-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrecmodels.py
294 lines (242 loc) · 11.7 KB
/
recmodels.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
from Recommenders.NonPersonalizedRecommender import TopPop
from Recommenders.KNN.ItemKNNCBFRecommender import ItemKNNCBFRecommender
from Recommenders.KNN.ItemKNNCFRecommender import ItemKNNCFRecommender
from Recommenders.KNN.UserKNNCFRecommender import UserKNNCFRecommender
from Recommenders.KNN.ItemKNNSimilarityHybridRecommender import ItemKNNSimilarityHybridRecommender
from Recommenders.MatrixFactorization.IALSRecommender import IALSRecommender
from Recommenders.SLIM.Cython.SLIM_BPR_Cython import SLIM_BPR_Cython
from Recommenders.GraphBased.P3alphaRecommender import P3alphaRecommender
from Recommenders.Hybrids.DiffStructHybridRecommender import DiffStructHybridRecommender
from Recommenders.EASE_R.EASE_R_Recommender import EASE_R_Recommender
from Recommenders.GraphBased.RP3betaRecommender import RP3betaRecommender
from Recommenders.MatrixFactorization.Cython.MatrixFactorization_Cython import MatrixFactorization_BPR_Cython, MatrixFactorization_FunkSVD_Cython, MatrixFactorization_AsySVD_Cython
from Recommenders.MatrixFactorization.PureSVDRecommender import PureSVDRecommender, PureSVDItemRecommender, ScaledPureSVDRecommender
from Recommenders.MatrixFactorization.SVDFeatureRecommender import SVDFeature
from Recommenders.FactorizationMachines.FMRecommender import LightFMRecommender
class TopPopRec(TopPop):
def fit(self):
super(TopPopRec, self).fit()
self.RECOMMENDER_VERSION = "classic"
class ItemKNNCBFRec(ItemKNNCBFRecommender):
def fit(self,
topK=50,
shrink=100,
normalize = True,
similarity = "cosine",
feature_weighting = "none",
ICM_bias = None,
**similarity_args):
super(ItemKNNCBFRec, self).fit(
topK,
shrink,similarity,
normalize,
feature_weighting,
ICM_bias,
**similarity_args)
self.RECOMMENDER_VERSION = "topK-" + str(topK) + "_shrink-" + str(shrink) +"_feature_weighting-"+feature_weighting+ "_sim-" + similarity
class ItemKNNCFRec(ItemKNNCFRecommender):
def fit(self,
topK=50,
shrink=100,
similarity='cosine',
normalize=True,
feature_weighting = "none",
URM_bias = False,
**similarity_args):
super(ItemKNNCFRec, self).fit(
topK,
shrink,
similarity,
normalize,
feature_weighting,
URM_bias,
**similarity_args)
self.RECOMMENDER_VERSION = "topK-" + str(topK) + "_shrink-" + str(shrink) +"_feature_weighting-"+feature_weighting+ "_sim-" + similarity
class UserKNNCFRec(UserKNNCFRecommender):
def fit(self,
topK=50,
shrink=100,
similarity='cosine',
normalize=True,
feature_weighting = "none",
URM_bias = False,
**similarity_args):
super(UserKNNCFRec, self).fit(
topK,
shrink,
similarity,
normalize,
feature_weighting,
URM_bias,
**similarity_args)
self.RECOMMENDER_VERSION = "topK-" + str(topK) + "_shrink-" + str(shrink) +"_feature_weighting-"+feature_weighting+ "_sim-" + similarity
class IALSRec(IALSRecommender):
def fit(self,
epochs = 300,
num_factors = 20,
confidence_scaling = "linear",
alpha = 1.0,
epsilon = 1.0,
reg = 1e-3,
init_mean=0.0,
init_std=0.1,
**earlystopping_kwargs):
super(IALSRec, self).fit(
epochs,
num_factors,
confidence_scaling,
alpha,
epsilon,
