-
Notifications
You must be signed in to change notification settings - Fork 0
/
lightfm.py
816 lines (697 loc) · 33.2 KB
/
lightfm.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
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
# coding=utf-8
from __future__ import print_function
import numpy as np
import scipy.sparse as sp
from ._lightfm_fast import (CSRMatrix, FastLightFM,
fit_bpr, fit_logistic, fit_warp,
fit_warp_kos, predict_lightfm, predict_ranks)
__all__ = ['LightFM']
CYTHON_DTYPE = np.float32
class LightFM(object):
"""
A hybrid recommender model.
Parameters
----------
no_components: int, optional
the dimensionality of the feature latent embeddings.
k: int, optional
for k-OS training, the k-th positive example will be selected from the
n positive examples sampled for every user.
n: int, optional
for k-OS training, maximum number of positives sampled for each update.
learning_schedule: string, optional
one of ('adagrad', 'adadelta').
loss: string, optional
one of ('logistic', 'bpr', 'warp', 'warp-kos'): the loss function.
learning_rate: float, optional
initial learning rate for the adagrad learning schedule.
rho: float, optional
moving average coefficient for the adadelta learning schedule.
epsilon: float, optional
conditioning parameter for the adadelta learning schedule.
item_alpha: float, optional
L2 penalty on item features
user_alpha: float, optional
L2 penalty on user features.
max_sampled: int, optional
maximum number of negative samples used during WARP fitting.
It requires a lot of sampling to find negative triplets for users that
are already well represented by the model; this can lead to very long
training times and overfitting. Setting this to a higher number will
generally lead to longer training times, but may in some cases improve
accuracy.
random_state: int seed, RandomState instance, or None
The seed of the pseudo random number generator to use when shuffling
the data and initializing the parameters.
Attributes
----------
item_embeddings: np.float32 array of shape [n_item_features, n_components]
Contains the estimated latent vectors for item features. The [i, j]-th
entry gives the value of the j-th component for the i-th item feature.
In the simplest case where the item feature matrix is an identity
matrix, the i-th row will represent the i-th item latent vector.
user_embeddings: np.float32 array of shape [n_user_features, n_components]
Contains the estimated latent vectors for user features. The [i, j]-th
entry gives the value of the j-th component for the i-th user feature.
In the simplest case where the user feature matrix is an identity
matrix, the i-th row will represent the i-th user latent vector.
item_biases: np.float32 array of shape [n_item_features,]
Contains the biases for item_features.
user_biases: np.float32 array of shape [n_user_features,]
Contains the biases for user_features.
Notes
-----
Four loss functions are available:
- logistic: useful when both positive (1) and negative (-1) interactions
are present.
- BPR: Bayesian Personalised Ranking [1]_ pairwise loss. Maximises the
prediction difference between a positive example and a randomly
chosen negative example. Useful when only positive interactions
are present and optimising ROC AUC is desired.
- WARP: Weighted Approximate-Rank Pairwise [2]_ loss. Maximises
the rank of positive examples by repeatedly sampling negative
examples until rank violating one is found. Useful when only
positive interactions are present and optimising the top of
the recommendation list (precision@k) is desired.
- k-OS WARP: k-th order statistic loss [3]_. A modification of WARP that
uses the k-th positive example for any given user as a basis for pairwise
updates.
Two learning rate schedules are available:
- adagrad: [4]_
- adadelta: [5]_
References
----------
.. [1] Rendle, Steffen, et al. "BPR: Bayesian personalized ranking from
implicit feedback."
Proceedings of the Twenty-Fifth Conference on Uncertainty in
Artificial Intelligence. AUAI Press, 2009.
.. [2] Weston, Jason, Samy Bengio, and Nicolas Usunier. "Wsabie: Scaling up
to large vocabulary image annotation." IJCAI. Vol. 11. 2011.
.. [3] Weston, Jason, Hector Yee, and Ron J. Weiss. "Learning to rank
recommendations with the k-order statistic loss."
Proceedings of the 7th ACM conference on Recommender systems. ACM,
2013.
