{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Prepare test set"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"slideshow": {
"slide_type": "-"
}
},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"import scipy.sparse as sparse\n",
"from collections import defaultdict\n",
"from itertools import chain\n",
"import random\n",
"from tqdm import tqdm\n",
"\n",
"# In evaluation we do not load train set - it is not needed\n",
"test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\\t', header=None)\n",
"test.columns=['user', 'item', 'rating', 'timestamp']\n",
"\n",
"test['user_code'] = test['user'].astype(\"category\").cat.codes\n",
"test['item_code'] = test['item'].astype(\"category\").cat.codes\n",
"\n",
"user_code_id = dict(enumerate(test['user'].astype(\"category\").cat.categories))\n",
"user_id_code = dict((v, k) for k, v in user_code_id.items())\n",
"item_code_id = dict(enumerate(test['item'].astype(\"category\").cat.categories))\n",
"item_id_code = dict((v, k) for k, v in item_code_id.items())\n",
"\n",
"test_ui = sparse.csr_matrix((test['rating'], (test['user_code'], test['item_code'])))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Estimations metrics"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"estimations_df=pd.read_csv('Recommendations generated/ml-100k/Ready_Baseline_estimations.csv', header=None)\n",
"estimations_df.columns=['user', 'item' ,'score']\n",
"\n",
"estimations_df['user_code']=[user_id_code[user] for user in estimations_df['user']]\n",
"estimations_df['item_code']=[item_id_code[item] for item in estimations_df['item']]\n",
"estimations=sparse.csr_matrix((estimations_df['score'], (estimations_df['user_code'], estimations_df['item_code'])), shape=test_ui.shape)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"def estimations_metrics(test_ui, estimations):\n",
" result=[]\n",
"\n",
" RMSE=(np.sum((estimations.data-test_ui.data)**2)/estimations.nnz)**(1/2)\n",
" result.append(['RMSE', RMSE])\n",
"\n",
" MAE=np.sum(abs(estimations.data-test_ui.data))/estimations.nnz\n",
" result.append(['MAE', MAE])\n",
" \n",
" df_result=(pd.DataFrame(list(zip(*result))[1])).T\n",
" df_result.columns=list(zip(*result))[0]\n",
" return df_result"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" RMSE | \n",
" MAE | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 0.949459 | \n",
" 0.752487 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" RMSE MAE\n",
"0 0.949459 0.752487"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# in case of error (in the laboratories) you might have to switch to the other version of pandas\n",
"# try !pip3 install pandas=='1.0.3' (or pip if you use python 2) and restart the kernel\n",
"\n",
"estimations_metrics(test_ui, estimations)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Ranking metrics"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[663, 475, 62, ..., 472, 269, 503],\n",
" [ 48, 313, 475, ..., 591, 175, 466],\n",
" [351, 313, 475, ..., 591, 175, 466],\n",
" ...,\n",
" [259, 313, 475, ..., 11, 591, 175],\n",
" [ 33, 313, 475, ..., 11, 591, 175],\n",
" [ 77, 313, 475, ..., 11, 591, 175]])"
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import numpy as np\n",
"reco = np.loadtxt('Recommendations generated/ml-100k/Ready_Baseline_reco.csv', delimiter=',')\n",
"# Let's ignore scores - they are not used in evaluation: \n",
"users=reco[:,:1]\n",
"items=reco[:,1::2]\n",
"# Let's use inner ids instead of real ones\n",
"users=np.vectorize(lambda x: user_id_code.setdefault(x, -1))(users)\n",
"items=np.vectorize(lambda x: item_id_code.setdefault(x, -1))(items) # maybe items we recommend are not in test set\n",
"# Let's put them into one array\n",
"reco=np.concatenate((users, items), axis=1)\n",
"reco"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"def ranking_metrics(test_ui, reco, super_reactions=[], topK=10):\n",
" \n",
" nb_items=test_ui.shape[1]\n",
" relevant_users, super_relevant_users, prec, rec, F_1, F_05, prec_super, rec_super, ndcg, mAP, MRR, LAUC, HR=\\\n",
" 0,0,0,0,0,0,0,0,0,0,0,0,0\n",
" \n",
" cg = (1.0 / np.log2(np.arange(2, topK + 2)))\n",
" cg_sum = np.cumsum(cg)\n",
" \n",
" for (nb_user, user) in tqdm(enumerate(reco[:,0])):\n",
" u_rated_items=test_ui.indices[test_ui.indptr[user]:test_ui.indptr[user+1]]\n",
" nb_u_rated_items=len(u_rated_items)\n",
" if nb_u_rated_items>0: # skip users with no items in test set (still possible that there will be no super items)\n",
" relevant_users+=1\n",
" \n",
" u_super_items=u_rated_items[np.vectorize(lambda x: x in super_reactions)\\\n",
