Self made RP3-beta

import helpers
import pandas as pd
import numpy as np
import scipy.sparse as sparse
from collections import defaultdict
from itertools import chain
import random
import time
import matplotlib.pyplot as plt

train_read=pd.read_csv('./Datasets/ml-100k/train.csv', sep='\t', header=None)
test_read=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\t', header=None)
train_ui, test_ui, user_code_id, user_id_code, item_code_id, item_id_code = helpers.data_to_csr(train_read, test_read)

class RP3Beta():
    def fit(self, train_ui, alpha, beta):
        """We weight our edges by user's explicit ratings so if user rated movie high we'll follow that path
        with higher probability."""
        self.train_ui=train_ui
        self.train_iu=train_ui.transpose()
        
        self.alpha = alpha
        self.beta = beta
        
        # Define Pui 
        Pui=sparse.csr_matrix(self.train_ui/self.train_ui.sum(axis=1))
        
        # Define Piu
        to_divide=np.vectorize(lambda x: x if x>0 else 1)(self.train_iu.sum(axis=1)) # to avoid dividing by zero
        Piu=sparse.csr_matrix(self.train_iu/to_divide)
        item_orders=(self.train_ui>0).sum(axis=0)
        
        Pui = Pui.power(self.alpha)
        Piu = Piu.power(self.alpha)

        P3=Pui*Piu*Pui
        
        P3/=np.power(np.vectorize(lambda x: x if x>0 else 1)(item_orders), self.beta)
        
        self.estimations=np.array(P3)
    
    def recommend(self, user_code_id, item_code_id, topK=10):
        
        top_k = defaultdict(list)
        for nb_user, user in enumerate(self.estimations):
            
            user_rated=self.train_ui.indices[self.train_ui.indptr[nb_user]:self.train_ui.indptr[nb_user+1]]
            for item, score in enumerate(user):
                if item not in user_rated and not np.isnan(score):
                    top_k[user_code_id[nb_user]].append((item_code_id[item], score))
        result=[]
        # Let's choose k best items in the format: (user, item1, score1, item2, score2, ...)
        for uid, item_scores in top_k.items():
            item_scores.sort(key=lambda x: x[1], reverse=True)
            result.append([uid]+list(chain(*item_scores[:topK])))
        return result
    
    def estimate(self, user_code_id, item_code_id, test_ui):
        result=[]
        for user, item in zip(*test_ui.nonzero()):
            result.append([user_code_id[user], item_code_id[item], 
                           self.estimations[user,item] if not np.isnan(self.estimations[user,item]) else 1])
        return result

model=RP3Beta()
model.fit(train_ui, alpha=1, beta=0)

top_n=pd.DataFrame(model.recommend(user_code_id, item_code_id, topK=10))

top_n.to_csv('Recommendations generated/ml-100k/Self_P3_reco.csv', index=False, header=False)

estimations=pd.DataFrame(model.estimate(user_code_id, item_code_id, test_ui))
estimations.to_csv('Recommendations generated/ml-100k/Self_P3_estimations.csv', index=False, header=False)

import evaluation_measures as ev
estimations_df=pd.read_csv('Recommendations generated/ml-100k/Self_P3_estimations.csv', header=None)
reco=np.loadtxt('Recommendations generated/ml-100k/Self_P3_reco.csv', delimiter=',')

ev.evaluate(test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\t', header=None),
            estimations_df=estimations_df, 
            reco=reco,
            super_reactions=[4,5])

943it [00:00, 2167.43it/s]

	RMSE	MAE	precision	recall	F_1	F_05	precision_super	recall_super	NDCG	mAP	MRR	LAUC	HR	HR2	Reco in test	Test coverage	Shannon	Gini
0	3.702446	3.527273	0.282185	0.192092	0.186749	0.21698	0.204185	0.240096	0.339114	0.204905	0.572157	0.593544	0.875928	0.685048	1.0	0.077201	3.875892	0.974947