.. [4] Duchi, John, Elad Hazan, and Yoram Singer. "Adaptive subgradient
methods for online learning and stochastic optimization."
The Journal of Machine Learning Research 12 (2011): 2121-2159.
.. [5] Zeiler, Matthew D. "ADADELTA: An adaptive learning rate method."
arXiv preprint arXiv:1212.5701 (2012).
"""
def __init__(self, no_components=10, k=5, n=10,
learning_schedule='adagrad',
loss='logistic',
learning_rate=0.05, rho=0.95, epsilon=1e-6,
item_alpha=0.0, user_alpha=0.0, max_sampled=10,
random_state=None):
assert item_alpha >= 0.0
assert user_alpha >= 0.0
assert no_components > 0
assert k > 0
assert n > 0
assert 0 < rho < 1
assert epsilon >= 0
assert learning_schedule in ('adagrad', 'adadelta')
assert loss in ('logistic', 'warp', 'bpr', 'warp-kos')
if max_sampled < 1:
raise ValueError('max_sampled must be a positive integer')
self.loss = loss
self.learning_schedule = learning_schedule
self.no_components = no_components
self.learning_rate = learning_rate
self.k = int(k)
self.n = int(n)
self.rho = rho
self.epsilon = epsilon
self.max_sampled = max_sampled
self.item_alpha = item_alpha
self.user_alpha = user_alpha
if random_state is None:
self.random_state = np.random.RandomState()
elif isinstance(random_state, np.random.RandomState):
self.random_state = random_state
else:
self.random_state = np.random.RandomState(random_state)
self._reset_state()
def _reset_state(self):
self.item_embeddings = None
self.item_embedding_gradients = None
self.item_embedding_momentum = None
self.item_biases = None
self.item_bias_gradients = None
self.item_bias_momentum = None
self.user_embeddings = None
self.user_embedding_gradients = None
self.user_embedding_momentum = None
self.user_biases = None
self.user_bias_gradients = None
self.user_bias_momentum = None
def _check_initialized(self):
for var in (self.item_embeddings,
self.item_embedding_gradients,
self.item_embedding_momentum,
self.item_biases,
self.item_bias_gradients,
self.item_bias_momentum,
self.user_embeddings,
self.user_embedding_gradients,
self.user_embedding_momentum,
self.user_biases,
self.user_bias_gradients,
self.user_bias_momentum):
if var is None:
raise ValueError('You must fit the model before '
'trying to obtain predictions.')
def _initialize(self, no_components, no_item_features, no_user_features):
"""
Initialise internal latent representations.
"""
# Initialise item features.
self.item_embeddings = (
(self.random_state.rand(no_item_features, no_components) - 0.5) /
no_components).astype(np.float32)
self.item_embedding_gradients = np.zeros_like(self.item_embeddings)
self.item_embedding_momentum = np.zeros_like(self.item_embeddings)
self.item_biases = np.zeros(no_item_features, dtype=np.float32)
self.item_bias_gradients = np.zeros_like(self.item_biases)
self.item_bias_momentum = np.zeros_like(self.item_biases)