" (test_ui.data[test_ui.indptr[user]:test_ui.indptr[user+1]])]\n",
" # more natural seems u_super_items=[item for item in u_rated_items if test_ui[user,item] in super_reactions]\n",
" # but accesing test_ui[user,item] is expensive -we should avoid doing it\n",
" if len(u_super_items)>0:\n",
" super_relevant_users+=1\n",
" \n",
" user_successes=np.zeros(topK)\n",
" nb_user_successes=0\n",
" user_super_successes=np.zeros(topK)\n",
" nb_user_super_successes=0\n",
" \n",
" # evaluation\n",
" for (item_position,item) in enumerate(reco[nb_user,1:topK+1]):\n",
" if item in u_rated_items:\n",
" user_successes[item_position]=1\n",
" nb_user_successes+=1\n",
" if item in u_super_items:\n",
" user_super_successes[item_position]=1\n",
" nb_user_super_successes+=1\n",
" \n",
" prec_u=nb_user_successes/topK \n",
" prec+=prec_u\n",
" \n",
" rec_u=nb_user_successes/nb_u_rated_items\n",
" rec+=rec_u\n",
" \n",
" F_1+=2*(prec_u*rec_u)/(prec_u+rec_u) if prec_u+rec_u>0 else 0\n",
" F_05+=(0.5**2+1)*(prec_u*rec_u)/(0.5**2*prec_u+rec_u) if prec_u+rec_u>0 else 0\n",
" \n",
" prec_super+=nb_user_super_successes/topK\n",
" rec_super+=nb_user_super_successes/max(len(u_super_items),1) # to set 0 if no super items\n",
" ndcg+=np.dot(user_successes,cg)/cg_sum[min(topK, nb_u_rated_items)-1]\n",
" \n",
" cumsum_successes=np.cumsum(user_successes)\n",
" mAP+=np.dot(cumsum_successes/np.arange(1,topK+1), user_successes)/min(topK, nb_u_rated_items)\n",
" MRR+=1/(user_successes.nonzero()[0][0]+1) if user_successes.nonzero()[0].size>0 else 0\n",
" LAUC+=(np.dot(cumsum_successes, 1-user_successes)+\\\n",
" (nb_user_successes+nb_u_rated_items)/2*((nb_items-nb_u_rated_items)-(topK-nb_user_successes)))/\\\n",
" ((nb_items-nb_u_rated_items)*nb_u_rated_items)\n",
" \n",
" HR+=nb_user_successes>0\n",
" \n",
" \n",
" result=[]\n",
" result.append(('precision', prec/relevant_users))\n",
" result.append(('recall', rec/relevant_users))\n",
" result.append(('F_1', F_1/relevant_users))\n",
" result.append(('F_05', F_05/relevant_users))\n",
" result.append(('precision_super', prec_super/super_relevant_users))\n",
" result.append(('recall_super', rec_super/super_relevant_users))\n",
" result.append(('NDCG', ndcg/relevant_users))\n",
" result.append(('mAP', mAP/relevant_users))\n",
" result.append(('MRR', MRR/relevant_users))\n",
" result.append(('LAUC', LAUC/relevant_users))\n",
" result.append(('HR', HR/relevant_users))\n",
"\n",
" df_result=(pd.DataFrame(list(zip(*result))[1])).T\n",
" df_result.columns=list(zip(*result))[0]\n",
" return df_result"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"943it [00:00, 10596.27it/s]\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" precision | \n",
" recall | \n",
" F_1 | \n",
" F_05 | \n",
" precision_super | \n",
" recall_super | \n",
" NDCG | \n",
" mAP | \n",
" MRR | \n",
" LAUC | \n",
" HR | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 0.09141 | \n",
" 0.037652 | \n",
" 0.04603 | \n",
" 0.061286 | \n",
" 0.079614 | \n",
" 0.056463 | \n",
" 0.095957 | \n",
" 0.043178 | \n",
" 0.198193 | \n",
" 0.515501 | \n",
" 0.437964 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" precision recall F_1 F_05 precision_super recall_super \\\n",
"0 0.09141 0.037652 0.04603 0.061286 0.079614 0.056463 \n",
"\n",
" NDCG mAP MRR LAUC HR \n",
"0 0.095957 0.043178 0.198193 0.515501 0.437964 "
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ranking_metrics(test_ui, reco, super_reactions=[4,5], topK=10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Diversity metrics"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"def diversity_metrics(test_ui, reco, topK=10):\n",
" \n",
" frequencies=defaultdict(int)\n",
" \n",
" # let's assign 0 to all items in test set\n",
" for item in list(set(test_ui.indices)):\n",
" frequencies[item]=0\n",
" \n",
" # counting frequencies\n",
" for item in reco[:,1:].flat:\n",
" frequencies[item]+=1\n",
" \n",
" nb_reco_outside_test=frequencies[-1]\n",
" del frequencies[-1]\n",
" \n",
" frequencies=np.array(list(frequencies.values()))\n",
" \n",
" nb_rec_items=len(frequencies[frequencies>0])\n",
" nb_reco_inside_test=np.sum(frequencies)\n",
" \n",
" frequencies=frequencies/np.sum(frequencies)\n",
" frequencies=np.sort(frequencies)\n",
" \n",
" with np.errstate(divide='ignore'): # let's put zeros put items with 0 frequency and ignore division warning\n",
" log_frequencies=np.nan_to_num(np.log(frequencies), posinf=0, neginf=0)\n",
" \n",
" result=[]\n",