reg,
init_mean,
init_std,
**earlystopping_kwargs)
self.RECOMMENDER_VERSION = "numfact-" + str(num_factors) + "_alpha-" + str(alpha) + "_reg-" + str(reg)
class SLIM_BPRRec(SLIM_BPR_Cython):
def fit(self,
epochs=300,
positive_threshold_BPR = None,
train_with_sparse_weights = None,
allow_train_with_sparse_weights = True,
symmetric = True,
random_seed = None,
lambda_i = 0.0,
lambda_j = 0.0,
learning_rate = 1e-4,
topK = 200,
sgd_mode='adagrad',
gamma=0.995,
beta_1=0.9,
beta_2=0.999,
**earlystopping_kwargs):
super(SLIM_BPRRec, self).fit(
epochs = epochs,
positive_threshold_BPR = positive_threshold_BPR,
train_with_sparse_weights = train_with_sparse_weights,
allow_train_with_sparse_weights = allow_train_with_sparse_weights,
symmetric = symmetric,
random_seed = random_seed,
lambda_i = lambda_i,
lambda_j = lambda_j,
learning_rate = learning_rate,
topK = topK,
sgd_mode = sgd_mode,
gamma = gamma,
beta_1 = beta_1,
beta_2 = beta_2,
**earlystopping_kwargs)
self.RECOMMENDER_VERSION = "epochs-" + str(epochs) + "_topK-" + str(topK) + "_lambda_i-" + str(lambda_i) + "_lambda_j-" + str(lambda_j) + "_learning_rate-" + str(learning_rate) + "_sym-" + str(symmetric) + "_sgd-" + str(sgd_mode)
class P3AlphaRec(P3alphaRecommender):
def fit(self,
topK = 100,
alpha = 1.,
min_rating = 0,
implicit = False,
normalize_similarity = False):
super(P3AlphaRec, self).fit(
topK,
alpha,
min_rating,
implicit,
normalize_similarity)
self.RECOMMENDER_VERSION = "topK-" + str(topK) + "_alpha-" + str(alpha) + "_min_rating-" + str(min_rating) + "_implicit-" + str(implicit) + "_normalize_similarity-" + str(normalize_similarity)
class RP3BetaRec(RP3betaRecommender):
def fit(self,
alpha = 1.,
beta = 0.6,
min_rating = 0,
topK = 100,
implicit = False,
normalize_similarity = True):
super(RP3BetaRec, self).fit(
alpha,
beta,
min_rating,
topK,
implicit,
normalize_similarity)
self.RECOMMENDER_VERSION = "topK-" + str(topK) + "_beta-" + str(beta) + "_alpha-" + str(alpha) + "_min_rating-" + str(min_rating) + "_implicit-" + str(implicit) + "_normalize_similarity-" + str(normalize_similarity)
class EASE_R_Rec(EASE_R_Recommender):
def fit(self,
topK = None,
l2_norm = 1e3,
normalize_metrics = False):
super(EASE_R_Rec, self).fit(
topK,
l2_norm,
normalize_metrics)
self.RECOMMENDER_VERSION = "topK-" + str(topK) + "_l2_norm-" + str(l2_norm) + "_normalize_metrics-" + str(normalize_metrics)
class MatrixFactorizationBPRRec(MatrixFactorization_BPR_Cython):
def fit(self,
epochs=300,
batch_size = 1000,
num_factors=10,
positive_threshold_BPR = None,
learning_rate = 0.001,
use_bias = True,
use_embeddings = True,
sgd_mode='sgd',
negative_interactions_quota = 0.0,
dropout_quota = None,
init_mean = 0.0,
init_std_dev = 0.1,
user_reg = 0.0,
item_reg = 0.0,
bias_reg = 0.0,
positive_reg = 0.0,
negative_reg = 0.0,
random_seed = None,
**earlystopping_kwargs):
super(MatrixFactorizationBPRRec, self).fit(epochs = epochs,
batch_size = batch_size,
num_factors=num_factors,
positive_threshold_BPR = positive_threshold_BPR,
learning_rate = learning_rate,
use_bias = use_bias,
use_embeddings = use_embeddings,
sgd_mode = sgd_mode,
negative_interactions_quota = negative_interactions_quota,