Let's check hiperparameters

Alpha

from tqdm import tqdm
result=[]
for alpha in tqdm([round(i,1) for i in np.arange(0.2,1.6001,0.2)]):
    model=RP3Beta()
    model.fit(train_ui, alpha=alpha, beta=0)
    reco=pd.DataFrame(model.recommend(user_code_id, item_code_id, topK=10))
    estimations_df=pd.DataFrame(model.estimate(user_code_id, item_code_id, test_ui))
    to_append=ev.evaluate(test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\t', header=None),
            estimations_df=estimations_df, 
            reco=np.array(reco),
            super_reactions=[4,5])
    to_append.insert(0, "Alpha", alpha)
    result.append(to_append)
    
result=pd.concat(result)
result

  0%|                                                                                            | 0/8 [00:00<?, ?it/s]
0it [00:00, ?it/s][A
280it [00:00, 2775.23it/s][A
526it [00:00, 2666.15it/s][A
943it [00:00, 2520.30it/s][A
 12%|██████████▌                                                                         | 1/8 [00:24<02:50, 24.33s/it]
0it [00:00, ?it/s][A
247it [00:00, 2452.10it/s][A
494it [00:00, 2452.09it/s][A
943it [00:00, 2484.84it/s][A
 25%|█████████████████████                                                               | 2/8 [00:49<02:28, 24.72s/it]
0it [00:00, ?it/s][A
159it [00:00, 1578.44it/s][A
413it [00:00, 1777.94it/s][A
584it [00:00, 1756.96it/s][A
712it [00:00, 1389.14it/s][A
943it [00:00, 1570.48it/s][A
 38%|███████████████████████████████▌                                                    | 3/8 [01:15<02:04, 24.87s/it]
0it [00:00, ?it/s][A
229it [00:00, 2273.39it/s][A
544it [00:00, 2476.21it/s][A
943it [00:00, 2626.45it/s][A
 50%|██████████████████████████████████████████                                          | 4/8 [01:39<01:38, 24.58s/it]
0it [00:00, ?it/s][A
232it [00:00, 2303.19it/s][A
472it [00:00, 2244.92it/s][A
618it [00:00, 1927.54it/s][A
943it [00:00, 2102.99it/s][A
 62%|████████████████████████████████████████████████████▌                               | 5/8 [02:02<01:12, 24.22s/it]
0it [00:00, ?it/s][A
187it [00:00, 1856.41it/s][A
347it [00:00, 1766.98it/s][A
473it [00:00, 1572.31it/s][A
633it [00:00, 1577.13it/s][A
943it [00:00, 1719.13it/s][A
 75%|███████████████████████████████████████████████████████████████                     | 6/8 [02:24<00:47, 23.62s/it]
0it [00:00, ?it/s][A
268it [00:00, 2660.58it/s][A
534it [00:00, 2658.40it/s][A
943it [00:00, 2583.15it/s][A
 88%|█████████████████████████████████████████████████████████████████████████▌          | 7/8 [02:49<00:23, 23.99s/it]
0it [00:00, ?it/s][A
167it [00:00, 1657.84it/s][A
276it [00:00, 1413.02it/s][A
368it [00:00, 1213.79it/s][A
489it [00:00, 1206.38it/s][A
672it [00:00, 1342.97it/s][A
943it [00:00, 1444.56it/s][A
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [03:13<00:00, 24.16s/it]

Alpha	RMSE	MAE	precision	recall	F_1	F_05	precision_super	recall_super	NDCG	mAP	MRR	LAUC	HR	HR2	Reco in test	Test coverage	Shannon	Gini
0.2	268.177832	211.732649	0.262672	0.166858	0.166277	0.197184	0.187661	0.203252	0.320910	0.196132	0.563378	0.580866	0.850477	0.629905	1.000000	0.060606	3.669627	0.979636
0.4	10.546689	7.792373	0.268505	0.172669	0.171569	0.202643	0.192489	0.212653	0.326760	0.200172	0.565148	0.583801	0.854719	0.644751	1.000000	0.064214	3.726996	0.978426
0.6	3.143988	2.948790	0.274655	0.180502	0.177820	0.208730	0.198176	0.222746	0.332872	0.203290	0.568872	0.587738	0.870626	0.657476	1.000000	0.065657	3.785282	0.977090
0.8	3.670728	3.495735	0.281972	0.189868	0.185300	0.216071	0.203541	0.236751	0.339867	0.206688	0.573729	0.592432	0.874867	0.685048	1.000000	0.070707	3.832415	0.975998
1.0	3.702446	3.527273	0.282185	0.192092	0.186749	0.216980	0.204185	0.240096	0.339114	0.204905	0.572157	0.593544	0.875928	0.685048	1.000000	0.077201	3.875892	0.974947
1.2	3.704441	3.529251	0.280912	0.193633	0.187311	0.216872	0.203004	0.240588	0.338049	0.203453	0.571830	0.594313	0.883351	0.681866	1.000000	0.085859	3.910718	0.974073
1.4	3.704580	3.529388	0.273595	0.190651	0.183874	0.212183	0.199464	0.239118	0.329550	0.195433	0.566171	0.592793	0.871686	0.675504	1.000000	0.107504	3.961915	0.972674
1.6	3.704591	3.529399	0.263097	0.186255	0.178709	0.205170	0.191094	0.232920	0.317439	0.184917	0.552349	0.590545	0.868505	0.669141	0.999576	0.156566	4.060156	0.969203