" result.append(('Reco in test', nb_reco_inside_test/(nb_reco_inside_test+nb_reco_outside_test)))\n",
" result.append(('Test coverage', nb_rec_items/test_ui.shape[1]))\n",
" result.append(('Shannon', -np.dot(frequencies, log_frequencies)))\n",
" result.append(('Gini', np.dot(frequencies, np.arange(1-len(frequencies), len(frequencies), 2))/(len(frequencies)-1)))\n",
" \n",
" df_result=(pd.DataFrame(list(zip(*result))[1])).T\n",
" df_result.columns=list(zip(*result))[0]\n",
" return df_result"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Reco in test | \n",
" Test coverage | \n",
" Shannon | \n",
" Gini | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 1.0 | \n",
" 0.033911 | \n",
" 2.836513 | \n",
" 0.991139 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Reco in test Test coverage Shannon Gini\n",
"0 1.0 0.033911 2.836513 0.991139"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# in case of errors try !pip3 install numpy==1.18.4 (or pip if you use python 2) and restart the kernel\n",
"\n",
"import evaluation_measures as ev\n",
"import imp\n",
"imp.reload(ev)\n",
"\n",
"x=diversity_metrics(test_ui, reco, topK=10)\n",
"x"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# To be used in other notebooks"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"943it [00:00, 9831.71it/s]\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" RMSE | \n",
" MAE | \n",
" precision | \n",
" recall | \n",
" F_1 | \n",
" F_05 | \n",
" precision_super | \n",
" recall_super | \n",
" NDCG | \n",
" mAP | \n",
" MRR | \n",
" LAUC | \n",
" HR | \n",
" Reco in test | \n",
" Test coverage | \n",
" Shannon | \n",
" Gini | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 0.949459 | \n",
" 0.752487 | \n",
" 0.09141 | \n",
" 0.037652 | \n",
" 0.04603 | \n",
" 0.061286 | \n",
" 0.079614 | \n",
" 0.056463 | \n",
" 0.095957 | \n",
" 0.043178 | \n",
" 0.198193 | \n",
" 0.515501 | \n",
" 0.437964 | \n",
" 1.0 | \n",
" 0.033911 | \n",
" 2.836513 | \n",
" 0.991139 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" RMSE MAE precision recall F_1 F_05 \\\n",
"0 0.949459 0.752487 0.09141 0.037652 0.04603 0.061286 \n",
"\n",
" precision_super recall_super NDCG mAP MRR LAUC \\\n",
"0 0.079614 0.056463 0.095957 0.043178 0.198193 0.515501 \n",
"\n",
" HR Reco in test Test coverage Shannon Gini \n",
"0 0.437964 1.0 0.033911 2.836513 0.991139 "
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import evaluation_measures as ev\n",
"import imp\n",
"imp.reload(ev)\n",
"\n",
"estimations_df=pd.read_csv('Recommendations generated/ml-100k/Ready_Baseline_estimations.csv', header=None)\n",
"reco=np.loadtxt('Recommendations generated/ml-100k/Ready_Baseline_reco.csv', delimiter=',')\n",
"\n",
"ev.evaluate(test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\\t', header=None),\n",
" estimations_df=estimations_df, \n",
" reco=reco,\n",
" super_reactions=[4,5])\n",
"#also you can just type ev.evaluate_all(estimations_df, reco) - I put above values as default"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"943it [00:00, 11236.19it/s]\n",
"943it [00:00, 10128.11it/s]\n",
"943it [00:00, 10299.38it/s]\n",
"943it [00:00, 10525.19it/s]\n"
]
},
{
"ename": "OSError",
"evalue": "Recommendations generated/ml-100k/Ready_I-KNNWithZScore.csv_reco.csv not found.",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mOSError\u001b[0m Traceback (most recent call last)",
"\u001b[1;32m\u001b[0m in \u001b[0;36m\u001b[1;34m\u001b[0m\n\u001b[0;32m 7\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'./Datasets/ml-100k/test.csv'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msep\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m'\\t'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mheader\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 8\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 9\u001b[1;33m \u001b[0mdf\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mev\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mevaluate_all\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdir_path\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msuper_reactions\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 10\u001b[0m \u001b[1;31m#also you can just type ev.evaluate_all() - I put above values as default\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32m~\\Desktop\\warsztaty\\moje\\evaluation_measures.py\u001b[0m in \u001b[0;36mevaluate_all\u001b[1;34m(test, dir_path, super_reactions, topK)\u001b[0m\n\u001b[0;32m 