# Initialise user features.
self.user_embeddings = (
(self.random_state.rand(no_user_features, no_components) - 0.5) /
no_components).astype(np.float32)
self.user_embedding_gradients = np.zeros_like(self.user_embeddings)
self.user_embedding_momentum = np.zeros_like(self.user_embeddings)
self.user_biases = np.zeros(no_user_features, dtype=np.float32)
self.user_bias_gradients = np.zeros_like(self.user_biases)
self.user_bias_momentum = np.zeros_like(self.user_biases)
if self.learning_schedule == 'adagrad':
self.item_embedding_gradients += 1
self.item_bias_gradients += 1
self.user_embedding_gradients += 1
self.user_bias_gradients += 1
def _construct_feature_matrices(self, n_users, n_items, user_features,
item_features):
if user_features is None:
user_features = sp.identity(n_users,
dtype=CYTHON_DTYPE,
format='csr')
else:
user_features = user_features.tocsr()
if item_features is None:
item_features = sp.identity(n_items,
dtype=CYTHON_DTYPE,
format='csr')
else:
item_features = item_features.tocsr()
if n_users > user_features.shape[0]:
raise Exception('Number of user feature rows does not equal '
'the number of users')
if n_items > item_features.shape[0]:
raise Exception('Number of item feature rows does not equal '
'the number of items')
# If we already have embeddings, verify that
# we have them for all the supplied features
if self.user_embeddings is not None:
assert self.user_embeddings.shape[0] >= user_features.shape[1]
if self.item_embeddings is not None:
assert self.item_embeddings.shape[0] >= item_features.shape[1]
user_features = self._to_cython_dtype(user_features)
item_features = self._to_cython_dtype(item_features)
return user_features, item_features
def _get_positives_lookup_matrix(self, interactions):
mat = interactions.tocsr()
if not mat.has_sorted_indices:
return mat.sorted_indices()
else:
return mat
def _to_cython_dtype(self, mat):
if mat.dtype != CYTHON_DTYPE:
return mat.astype(CYTHON_DTYPE)
else:
return mat
def _process_sample_weight(self, interactions, sample_weight):
if sample_weight is not None:
if self.loss == 'warp-kos':
raise NotImplementedError('k-OS loss with sample weights '
'not implemented.')
if not isinstance(sample_weight, sp.coo_matrix):
raise ValueError('Sample_weight must be a COO matrix.')
if sample_weight.shape != interactions.shape:
raise ValueError('Sample weight and interactions '
'matrices must be the same shape')
if not (np.array_equal(interactions.row,
sample_weight.row) and
np.array_equal(interactions.col,
sample_weight.col)):
raise ValueError('Sample weight and interaction matrix '
'entries must be in the same order')
if sample_weight.data.dtype != CYTHON_DTYPE:
sample_weight_data = sample_weight.data.astype(CYTHON_DTYPE)
else:
sample_weight_data = sample_weight.data
else:
if np.array_equiv(interactions.data, 1.0):
# Re-use interactions data if they are all
# ones
sample_weight_data = interactions.data
else:
# Otherwise allocate a new array of ones
sample_weight_data = np.ones_like(interactions.data,
dtype=CYTHON_DTYPE)
return sample_weight_data
def _get_lightfm_data(self):
lightfm_data = FastLightFM(self.item_embeddings,
self.item_embedding_gradients,
self.item_embedding_momentum,
self.item_biases,
self.item_bias_gradients,
self.item_bias_momentum,
self.user_embeddings,
self.user_embedding_gradients,
self.user_embedding_momentum,
self.user_biases,
self.user_bias_gradients,
self.user_bias_momentum,
self.no_components,
int(self.learning_schedule == 'adadelta'),
self.learning_rate,
self.rho,
self.epsilon,
self.max_sampled)
return lightfm_data
def _check_finite(self):
for parameter in (self.item_embeddings,
self.item_biases,
self.user_embeddings,
self.user_biases):
# A sum of an array that contains non-finite values
# will also be non-finite, and we avoid creating a
# large boolean temporary.
if not np.isfinite(np.sum(parameter)):
raise ValueError("Not all estimated parameters are finite,"
" your model may have diverged. Try decreasing"
" the learning rate or normalising feature values"
" and sample weights")
def _check_input_finite(self, data):
if not np.isfinite(np.sum(data)):
raise ValueError('Not all input values are finite. '
'Check the input for NaNs and infinite values.')
def fit(self, interactions,
user_features=None, item_features=None,
sample_weight=None,
epochs=1, num_threads=1, verbose=False):
"""
Fit the model.
Arguments
---------
interactions: np.float32 coo_matrix of shape [n_users, n_items]
the matrix containing
user-item interactions. Will be converted to
numpy.float32 dtype if it is not of that type.
user_features: np.float32 csr_matrix of shape [n_users, n_user_features], optional
Each row contains that user's weights over features.
item_features: np.float32 csr_matrix of shape [n_items, n_item_features], optional
Each row contains that item's weights over features.
sample_weight: np.float32 coo_matrix of shape [n_users, n_items], optional
matrix with entries expressing weights of individual
interactions from the interactions matrix.