metrics=list(result.columns[[i not in ['Alpha'] for i in result.columns]])

charts_per_row=6
charts_per_column=3

fig, axes = plt.subplots(nrows=charts_per_row, ncols=charts_per_column,figsize=(18, 7*charts_per_row ))
import itertools
to_iter=[i for i in itertools.product(range(charts_per_row), range(charts_per_column))]

for i in range(len(metrics)):
    df=result[['Alpha', metrics[i]]]
    df.plot(ax=axes[to_iter[i]], title=metrics[i], x=0, y=1)

Beta

from tqdm import tqdm
result=[]
for beta in tqdm([round(i,1) for i in np.arange(0,1,0.1)]):
    model=RP3Beta()
    model.fit(train_ui, alpha=1, beta=beta)
    reco=pd.DataFrame(model.recommend(user_code_id, item_code_id, topK=10))
    estimations_df=pd.DataFrame(model.estimate(user_code_id, item_code_id, test_ui))
    to_append=ev.evaluate(test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\t', header=None),
            estimations_df=estimations_df, 
            reco=np.array(reco),
            super_reactions=[4,5])
    to_append.insert(0, "Beta", beta)
    result.append(to_append)
    
result=pd.concat(result)
result

  0%|                                                                                           | 0/10 [00:00<?, ?it/s]
0it [00:00, ?it/s][A
289it [00:00, 2869.05it/s][A
943it [00:00, 3796.57it/s][A
 10%|████████▎                                                                          | 1/10 [00:13<02:03, 13.71s/it]
0it [00:00, ?it/s][A
462it [00:00, 4605.96it/s][A
943it [00:00, 3601.05it/s][A
 20%|████████████████▌                                                                  | 2/10 [00:35<02:09, 16.14s/it]
0it [00:00, ?it/s][A
943it [00:00, 5872.90it/s][A
 30%|████████████████████████▉                                                          | 3/10 [00:50<01:50, 15.73s/it]
0it [00:00, ?it/s][A
351it [00:00, 3484.49it/s][A
943it [00:00, 3341.08it/s][A
 40%|█████████████████████████████████▏                                                 | 4/10 [01:05<01:33, 15.65s/it]
0it [00:00, ?it/s][A
485it [00:00, 4812.21it/s][A
943it [00:00, 4202.34it/s][A
 50%|█████████████████████████████████████████▌                                         | 5/10 [01:23<01:21, 16.34s/it]
0it [00:00, ?it/s][A
943it [00:00, 4873.77it/s][A
 60%|█████████████████████████████████████████████████▊                                 | 6/10 [01:38<01:04, 16.00s/it]
0it [00:00, ?it/s][A
258it [00:00, 2561.22it/s][A
555it [00:00, 2666.26it/s][A
943it [00:00, 2693.80it/s][A
 70%|██████████████████████████████████████████████████████████                         | 7/10 [01:57<00:50, 16.73s/it]
0it [00:00, ?it/s][A
943it [00:00, 5416.39it/s][A
 80%|██████████████████████████████████████████████████████████████████▍                | 8/10 [02:17<00:35, 17.72s/it]
0it [00:00, ?it/s][A
158it [00:00, 1568.57it/s][A
295it [00:00, 1499.58it/s][A
943it [00:00, 2565.22it/s][A
 90%|██████████████████████████████████████████████████████████████████████████▋        | 9/10 [02:33<00:17, 17.21s/it]
0it [00:00, ?it/s][A
262it [00:00, 2600.87it/s][A
572it [00:00, 2728.34it/s][A
943it [00:00, 2883.56it/s][A
100%|██████████████████████████████████████████████████████████████████████████████████| 10/10 [02:51<00:00, 17.16s/it]