205\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mmodel\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mmodels\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 206\u001b[0m \u001b[0mestimations_df\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'{}{}_estimations.csv'\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdir_path\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mreplace\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\".csv\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mheader\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 207\u001b[1;33m \u001b[0mreco\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mloadtxt\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'{}{}_reco.csv'\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdir_path\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdelimiter\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m','\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 208\u001b[0m \u001b[0mto_append\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mestimations_df\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreco\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msuper_reactions\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtopK\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 209\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\adrian\\appdata\\local\\programs\\python\\python38-32\\lib\\site-packages\\numpy\\lib\\npyio.py\u001b[0m in \u001b[0;36mloadtxt\u001b[1;34m(fname, dtype, comments, delimiter, converters, skiprows, usecols, unpack, ndmin, encoding, max_rows)\u001b[0m\n\u001b[0;32m 979\u001b[0m \u001b[0mfname\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mos_fspath\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 980\u001b[0m \u001b[1;32mif\u001b[0m \u001b[0m_is_string_like\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 981\u001b[1;33m \u001b[0mfh\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlib\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_datasource\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfname\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'rt'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mencoding\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mencoding\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 982\u001b[0m \u001b[0mfencoding\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfh\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'encoding'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'latin1'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 983\u001b[0m \u001b[0mfh\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0miter\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfh\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\adrian\\appdata\\local\\programs\\python\\python38-32\\lib\\site-packages\\numpy\\lib\\_datasource.py\u001b[0m in \u001b[0;36mopen\u001b[1;34m(path, mode, destpath, encoding, newline)\u001b[0m\n\u001b[0;32m 267\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 268\u001b[0m \u001b[0mds\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mDataSource\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdestpath\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 269\u001b[1;33m \u001b[1;32mreturn\u001b[0m \u001b[0mds\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpath\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mencoding\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mencoding\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnewline\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mnewline\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 270\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 271\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\adrian\\appdata\\local\\programs\\python\\python38-32\\lib\\site-packages\\numpy\\lib\\_datasource.py\u001b[0m in \u001b[0;36mopen\u001b[1;34m(self, path, mode, encoding, newline)\u001b[0m\n\u001b[0;32m 621\u001b[0m encoding=encoding, newline=newline)\n\u001b[0;32m 622\u001b[0m \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 623\u001b[1;33m \u001b[1;32mraise\u001b[0m \u001b[0mIOError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"%s not found.\"\u001b[0m \u001b[1;33m%\u001b[0m \u001b[0mpath\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 624\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 625\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;31mOSError\u001b[0m: Recommendations generated/ml-100k/Ready_I-KNNWithZScore.csv_reco.csv not found."