Its row and col arrays must be the same as
those of the interactions matrix. For memory
efficiency its possible to use the same arrays
for both weights and interaction matrices.
Defaults to weight 1.0 for all interactions.
Not implemented for the k-OS loss.
epochs: int, optional
number of epochs to run
num_threads: int, optional
Number of parallel computation threads to use. Should
not be higher than the number of physical cores.
verbose: bool, optional
whether to print progress messages.
Returns
-------
LightFM instance
the fitted model
"""
# Discard old results, if any
self._reset_state()
return self.fit_partial(interactions,
user_features=user_features,
item_features=item_features,
sample_weight=sample_weight,
epochs=epochs,
num_threads=num_threads,
verbose=verbose)
def fit_partial(self, interactions,
user_features=None, item_features=None,
sample_weight=None,
epochs=1, num_threads=1, verbose=False):
"""
Fit the model.
Fit the model. Unlike fit, repeated calls to this method will
cause training to resume from the current model state.
Arguments
---------
interactions: np.float32 coo_matrix of shape [n_users, n_items]
the matrix containing
user-item interactions. Will be converted to
numpy.float32 dtype if it is not of that type.
user_features: np.float32 csr_matrix of shape [n_users, n_user_features], optional
Each row contains that user's weights over features.
item_features: np.float32 csr_matrix of shape [n_items, n_item_features], optional
Each row contains that item's weights over features.
sample_weight: np.float32 coo_matrix of shape [n_users, n_items], optional
matrix with entries expressing weights of individual
interactions from the interactions matrix.
Its row and col arrays must be the same as
those of the interactions matrix. For memory
efficiency its possible to use the same arrays
for both weights and interaction matrices.
Defaults to weight 1.0 for all interactions.
Not implemented for the k-OS loss.
epochs: int, optional
number of epochs to run
num_threads: int, optional
Number of parallel computation threads to use. Should
not be higher than the number of physical cores.
verbose: bool, optional
whether to print progress messages.
Returns
-------
LightFM instance
the fitted model
"""
# We need this in the COO format.
# If that's already true, this is a no-op.
interactions = interactions.tocoo()
if interactions.dtype != CYTHON_DTYPE:
interactions.data = interactions.data.astype(CYTHON_DTYPE)
sample_weight_data = self._process_sample_weight(interactions,
sample_weight)
n_users, n_items = interactions.shape
(user_features,
item_features) = self._construct_feature_matrices(n_users,
n_items,
user_features,
item_features)
sample_weight = (self._to_cython_dtype(sample_weight)
if sample_weight is not None else
np.ones(interactions.getnnz(),
dtype=CYTHON_DTYPE))
for input_data in (user_features.data,
item_features.data,
sample_weight):
self._check_input_finite(input_data)
if self.item_embeddings is None:
# Initialise latent factors only if this is the first call
# to fit_partial.
self._initialize(self.no_components,
item_features.shape[1],
user_features.shape[1])
# Check that the dimensionality of the feature matrices has
# not changed between runs.
if not item_features.shape[1] == self.item_embeddings.shape[0]:
raise ValueError('Incorrect number of features in item_features')
if not user_features.shape[1] == self.user_embeddings.shape[0]:
raise ValueError('Incorrect number of features in user_features')
for epoch in range(epochs):
if verbose:
print('Epoch %s' % epoch)
self._run_epoch(item_features,
user_features,
interactions,
sample_weight_data,
num_threads,
self.loss)
self._check_finite()
return self
def _run_epoch(self, item_features, user_features, interactions,
sample_weight, num_threads, loss):
"""
Run an individual epoch.
