ScaledHybridSVD.ipynb

{
 "cells": [
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "%env MKL_NUM_THREADS=16\n",
    "%env OMP_NUM_THREADS=16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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_movielens_data,\n",
    "                                get_bookcrossing_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": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from polara.recommender import defaults\n",
    "defaults.memory_hard_limit = 3 # allowed memory usage during recommendations generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "seed = 42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment_name = 'scaledhybridsvd'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Experiment setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_labels = ['ML1M', 'ML10M', 'BX']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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 = {'ML1M': [r for r in ranks_grid if r <= 1000],\n",
    "             'ML10M': [r for r in ranks_grid if r <= 1000],\n",
    "             'BX': [r for r in ranks_grid if r <= 2000]}\n",
    "hybridsvd_weights = [0.1, 0.3, 0.5, 0.7, 0.9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "topk_values = [1, 3, 10, 20, 30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_metric = 'mrr'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_data = [data_dict, similarities, sim_indices, meta_dict]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Movielens1M"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "lbl = 'ML1M'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict[lbl], meta_dict[lbl] = get_movielens_data('/mnt/c/Users/evfro/Downloads/ml-1m.zip',\n",
    "                                                    meta_path='data/meta_info_ml1m.csv',\n",
    "                                                    implicit=True,\n",
    "                                                    filter_no_meta=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "similarities[lbl], sim_indices[lbl], feature_idx[lbl] = get_similarity_data(meta_dict[lbl])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(meta_dict[lbl].applymap(len).sum(axis=1)==0).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Movielens10M"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "lbl = 'ML10M'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict[lbl], meta_dict[lbl] = get_movielens_data('/mnt/c/Users/evfro/Downloads/ml-10m.zip',\n",
    "                                                    meta_path='data/meta_info_ml10m.csv',\n",
    "                                                    implicit=True,\n",
    "                                                    filter_no_meta=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "similarities[lbl], sim_indices[lbl], feature_idx[lbl] = get_similarity_data(meta_dict[lbl])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(meta_dict[lbl].applymap(len).sum(axis=1)==0).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BookCrossing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "lbl = 'BX'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dict[lbl], meta_dict[lbl] = get_bookcrossing_data('/mnt/c/Users/evfro/Downloads/BX-CSV-Dump.zip',\n",
    "                                                       get_books_meta=True,\n",
    "                                                       implicit=True,\n",
    "                                                       pcore=5,\n",
    "                                                       filter_no_meta=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "similarities[lbl], sim_indices[lbl], feature_idx[lbl] = get_similarity_data(meta_dict[lbl])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 22,
     "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": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ML1M\n",
      "{'userid': 6038, 'movieid': 3522}\n",
      "density 2.699052132255699\n",
      "similarity matrix density 51.672185638677114\n",
      "ML10M\n",
      "{'userid': 69797, 'movieid': 10258}\n",
      "density 0.6991397242349022\n",
      "similarity matrix density 51.91219966091452\n",
      "BX\n",
      "{'userid': 7160, 'isbn': 16273}\n",
      "density 0.18925598044812894\n",
      "similarity matrix density 0.4827469267285518\n"
     ]
    }
   ],
   "source": [
    "print_data_stats(data_labels, all_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Standard experiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaling_params = {lbl: read_config('scaledsvd', lbl)['PureSVDs']['col_scaling'] for lbl in data_labels}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ML1M': 0.4, 'ML10M': 0.2, 'BX': 0.4}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaling_params"
   ]
  },
  {
   "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=\"3\" value=\"3\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">3/3</span>\n",
       "<span class=\"Time-label\">[01:44:47<35:54, 2095.63s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 3/3 [01:44:47<35:54, 2095.63s/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\">ML1M</span>\n",
       "<progress style=\"width:45ex\" max=\"100\" value=\"100\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">100/100</span>\n",
       "<span class=\"Time-label\">[05:43<00:00, 3.43s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "           ML1M [█████████████████████████████████████████████] 100/100 [05:43<00:00, 3.43s/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\">ML10M</span>\n",
       "<progress style=\"width:45ex\" max=\"100\" value=\"100\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">100/100</span>\n",
       "<span class=\"Time-label\">[01:02:59<00:05, 37.79s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "          ML10M [█████████████████████████████████████████████] 100/100 [01:02:59<00:05, 37.79s/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\">BX</span>\n",
       "<progress style=\"width:45ex\" max=\"120\" value=\"120\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">120/120</span>\n",
       "<span class=\"Time-label\">[35:53<00:01, 17.94s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "             BX [█████████████████████████████████████████████] 120/120 [35:53<00:01, 17.94s/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": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/winbash/miniconda3/envs/lab/lib/python3.6/site-packages/pandas/plotting/_core.py:1001: UserWarning: Attempted to set non-positive left xlim on a log-scaled axis.\n",