]
}
],
"source": [
"import evaluation_measures as ev\n",
"import imp\n",
"imp.reload(ev)\n",
"\n",
"dir_path=\"Recommendations generated/ml-100k/\"\n",
"super_reactions=[4,5]\n",
"test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\\t', header=None)\n",
"\n",
"df=ev.evaluate_all(test, dir_path, super_reactions)\n",
"#also you can just type ev.evaluate_all() - I put above values as default"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"df.iloc[:,:9]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"df.iloc[:,np.append(0,np.arange(9, df.shape[1]))]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Check metrics on toy dataset"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import evaluation_measures as ev\n",
"import imp\n",
"import helpers\n",
"imp.reload(ev)\n",
"\n",
"dir_path=\"Recommendations generated/toy-example/\"\n",
"super_reactions=[4,5]\n",
"test=pd.read_csv('./Datasets/toy-example/test.csv', sep='\\t', header=None)\n",
"\n",
"display(ev.evaluate_all(test, dir_path, super_reactions, topK=3))\n",
"#also you can just type ev.evaluate_all() - I put above values as default\n",
"\n",
"toy_train_read=pd.read_csv('./Datasets/toy-example/train.csv', sep='\\t', header=None, names=['user', 'item', 'rating', 'timestamp'])\n",
"toy_test_read=pd.read_csv('./Datasets/toy-example/test.csv', sep='\\t', header=None, names=['user', 'item', 'rating', 'timestamp'])\n",
"reco=pd.read_csv('Recommendations generated/toy-example/Self_BaselineUI_reco.csv', header=None)\n",
"estimations=pd.read_csv('Recommendations generated/toy-example/Self_BaselineUI_estimations.csv', names=['user', 'item', 'est_score'])\n",
"toy_train_ui, toy_test_ui, toy_user_code_id, toy_user_id_code, \\\n",
"toy_item_code_id, toy_item_id_code = helpers.data_to_csr(toy_train_read, toy_test_read)\n",
"\n",
"print('Training data:')\n",
"display(toy_train_ui.todense())\n",
"\n",
"print('Test data:')\n",
"display(toy_test_ui.todense())\n",
"\n",
"print('Recommendations:')\n",
"display(reco)\n",
"\n",
"print('Estimations:')\n",
"display(estimations)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Sample recommendations"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Here is what user rated high:\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" user | \n",
" rating | \n",
" title | \n",
" genres | \n",
"
\n",
" \n",
" \n",
" \n",
" 153 | \n",
" 1 | \n",
" 5 | \n",
" Toy Story (1995) | \n",
" Animation, Children's, Comedy | \n",
"
\n",
" \n",
" 57962 | \n",
" 1 | \n",
" 5 | \n",
" Wrong Trousers, The (1993) | \n",
" Animation, Comedy | \n",
"
\n",
" \n",
" 52845 | \n",
" 1 | \n",
" 5 | \n",
" Antonia's Line (1995) | \n",
" Drama | \n",
"
\n",
" \n",
" 23811 | \n",
" 1 | \n",
" 5 | \n",
" Chasing Amy (1997) | \n",
" Drama, Romance | \n",
"
\n",
" \n",
" 53494 | \n",
" 1 | \n",
" 5 | \n",
" Nightmare Before Christmas, The (1993) | \n",
" Children's, Comedy, Musical | \n",
"
\n",
" \n",
" 23377 | \n",
" 1 | \n",
" 5 | \n",
" Eat Drink Man Woman (1994) | \n",
" Comedy, Drama | \n",
"
\n",
" \n",
" 54249 | \n",
" 1 | \n",
" 5 | \n",
" 12 Angry Men (1957) | \n",
" Drama | \n",
"
\n",
" \n",
" 56869 | \n",
" 1 | \n",
" 5 | \n",
" Henry V (1989) | \n",
" Drama, War | \n",
"
\n",
" \n",
" 22174 | \n",
" 1 | \n",
" 5 | \n",
" Groundhog Day (1993) | \n",
" Comedy, Romance | \n",
"
\n",
" \n",
" 21742 | \n",
" 1 | \n",
" 5 | \n",
" Good, The Bad and The Ugly, The (1966) | \n",
" Action, Western | \n",
"
\n",
" \n",
" 51915 | \n",
" 1 | \n",
" 5 | \n",
" Mighty Aphrodite (1995) | \n",
" Comedy | \n",
"
\n",
" \n",
" 58218 | \n",
" 1 | \n",
" 5 | \n",
" Chasing Amy (1997) | \n",
" Drama, Romance | \n",
"
\n",
" \n",
" 58440 | \n",
" 1 | \n",
" 5 | \n",
" Manon of the Spring (Manon des sources) (1986) | \n",
" Drama | \n",
"
\n",
" \n",
" 20120 | \n",
" 1 | \n",
" 5 | \n",
" Return of the Jedi (1983) | \n",