"""
if loss in ('warp', 'bpr', 'warp-kos'):
# The CSR conversion needs to happen before shuffle indices are created.
# Calling .tocsr may result in a change in the data arrays of the COO matrix,
positives_lookup = CSRMatrix(
self._get_positives_lookup_matrix(interactions))
# Create shuffle indexes.
shuffle_indices = np.arange(len(interactions.data), dtype=np.int32)
self.random_state.shuffle(shuffle_indices)
lightfm_data = self._get_lightfm_data()
# Call the estimation routines.
if loss == 'warp':
fit_warp(CSRMatrix(item_features),
CSRMatrix(user_features),
positives_lookup,
interactions.row,
interactions.col,
interactions.data,
sample_weight,
shuffle_indices,
lightfm_data,
self.learning_rate,
self.item_alpha,
self.user_alpha,
num_threads,
self.random_state)
elif loss == 'bpr':
fit_bpr(CSRMatrix(item_features),
CSRMatrix(user_features),
positives_lookup,
interactions.row,
interactions.col,
interactions.data,
sample_weight,
shuffle_indices,
lightfm_data,
self.learning_rate,
self.item_alpha,
self.user_alpha,
num_threads,
self.random_state)
elif loss == 'warp-kos':
fit_warp_kos(CSRMatrix(item_features),
CSRMatrix(user_features),
positives_lookup,
interactions.row,
shuffle_indices,
lightfm_data,
self.learning_rate,
self.item_alpha,
self.user_alpha,
self.k,
self.n,
num_threads,
self.random_state)
else:
fit_logistic(CSRMatrix(item_features),
CSRMatrix(user_features),
interactions.row,
interactions.col,
interactions.data,
sample_weight,
shuffle_indices,
lightfm_data,
self.learning_rate,
self.item_alpha,
self.user_alpha,
num_threads)
def predict(self, user_ids, item_ids, item_features=None,
user_features=None, num_threads=1):
"""
Compute the recommendation score for user-item pairs.
Arguments
---------
user_ids: integer or np.int32 array of shape [n_pairs,]
single user id or an array containing the user ids for the
user-item pairs for which a prediction is to be computed
item_ids: np.int32 array of shape [n_pairs,]
an array containing the item ids for the user-item pairs for which
a prediction is to be computed.
user_features: np.float32 csr_matrix of shape [n_users, n_user_features], optional
Each row contains that user's weights over features.
item_features: np.float32 csr_matrix of shape [n_items, n_item_features], optional
Each row contains that item's weights over features.
num_threads: int, optional
Number of parallel computation threads to use. Should
not be higher than the number of physical cores.
Returns
-------
np.float32 array of shape [n_pairs,]
Numpy array containing the recommendation scores for pairs defined
by the inputs.
"""
self._check_initialized()
if not isinstance(user_ids, np.ndarray):
user_ids = np.repeat(np.int32(user_ids), len(item_ids))
assert len(user_ids) == len(item_ids)
if user_ids.dtype != np.int32:
user_ids = user_ids.astype(np.int32)
if item_ids.dtype != np.int32:
item_ids = item_ids.astype(np.int32)
n_users = user_ids.max() + 1
n_items = item_ids.max() + 1
(user_features,
item_features) = self._construct_feature_matrices(n_users,
n_items,
user_features,
item_features)
lightfm_data = self._get_lightfm_data()
predictions = np.empty(len(user_ids), dtype=np.float64)
predict_lightfm(CSRMatrix(item_features),
CSRMatrix(user_features),
user_ids,
item_ids,
predictions,
lightfm_data,
num_threads)
return predictions
def predict_rank(self, test_interactions, train_interactions=None,
item_features=None, user_features=None, num_threads=1):
"""
Predict the rank of selected interactions. Computes recommendation
rankings across all items for every user in interactions and calculates
the rank of all non-zero entries in the recommendation ranking, with 0
meaning the top of the list (most recommended) and n_items - 1 being
the end of the list (least recommended).
Performs best when only a handful of interactions need to be evaluated
per user. If you need to compute predictions for many items for every
user, use the predict method instead.
Arguments
---------
test_interactions: np.float32 csr_matrix of shape [n_users, n_items]
Non-zero entries denote the user-item pairs
whose rank will be computed.
train_interactions: np.float32 csr_matrix of shape [n_users, n_items], optional
Non-zero entries denote the user-item pairs which will be excluded
from rank computation. Use to exclude training set interactions
from being scored and ranked for evaluation.
user_features: np.float32 csr_matrix of shape [n_users, n_user_features], optional
Each row contains that user's weights over features.
item_features: np.float32 csr_matrix of shape [n_items, n_item_features], optional
Each row contains that item's weights over features.
num_threads: int, optional
Number of parallel computation threads to use.
