZvg [█████████████████████████████████████████████] 5/5 [03:34<00:44, 42.81s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result = {}\n",
    "for label in track(data_labels):\n",
    "    models = prepare_recommender_models(label, data_models, config)\n",
    "    result[label] = ee.run_cv_experiment(models,\n",
    "                                         fold_experiment=ee.topk_test,\n",
    "                                         topk_list=topk_values,\n",
    "                                         ignore_feedback=True,\n",
    "                                         iterator=lambda x: track(x, label=label))\n",
    "    save_cv_training_time(experiment_name, models, label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "report_results('topn', result, target_metric);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### saving data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_results(experiment_name, cv=result)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cold start"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "import gc\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaling_params = {lbl: read_config('scaledsvd_coldstart', lbl)['PureSVDs(cs)']['col_scaling'] for lbl in data_labels}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_cold_start_recommender_models(data_label, data_models, config):\n",
    "    data_model = data_models[data_label]\n",
    "    svd = ScaledHybridSVDColdStart(data_model, item_features=meta_dict[data_label])\n",
    "    svd.col_scaling = scaling_params[data_label]\n",
    "    models = [svd]\n",
    "    for model in models:\n",
    "        model.verbose = False\n",
    "        model_config = config[data_label].get(model.method, None)\n",
    "        if model_config:\n",
    "            set_config(model, *list(zip(*model_config.items())))\n",
    "    return models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "config_cold = {}\n",
    "scores_cold = {}\n",
    "data_models_cold = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:0; max-width:15ex; vertical-align:middle; text-align:right\"></span>\n",
       "<progress style=\"width:60ex\" max=\"2\" value=\"2\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">2/2</span>\n",
       "<span class=\"Time-label\">[11:00:56<05:09, 19828.00s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 2/2 [11:00:56<05:09, 19828.00s/it]\u001b[B"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZe</span>\n",
       "<progress style=\"width:45ex\" max=\"130\" value=\"130\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">130/130</span>\n",
       "<span class=\"Time-label\">[10:55:42<00:10, 302.63s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "           AMZe [█████████████████████████████████████████████] 130/130 [10:55:42<00:10, 302.63s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZvg</span>\n",
       "<progress style=\"width:45ex\" max=\"90\" value=\"90\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">90/90</span>\n",
       "<span class=\"Time-label\">[05:08<00:00, 3.43s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "          AMZvg [█████████████████████████████████████████████] 90/90 [05:08<00:00, 3.43s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for label in track(data_labels):\n",
    "    data_models_cold[label] = prepare_cold_start_data_model(label, *all_data, seed)\n",
    "    model = ScaledHybridSVDColdStart(data_models_cold[label],\n",
    "                                     item_features=meta_dict[label])\n",
    "    model.use_raw_features = True\n",
    "    config_cold[label], scores_cold[label] = fine_tune_hybridsvd(model,\n",
    "                                                                 svd_ranks[label],\n",
    "                                                                 hybridsvd_weights,\n",
    "                                                                 label)\n",
    "    del model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "report_results('rank', scores_cold);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'AMZe': {'HybridSVDs(cs)': {'features_weight': 0.9, 'rank': 750}},\n",
       " 'AMZvg': {'HybridSVDs(cs)': {'features_weight': 0.3, 'rank': 500}}}"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "config_cold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### saving data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_results(experiment_name+'_coldstart', config=config_cold, tuning=scores_cold)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:0; max-width:15ex; vertical-align:middle; text-align:right\"></span>\n",
       "<progress style=\"width:60ex\" max=\"2\" value=\"2\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">2/2</span>\n",
       "<span class=\"Time-label\">[01:33:58<06:41, 2818.88s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 2/2 [01:33:58<06:41, 2818.88s/it]\u001b[B"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZe folds</span>\n",
       "<progress style=\"width:45ex\" max=\"5\" value=\"5\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">5/5</span>\n",
       "<span class=\"Time-label\">[01:27:17<20:27, 1047.38s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "     AMZe folds [█████████████████████████████████████████████] 5/5 [01:27:17<20:27, 1047.38s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div><span class=\"Text-label\" style=\"display:inline-block; overflow:hidden; white-space:nowrap; text-overflow:ellipsis; min-width:15ex; max-width:15ex; vertical-align:middle; text-align:right\">AMZvg folds</span>\n",
       "<progress style=\"width:45ex\" max=\"5\" value=\"5\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">5/5</span>\n",
       "<span class=\"Time-label\">[06:41<01:17, 80.16s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "    AMZvg folds [█████████████████████████████████████████████] 5/5 [06:41<01:17, 80.16s/it]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result_cold = {}\n",
    "for label in track(data_labels):\n",
    "    models_cold = prepare_cold_start_recommender_models(label, data_models_cold, config_cold)\n",
    "    result_cold[label] = ee.run_cv_experiment(models_cold,\n",
    "                                              fold_experiment=ee.topk_test,\n",
    "                                              topk_list=topk_values,\n",
    "                                              ignore_feedback=True,\n",
    "                                              iterator=lambda x: track(x, label=f'{label} folds'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "report_results('topn', result_cold, target_metric);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "report_results('topn', result_cold, 'coverage');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### saving data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_results(experiment_name+'_coldstart', cv=result_cold)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}