" Action, Adventure, Romance, Sci-Fi, War | \n",
"
\n",
" \n",
" 19043 | \n",
" 1 | \n",
" 5 | \n",
" Mars Attacks! (1996) | \n",
" Action, Comedy, Sci-Fi, War | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" user rating title \\\n",
"153 1 5 Toy Story (1995) \n",
"57962 1 5 Wrong Trousers, The (1993) \n",
"52845 1 5 Antonia's Line (1995) \n",
"23811 1 5 Chasing Amy (1997) \n",
"53494 1 5 Nightmare Before Christmas, The (1993) \n",
"23377 1 5 Eat Drink Man Woman (1994) \n",
"54249 1 5 12 Angry Men (1957) \n",
"56869 1 5 Henry V (1989) \n",
"22174 1 5 Groundhog Day (1993) \n",
"21742 1 5 Good, The Bad and The Ugly, The (1966) \n",
"51915 1 5 Mighty Aphrodite (1995) \n",
"58218 1 5 Chasing Amy (1997) \n",
"58440 1 5 Manon of the Spring (Manon des sources) (1986) \n",
"20120 1 5 Return of the Jedi (1983) \n",
"19043 1 5 Mars Attacks! (1996) \n",
"\n",
" genres \n",
"153 Animation, Children's, Comedy \n",
"57962 Animation, Comedy \n",
"52845 Drama \n",
"23811 Drama, Romance \n",
"53494 Children's, Comedy, Musical \n",
"23377 Comedy, Drama \n",
"54249 Drama \n",
"56869 Drama, War \n",
"22174 Comedy, Romance \n",
"21742 Action, Western \n",
"51915 Comedy \n",
"58218 Drama, Romance \n",
"58440 Drama \n",
"20120 Action, Adventure, Romance, Sci-Fi, War \n",
"19043 Action, Comedy, Sci-Fi, War "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Here is what we recommend:\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" user | \n",
" rec_nb | \n",
" title | \n",
" genres | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 1.0 | \n",
" 1 | \n",
" Great Day in Harlem, A (1994) | \n",
" Documentary | \n",
"
\n",
" \n",
" 942 | \n",
" 1.0 | \n",
" 2 | \n",
" Tough and Deadly (1995) | \n",
" Action, Drama, Thriller | \n",
"
\n",
" \n",
" 1884 | \n",
" 1.0 | \n",
" 3 | \n",
" Aiqing wansui (1994) | \n",
" Drama | \n",
"
\n",
" \n",
" 2826 | \n",
" 1.0 | \n",
" 4 | \n",
" Delta of Venus (1994) | \n",
" Drama | \n",
"
\n",
" \n",
" 3768 | \n",
" 1.0 | \n",
" 5 | \n",
" Someone Else's America (1995) | \n",
" Drama | \n",
"
\n",
" \n",
" 4710 | \n",
" 1.0 | \n",
" 6 | \n",
" Saint of Fort Washington, The (1993) | \n",
" Drama | \n",
"
\n",
" \n",
" 5652 | \n",
" 1.0 | \n",
" 7 | \n",
" Celestial Clockwork (1994) | \n",
" Comedy | \n",
"
\n",
" \n",
" 6594 | \n",
" 1.0 | \n",
" 8 | \n",
" Some Mother's Son (1996) | \n",
" Drama | \n",
"
\n",
" \n",
" 7535 | \n",
" 1.0 | \n",
" 9 | \n",
" Prefontaine (1997) | \n",
" Drama | \n",
"
\n",
" \n",
" 8477 | \n",
" 1.0 | \n",
" 10 | \n",
" Santa with Muscles (1996) | \n",
" Comedy | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" user rec_nb title \\\n",
"0 1.0 1 Great Day in Harlem, A (1994) \n",
"942 1.0 2 Tough and Deadly (1995) \n",
"1884 1.0 3 Aiqing wansui (1994) \n",
"2826 1.0 4 Delta of Venus (1994) \n",
"3768 1.0 5 Someone Else's America (1995) \n",
"4710 1.0 6 Saint of Fort Washington, The (1993) \n",
"5652 1.0 7 Celestial Clockwork (1994) \n",
"6594 1.0 8 Some Mother's Son (1996) \n",
"7535 1.0 9 Prefontaine (1997) \n",
"8477 1.0 10 Santa with Muscles (1996) \n",
"\n",
" genres \n",
"0 Documentary \n",
"942 Action, Drama, Thriller \n",
"1884 Drama \n",
"2826 Drama \n",
"3768 Drama \n",
"4710 Drama \n",
"5652 Comedy \n",
"6594 Drama \n",
"7535 Drama \n",
"8477 Comedy "
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train=pd.read_csv('./Datasets/ml-100k/train.csv', sep='\\t', header=None, names=['user', 'item', 'rating', 'timestamp'])\n",
"items=pd.read_csv('./Datasets/ml-100k/movies.csv')\n",
"\n",
"user=random.choice(list(set(train['user'])))\n",
"\n",
"train_content=pd.merge(train, items, left_on='item', right_on='id')\n",
"\n",
"print('Here is what user rated high:')\n",
"display(train_content[train_content['user']==user][['user', 'rating', 'title', 'genres']]\\\n",
" .sort_values(by='rating', ascending=False)[:15])\n",
"\n",