dropout_quota = dropout_quota,
init_mean = init_mean,
init_std_dev = init_std_dev,
user_reg = user_reg,
item_reg = item_reg,
bias_reg = bias_reg,
positive_reg = positive_reg,
negative_reg = negative_reg,
random_seed = 1,
**earlystopping_kwargs)
self.RECOMMENDER_VERSION = "epochs-" + str(epochs) + "_batch-" + str(batch_size) + "_nfactors-" + str(num_factors) + "_learnrate-" + str(learning_rate) + "_usebias-" + str(use_bias) + "_useembed-" + str(use_embeddings) + "_sgdmode-" + str(sgd_mode) + "_neginter-" + str(negative_interactions_quota) + "_dropout-" + str(dropout_quota) + "_userreg-" + str(user_reg) + "_itemreg-" + str(item_reg) + "_biasreg-" + str(bias_reg) + "_posreg-" + str(positive_reg) + "_negreg-" + str(negative_reg)
class FunkSVDRec(MatrixFactorizationBPRRec):
RECOMMENDER_NAME = "MatrixFactorization_FunkSVD_Cython_Recommender"
class AsySVDRec(MatrixFactorizationBPRRec):
RECOMMENDER_NAME = "MatrixFactorization_AsySVD_Cython_Recommender"
class PureSVDRec(PureSVDRecommender):
def fit(self, num_factors=100, random_seed = None):
super(PureSVDRec, self).fit(num_factors=num_factors, random_seed=random_seed)
self.RECOMMENDER_VERSION = "numfactors-" + str(num_factors)
class PureSVDItemRec(PureSVDItemRecommender):
def fit(self, num_factors=100, topK = None, random_seed = None):
super(PureSVDItemRec, self).fit(num_factors=num_factors, topK=topK, random_seed=random_seed)
self.RECOMMENDER_VERSION = "numfactors-" + str(num_factors) + "_topK-" + str(topK)
class ScaledPureSVDRec(ScaledPureSVDRecommender):
def fit(self, num_factors = 100, random_seed = None, scaling_items = 1.0, scaling_users = 1.0):
super(ScaledPureSVDRec, self).fit(num_factors=num_factors, random_seed=random_seed, scaling_items=scaling_items, scaling_users=scaling_users)
self.RECOMMENDER_VERSION = "numfactors-" + str(num_factors) + "_scalingitems-" + str(scaling_items) + "_scalingusers-" + str(scaling_users)
class SVDFeatureRec(SVDFeature):
def fit(self, epochs=30, num_factors=32, learning_rate=0.01,
user_reg=0.0, item_reg=0.0, user_bias_reg=0.0, item_bias_reg=0.0,
temp_file_folder = None):
super(SVDFeatureRec, self).fit(epochs=epochs, num_factors=num_factors, learning_rate=learning_rate,
user_reg=user_reg, item_reg=item_reg, user_bias_reg=user_bias_reg, item_bias_reg=item_bias_reg,
temp_file_folder = temp_file_folder)
self.RECOMMENDER_VERSION = "epochs-" + str(epochs) + "_nfactors-" + str(num_factors) + "_learnrate-" + str(learning_rate) + "_userreg-" + str(user_reg) + "_itemreg-" + str(item_reg) + "_userbiasreg-" + str(user_bias_reg) + "__itembiasreg-" + str(item_bias_reg)
class ItemKNNSimilarityHybridRec(ItemKNNSimilarityHybridRecommender):
RECOMMENDER_VERSION = "best_version"
def __init__(self, URM_train, verbose = True):
self.URM_train = URM_train
self.verbose = verbose
def fit(self, Similarity_1, Similarity_2, topK=100, alpha = 0.5, similarities_string = ""):
super(ItemKNNSimilarityHybridRec, self).__init__(self.URM_train, Similarity_1, Similarity_2, verbose = self.verbose)
super(ItemKNNSimilarityHybridRec, self).fit(topK=topK, alpha=alpha)
#self.RECOMMENDER_VERSION = similarities_string + "topK-" + str(topK) + "_alpha-" + str(alpha)