Beta	RMSE	MAE	precision	recall	F_1	F_05	precision_super	recall_super	NDCG	mAP	MRR	LAUC	HR	HR2	Reco in test	Test coverage	Shannon	Gini
0.0	3.702446	3.527273	0.282185	0.192092	0.186749	0.216980	0.204185	0.240096	0.339114	0.204905	0.572157	0.593544	0.875928	0.685048	1.000000	0.077201	3.875892	0.974947
0.1	3.703312	3.528128	0.290138	0.197597	0.192259	0.223336	0.210944	0.246153	0.347768	0.212034	0.581038	0.596328	0.884411	0.695652	1.000000	0.085137	3.957416	0.972784
0.2	3.703825	3.528636	0.297137	0.201202	0.196067	0.228169	0.218026	0.252767	0.355655	0.219909	0.588904	0.598160	0.886532	0.697773	1.000000	0.094517	4.053212	0.969980
0.3	3.704130	3.528939	0.303499	0.204749	0.199901	0.232829	0.225107	0.260797	0.363757	0.226825	0.599969	0.599964	0.888653	0.707317	1.000000	0.105339	4.147779	0.966948
0.4	3.704313	3.529120	0.308908	0.208811	0.203854	0.237241	0.229614	0.266918	0.370758	0.232673	0.609385	0.602014	0.895016	0.718982	0.999894	0.132035	4.259682	0.962989
0.5	3.704422	3.529229	0.314316	0.211411	0.206768	0.240986	0.237124	0.273416	0.378307	0.239297	0.622792	0.603327	0.903499	0.724284	0.999046	0.168831	4.411281	0.956648
0.6	3.704488	3.529295	0.314634	0.206209	0.204818	0.240159	0.242489	0.273850	0.376438	0.238428	0.622042	0.600721	0.897137	0.720042	0.996394	0.212843	4.621938	0.945932
0.7	3.704528	3.529335	0.304136	0.187298	0.191990	0.228749	0.238305	0.256201	0.358807	0.226808	0.593897	0.591207	0.868505	0.693531	0.983033	0.256854	4.898568	0.928065
0.8	3.704552	3.529360	0.266384	0.147571	0.158660	0.194838	0.214485	0.209336	0.299850	0.184356	0.492852	0.571152	0.803818	0.604454	0.936373	0.341270	5.257397	0.895882
0.9	3.704567	3.529375	0.162354	0.076967	0.089233	0.114583	0.134657	0.113253	0.160868	0.085486	0.243590	0.535405	0.580064	0.400848	0.800106	0.415584	5.563910	0.857396

### import matplotlib.pyplot as plt

metrics=list(result.columns[[i not in ['Beta'] for i in result.columns]])

charts_per_row=6
charts_per_column=3

fig, axes = plt.subplots(nrows=charts_per_row, ncols=charts_per_column,figsize=(18, 7*charts_per_row ))
import itertools
to_iter=[i for i in itertools.product(range(charts_per_row), range(charts_per_column))]

for i in range(len(metrics)):
    df=result[['Beta', metrics[i]]]
    df.plot(ax=axes[to_iter[i]], title=metrics[i], x=0, y=1)

Check sample recommendations

train=pd.read_csv('./Datasets/ml-100k/train.csv', sep='\t', header=None, names=['user', 'item', 'rating', 'timestamp'])
items=pd.read_csv('./Datasets/ml-100k/movies.csv')

user=random.choice(list(set(train['user'])))

train_content=pd.merge(train, items, left_on='item', right_on='id')
display(train_content[train_content['user']==user][['user', 'rating', 'title', 'genres']]\
        .sort_values(by='rating', ascending=False)[:15])

reco = np.loadtxt('Recommendations generated/ml-100k/Self_P3_reco.csv', delimiter=',')
items=pd.read_csv('./Datasets/ml-100k/movies.csv')

# Let's ignore scores - they are not used in evaluation: 
reco_users=reco[:,:1]
reco_items=reco[:,1::2]
# Let's put them into one array
reco=np.concatenate((reco_users, reco_items), axis=1)

# Let's rebuild it user-item dataframe
recommended=[]
for row in reco:
    for rec_nb, entry in enumerate(row[1:]):
        recommended.append((row[0], rec_nb+1, entry))
recommended=pd.DataFrame(recommended, columns=['user','rec_nb', 'item'])

recommended_content=pd.merge(recommended, items, left_on='item', right_on='id')
recommended_content[recommended_content['user']==user][['user', 'rec_nb', 'title', 'genres']].sort_values(by='rec_nb')