"reco = np.loadtxt('Recommendations generated/ml-100k/Self_BaselineUI_reco.csv', delimiter=',')\n",
"items=pd.read_csv('./Datasets/ml-100k/movies.csv')\n",
"\n",
"# Let's ignore scores - they are not used in evaluation: \n",
"reco_users=reco[:,:1]\n",
"reco_items=reco[:,1::2]\n",
"# Let's put them into one array\n",
"reco=np.concatenate((reco_users, reco_items), axis=1)\n",
"\n",
"# Let's rebuild it user-item dataframe\n",
"recommended=[]\n",
"for row in reco:\n",
" for rec_nb, entry in enumerate(row[1:]):\n",
" recommended.append((row[0], rec_nb+1, entry))\n",
"recommended=pd.DataFrame(recommended, columns=['user','rec_nb', 'item'])\n",
"\n",
"recommended_content=pd.merge(recommended, items, left_on='item', right_on='id')\n",
"\n",
"print('Here is what we recommend:')\n",
"recommended_content[recommended_content['user']==user][['user', 'rec_nb', 'title', 'genres']].sort_values(by='rec_nb')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# project task 3: implement some other evaluation measure"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [],
"source": [
"# it may be your idea, modification of what we have already implemented \n",
"# (for example Hit2 rate which would count as a success users whoreceived at least 2 relevant recommendations) \n",
"# or something well-known\n",
"# expected output: modification of evaluation_measures.py such that evaluate_all will also display your measure"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"943it [00:00, 10400.09it/s]\n",
"943it [00:00, 9776.40it/s]\n",
"943it [00:00, 10473.38it/s]\n",
"943it [00:00, 10505.90it/s]\n"
]
},
{
"ename": "OSError",
"evalue": "Recommendations generated/ml-100k/Ready_I-KNNWithZScore.csv_reco.csv not found.",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mOSError\u001b[0m Traceback (most recent call last)",
"\u001b[1;32m\u001b[0m in \u001b[0;36m\u001b[1;34m\u001b[0m\n\u001b[0;32m 3\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'./Datasets/ml-100k/test.csv'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msep\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m'\\t'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mheader\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 4\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 5\u001b[1;33m \u001b[0mev\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mevaluate_all\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdir_path\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msuper_reactions\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[1;32m~\\Desktop\\warsztaty\\moje\\evaluation_measures.py\u001b[0m in \u001b[0;36mevaluate_all\u001b[1;34m(test, dir_path, super_reactions, topK)\u001b[0m\n\u001b[0;32m 205\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mmodel\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mmodels\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 206\u001b[0m \u001b[0mmodel\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mreplace\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\".csv\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 207\u001b[1;33m \u001b[0mestimations_df\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'{}{}_estimations.csv'\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdir_path\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mheader\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 208\u001b[0m \u001b[0mreco\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mloadtxt\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'{}{}_reco.csv'\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdir_path\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdelimiter\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m','\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 209\u001b[0m \u001b[0mto_append\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mestimations_df\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreco\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msuper_reactions\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtopK\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\adrian\\appdata\\local\\programs\\python\\python38-32\\lib\\site-packages\\numpy\\lib\\npyio.py\u001b[0m in \u001b[0;36mloadtxt\u001b[1;34m(fname, dtype, comments, delimiter, converters, skiprows, usecols, unpack, ndmin, encoding, max_rows)\u001b[0m\n\u001b[0;32m 979\u001b[0m \u001b[0mfname\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mos_fspath\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 980\u001b[0m \u001b[1;32mif\u001b[0m \u001b[0m_is_string_like\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 981\u001b[1;33m \u001b[0mfh\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlib\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_datasource\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfname\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'rt'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mencoding\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mencoding\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 982\u001b[0m \u001b[0mfencoding\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfh\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'encoding'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'latin1'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 983\u001b[0m \u001b[0mfh\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0miter\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfh\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\adrian\\appdata\\local\\programs\\python\\python38-32\\lib\\site-packages\\numpy\\lib\\_datasource.py\u001b[0m in \u001b[0;36mopen\u001b[1;34m(path, mode, destpath, encoding, newline)\u001b[0m\n\u001b[0;32m 267\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 268\u001b[0m \u001b[0mds\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mDataSource\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdestpath\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 269\u001b[1;33m \u001b[1;32mreturn\u001b[0m \u001b[0mds\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpath\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mencoding\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mencoding\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnewline\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mnewline\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 270\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 271\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\adrian\\appdata\\local\\programs\\python\\python38-32\\lib\\site-packages\\numpy\\lib\\_datasource.py\u001b[0m in \u001b[0;36mopen\u001b[1;34m(self, path, mode, encoding, newline)\u001b[0m\n\u001b[0;32m 621\u001b[0m encoding=encoding, newline=newline)\n\u001b[0;32m 622\u001b[0m \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 623\u001b[1;33m \u001b[1;32mraise\u001b[0m \u001b[0mIOError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"%s not found.\"\u001b[0m \u001b[1;33m%\u001b[0m \u001b[0mpath\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 624\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 625\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;31mOSError\u001b[0m: Recommendations generated/ml-100k/Ready_I-KNNWithZScore.csv_reco.csv not found."
]
}
],
"source": [
"dir_path='Recommendations generated/ml-100k/'\n",
"super_reactions=[4,5]\n",
"test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\\t', header=None)\n",
"\n",
"ev.evaluate_all(test, dir_path, super_reactions)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"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": "3.8.3"
}
},
"nbformat": 4,
"nbformat_minor": 4
}