Should not be higher than the number of physical cores.
Returns
-------
np.float32 csr_matrix of shape [n_users, n_items]
the [i, j]-th entry of the matrix will contain the rank of the j-th
item in the sorted recommendations list for the i-th user.
The degree of sparsity of this matrix will be equal to that of the
input interactions matrix.
"""
self._check_initialized()
n_users, n_items = test_interactions.shape
(user_features,
item_features) = self._construct_feature_matrices(n_users,
n_items,
user_features,
item_features)
if not item_features.shape[1] == self.item_embeddings.shape[0]:
raise ValueError('Incorrect number of features in item_features')
if not user_features.shape[1] == self.user_embeddings.shape[0]:
raise ValueError('Incorrect number of features in user_features')
test_interactions = test_interactions.tocsr()
test_interactions = self._to_cython_dtype(test_interactions)
if train_interactions is None:
train_interactions = sp.csr_matrix((n_users, n_items),
dtype=CYTHON_DTYPE)
else:
train_interactions = train_interactions.tocsr()
train_interactions = self._to_cython_dtype(train_interactions)
ranks = sp.csr_matrix((np.zeros_like(test_interactions.data),
test_interactions.indices,
test_interactions.indptr),
shape=test_interactions.shape)
lightfm_data = self._get_lightfm_data()
predict_ranks(CSRMatrix(item_features),
CSRMatrix(user_features),
CSRMatrix(test_interactions),
CSRMatrix(train_interactions),
ranks.data,
lightfm_data,
num_threads)
return ranks
def get_item_representations(self, features=None):
"""
Get the latent representations for items given model and features.
Arguments
---------
features: np.float32 csr_matrix of shape [n_items, n_item_features], optional
Each row contains that item's weights over features.
An identity matrix will be used if not supplied.
Returns
-------
(item_biases, item_embeddings):
(np.float32 array of shape n_items,
np.float32 array of shape [n_items, num_components]
Biases and latent representations for items.
"""
self._check_initialized()
if features is None:
return self.item_biases, self.item_embeddings
features = sp.csr_matrix(features, dtype=CYTHON_DTYPE)
return features * self.item_biases, features * self.item_embeddings
def get_user_representations(self, features=None):
"""
Get the latent representations for users given model and features.
Arguments
---------
features: np.float32 csr_matrix of shape [n_users, n_user_features], optional
Each row contains that user's weights over features.
An identity matrix will be used if not supplied.
Returns
-------
(user_biases, user_embeddings):
(np.float32 array of shape n_users
np.float32 array of shape [n_users, num_components]
Biases and latent representations for users.
"""
self._check_initialized()
if features is None:
return self.user_biases, self.user_embeddings
features = sp.csr_matrix(features, dtype=CYTHON_DTYPE)
return features * self.user_biases, features * self.user_embeddings
def get_params(self, deep=True):
"""
Get parameters for this estimator.
Arguments
---------
deep: boolean, optional
If True, will return the parameters for this estimator and
contained subobjects that are estimators.
Returns
-------
params : mapping of string to any
Parameter names mapped to their values.
"""
params = {'loss': self.loss,
'learning_schedule': self.learning_schedule,
'no_components': self.no_components,
'learning_rate': self.learning_rate,
'k': self.k,
'n': self.n,
'rho': self.rho,
'epsilon': self.epsilon,
'max_sampled': self.max_sampled,
'item_alpha': self.item_alpha,
'user_alpha': self.user_alpha,
'random_state': self.random_state}
return params
def set_params(self, **params):
"""
Set the parameters of this estimator.
Returns
-------
self
"""
valid_params = self.get_params()
for key, value in params.items():
if key not in valid_params:
raise ValueError('Invalid parameter %s for estimator %s. '
'Check the list of available parameters '
'with `estimator.get_params().keys()`.' %
(key, self.__class__.__name__))
setattr(self, key, value)
return self