      "Invalid limit will be ignored.\n",
      "  ax.set_xlim(left, right)\n"
     ]
    },
    {
     "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"
    },
    {
     "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": [
       "{'ML1M': {'HybridSVDs': {'features_weight': 0.7, 'rank': 100}},\n",
       " 'ML10M': {'HybridSVDs': {'features_weight': 0.1, 'rank': 1000}},\n",
       " 'BX': {'HybridSVDs': {'features_weight': 0.5, 'rank': 1750}}}"
      ]
     },
     "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=\"3\" value=\"3\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">3/3</span>\n",
       "<span class=\"Time-label\">[01:23:22<27:12, 1667.34s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 3/3 [01:23:22<27:12, 1667.34s/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\">ML1M</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\">[00:51<00:10, 10.11s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "           ML1M [█████████████████████████████████████████████] 5/5 [00:51<00:10, 10.11s/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\">ML10M</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\">[55:19<11:29, 663.81s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "          ML10M [█████████████████████████████████████████████] 5/5 [55:19<11:29, 663.81s/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\">BX</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\">[27:12<05:29, 326.42s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "             BX [█████████████████████████████████████████████] 5/5 [27:12<05:29, 326.42s/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"
    },
    {
     "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": 39,
   "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": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "config_cold = {}\n",
    "scores_cold = {}\n",
    "data_models_cold = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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=\"3\" value=\"3\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">3/3</span>\n",
       "<span class=\"Time-label\">[01:49:10<37:42, 2183.33s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 3/3 [01:49:10<37:42, 2183.33s/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\">ML1M</span>\n",
       "<progress style=\"width:45ex\" max=\"100\" value=\"100\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">100/100</span>\n",
       "<span class=\"Time-label\">[06:59<00:00, 4.19s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "           ML1M [█████████████████████████████████████████████] 100/100 [06:59<00:00, 4.19s/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\">ML10M</span>\n",
       "<progress style=\"width:45ex\" max=\"100\" value=\"100\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">100/100</span>\n",
       "<span class=\"Time-label\">[01:04:23<00:05, 38.63s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "          ML10M [█████████████████████████████████████████████] 100/100 [01:04:23<00:05, 38.63s/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\">BX</span>\n",
       "<progress style=\"width:45ex\" max=\"120\" value=\"120\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">120/120</span>\n",
       "<span class=\"Time-label\">[37:40<00:01, 18.84s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "             BX [█████████████████████████████████████████████] 120/120 [37:40<00:01, 18.84s/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": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/winbash/miniconda3/envs/lab/lib/python3.6/site-packages/pandas/plotting/_core.py:1001: UserWarning: Attempted to set non-positive left xlim on a log-scaled axis.\n",
      "Invalid limit will be ignored.\n",
      "  ax.set_xlim(left, right)\n"
     ]
    },
    {
     "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"
    },
    {
     "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": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ML1M': {'HybridSVDs(cs)': {'features_weight': 0.7, 'rank': 5}},\n",
       " 'ML10M': {'HybridSVDs(cs)': {'features_weight': 0.9, 'rank': 50}},\n",
       " 'BX': {'HybridSVDs(cs)': {'features_weight': 0.9, 'rank': 500}}}"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "config_cold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### saving data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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": 48,
   "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=\"3\" value=\"3\" class=\"Progress-main\"/></progress>\n",
       "<span class=\"Progress-label\"><strong>100%</strong></span>\n",
       "<span class=\"Iteration-label\">3/3</span>\n",
       "<span class=\"Time-label\">[10:24<05:12, 208.11s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[A\u001b[2K\r",
       " [████████████████████████████████████████████████████████████] 3/3 [10:24<05:12, 208.11s/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\">ML1M 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\">[00:16<00:03, 3.11s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "     ML1M folds [█████████████████████████████████████████████] 5/5 [00:16<00:03, 3.11s/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\">ML10M 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\">[04:57<01:09, 59.34s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "    ML10M folds [█████████████████████████████████████████████] 5/5 [04:57<01:09, 59.34s/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\">BX 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\">[05:12<01:03, 62.36s/it]</span></div>"
      ],
      "text/plain": [
       "\u001b[A\u001b[2K\r",
       "       BX folds [█████████████████████████████████████████████] 5/5 [05:12<01:03, 62.36s/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": 49,
   "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"
    },
    {
     "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": 50,
   "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"
    },
    {
     "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": 51,
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}