	user	rating	title	genres
82	200	5	Toy Story (1995)	Animation, Children's, Comedy
22169	200	5	Groundhog Day (1993)	Comedy, Romance
41229	200	5	Star Trek: Generations (1994)	Action, Adventure, Sci-Fi
40846	200	5	Miracle on 34th Street (1994)	Drama
38062	200	5	Beauty and the Beast (1991)	Animation, Children's, Musical
37534	200	5	Aladdin (1992)	Animation, Children's, Comedy, Musical
36306	200	5	Silence of the Lambs, The (1991)	Drama, Thriller
33862	200	5	Independence Day (ID4) (1996)	Action, Sci-Fi, War
32487	200	5	Phenomenon (1996)	Drama, Romance
32420	200	5	Spawn (1997)	Action, Adventure, Sci-Fi, Thriller
31448	200	5	Seven (Se7en) (1995)	Crime, Thriller
30214	200	5	Star Trek: First Contact (1996)	Action, Adventure, Sci-Fi
27813	200	5	Day the Earth Stood Still, The (1951)	Drama, Sci-Fi
27483	200	5	Young Frankenstein (1974)	Comedy, Horror
27366	200	5	Empire Strikes Back, The (1980)	Action, Adventure, Drama, Romance, Sci-Fi, War

	user	rec_nb	title	genres
2710	200.0	1	Return of the Jedi (1983)	Action, Adventure, Romance, Sci-Fi, War
7001	200.0	2	Fargo (1996)	Crime, Drama, Thriller
2272	200.0	3	Godfather, The (1972)	Action, Crime, Drama
6393	200.0	4	Pulp Fiction (1994)	Crime, Drama
7714	200.0	5	Princess Bride, The (1987)	Action, Adventure, Comedy, Romance
3434	200.0	6	Jerry Maguire (1996)	Drama, Romance
9101	200.0	7	Mission: Impossible (1996)	Action, Adventure, Mystery
4491	200.0	8	Air Force One (1997)	Action, Thriller
2026	200.0	9	Back to the Future (1985)	Comedy, Sci-Fi
8048	200.0	10	Monty Python and the Holy Grail (1974)	Comedy

project task 6: generate recommendations of RP3Beta for hiperparameters found to optimize recall

# use better values than (1,0) for alpha and beta
# if you want you can also modify the model to consider different weights (we took as weights user ratings, maybe take ones or squares of ratings instead)
# save the outptut in 'Recommendations generated/ml-100k/Self_RP3Beta_estimations.csv'
# and 'Recommendations generated/ml-100k/Self_RP3Beta_reco.csv'

model=RP3Beta()
model.fit(train_ui, alpha=1.2, beta=0.5)

top_n=pd.DataFrame(model.recommend(user_code_id, item_code_id, topK=10))

top_n.to_csv('Recommendations generated/ml-100k/Self_RP3Beta_reco.csv', index=False, header=False)

estimations=pd.DataFrame(model.estimate(user_code_id, item_code_id, test_ui))
estimations.to_csv('Recommendations generated/ml-100k/Self_RP3Beta_estimations.csv', index=False, header=False)

import evaluation_measures as ev
estimations_df=pd.read_csv('Recommendations generated/ml-100k/Self_RP3Beta_estimations.csv', header=None)
reco=np.loadtxt('Recommendations generated/ml-100k/Self_RP3Beta_reco.csv', delimiter=',')

ev.evaluate(test=pd.read_csv('./Datasets/ml-100k/test.csv', sep='\t', header=None),
            estimations_df=estimations_df, 
            reco=reco,
            super_reactions=[4,5])

943it [00:00, 3493.16it/s]

	RMSE	MAE	precision	recall	F_1	F_05	precision_super	recall_super	NDCG	mAP	MRR	LAUC	HR	HR2	Reco in test	Test coverage	Shannon	Gini
0	3.70458	3.529388	0.302969	0.199894	0.197705	0.231449	0.231438	0.263787	0.362426	0.226406	0.601293	0.597526	0.889714	0.700954	0.996819	0.212121	4.509878	0.951344

project task 7 (optional): implement graph-based model of your choice

# for example change length of paths in RP3beta
# save the outptut in 'Recommendations generated/ml-100k/Self_GraphTask_estimations.csv'
# and 'Recommendations generated/ml-100k/Self_GraphTask_reco.csv'

576 KiB Raw Blame History Unescape Escape

Self made RP3-beta

Let's check hiperparameters

Alpha

Beta

Check sample recommendations

project task 6: generate recommendations of RP3Beta for hiperparameters found to optimize recall

project task 7 (optional): implement graph-based model of your choice

576 KiB

Raw Blame History