tf-models-nightly 2.13.1.dev20230811__py2.py3-none-any.whl → 2.13.1.dev20230813__py2.py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -214,6 +214,7 @@ class TrainerConfig(base_config.Config):
214
214
  train_tf_while_loop: whether or not to use tf while loop.
215
215
  train_tf_function: whether or not to use tf_function for training loop.
216
216
  eval_tf_function: whether or not to use tf_function for eval.
217
+ eval_tf_while_loop: whether or not to use tf while loop for eval.
217
218
  allow_tpu_summary: Whether to allow summary happen inside the XLA program
218
219
  runs on TPU through automatic outside compilation.
219
220
  steps_per_loop: number of steps per loop to report training metrics. This
@@ -528,9 +528,15 @@ class FunnelTransformerEncoder(tf.keras.layers.Layer):
528
528
  axes=[1])
529
529
 
530
530
  for i, layer in enumerate(self._transformer_layers):
531
+ transformer_output_range = None
532
+ if i == self._num_layers - 1:
533
+ transformer_output_range = output_range
534
+
531
535
  # Bypass no pooling cases.
532
536
  if self._pool_strides[i] == 1:
533
- x = layer([x, x, attention_mask])
537
+ x = layer(
538
+ [x, x, attention_mask], output_range=transformer_output_range
539
+ )
534
540
  else:
535
541
  # Pools layer for compressing the query length.
536
542
  pooled_inputs = self._att_input_pool_layers[i](
@@ -541,8 +547,7 @@ class FunnelTransformerEncoder(tf.keras.layers.Layer):
541
547
  dtype=pooled_inputs.dtype), pooled_inputs),
542
548
  axis=1)
543
549
  x = layer([query_inputs, x, attention_mask],
544
- output_range=output_range if i == self._num_layers -
545
- 1 else None)
550
+ output_range=transformer_output_range)
546
551
  # Pools the corresponding attention_mask.
547
552
  if i < len(self._transformer_layers) - 1:
548
553
  attention_mask = _pool_and_concat(
@@ -245,24 +245,32 @@ class FunnelTransformerEncoderTest(parameterized.TestCase, tf.test.TestCase):
245
245
  self.assertAllEqual(tf.float32, pooled.dtype)
246
246
 
247
247
  @parameterized.named_parameters(
248
- ("all_sequence", None, 3, 0),
249
- ("output_range", 1, 1, 0),
250
- ("all_sequence_with_unpool", None, 4, 1),
251
- ("output_range_with_unpool", 1, 1, 1),
252
- ("output_range_with_large_unpool", 1, 1, 2),
248
+ ("all_sequence", None, 3, 0, 2),
249
+ ("output_range", 1, 1, 0, 2),
250
+ ("all_sequence_with_unpool", None, 4, 1, 2),
251
+ ("output_range_with_unpool", 1, 1, 1, 2),
252
+ ("output_range_with_large_unpool", 1, 1, 2, 2),
253
+ ("output_range_with_no_pooling", 1, 1, 0, 1),
254
+ ("output_range_with_unpool_and_no_pooling", 1, 1, 1, 1),
253
255
  )
254
- def test_network_invocation(self, output_range, out_seq_len, unpool_length):
256
+ def test_network_invocation(
257
+ self,
258
+ output_range,
259
+ out_seq_len,
260
+ unpool_length,
261
+ pool_stride,
262
+ ):
255
263
  hidden_size = 32
256
264
  sequence_length = 21
257
265
  vocab_size = 57
258
266
  num_types = 7
259
- pool_stride = 2
267
+ num_layers = 3
260
268
  # Create a small FunnelTransformerEncoder for testing.
261
269
  test_network = funnel_transformer.FunnelTransformerEncoder(
262
270
  vocab_size=vocab_size,
263
271
  hidden_size=hidden_size,
264
272
  num_attention_heads=2,
265
- num_layers=3,
273
+ num_layers=num_layers,
266
274
  type_vocab_size=num_types,
267
275
  pool_stride=pool_stride,
268
276
  unpool_length=unpool_length)
@@ -297,7 +305,7 @@ class FunnelTransformerEncoderTest(parameterized.TestCase, tf.test.TestCase):
297
305
  hidden_size=hidden_size,
298
306
  max_sequence_length=max_sequence_length,
299
307
  num_attention_heads=2,
300
- num_layers=3,
308
+ num_layers=num_layers,
301
309
  type_vocab_size=num_types,
302
310
  pool_stride=pool_stride)
303
311
  dict_outputs = test_network([word_ids, mask, type_ids])
@@ -305,7 +313,10 @@ class FunnelTransformerEncoderTest(parameterized.TestCase, tf.test.TestCase):
305
313
  pooled = dict_outputs["pooled_output"]
306
314
  model = tf.keras.Model([word_ids, mask, type_ids], [data, pooled])
307
315
  outputs = model.predict([word_id_data, mask_data, type_id_data])
308
- self.assertEqual(outputs[0].shape[1], 3)
316
+ expected_sequence_length = float(sequence_length)
317
+ for _ in range(num_layers):
318
+ expected_sequence_length = np.ceil(expected_sequence_length / pool_stride)
319
+ self.assertEqual(outputs[0].shape[1], expected_sequence_length)
309
320
 
310
321
  # Creates a FunnelTransformerEncoder with embedding_width != hidden_size
311
322
  test_network = funnel_transformer.FunnelTransformerEncoder(
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: tf-models-nightly
3
- Version: 2.13.1.dev20230811
3
+ Version: 2.13.1.dev20230813
4
4
  Summary: TensorFlow Official Models
5
5
  Home-page: https://github.com/tensorflow/models
6
6
  Author: Google Inc.
@@ -12,7 +12,7 @@ official/core/actions_test.py,sha256=NfT4ywnRcQNFY7i4xm_uE9qPinON-20lidBkkRYVEuo
12
12
  official/core/base_task.py,sha256=6zy7OldZtiYvH2o4TPwzkxqZd763u5P2tXN8H9aeC30,12958
13
13
  official/core/base_trainer.py,sha256=n_dxBOifRBD7G24W_8N2_Jj9kHUVSadvRaB9eEGA0pk,18115
14
14
  official/core/base_trainer_test.py,sha256=jijBCetnQgVl0yDSUq8NbtW8eOn2PfStkapW5Z1AZSs,13011
15
- official/core/config_definitions.py,sha256=cGUHTe0NaFnxKpZAUBWpEvraJyJdkKlWa-3-hKySG3w,15333
15
+ official/core/config_definitions.py,sha256=yYwFEQJgu_zCx6DJZSywi0IBrs1WbPpWazU2sn2G-6s,15403
16
16
  official/core/exp_factory.py,sha256=PzCIJEDseAhPUGwEOq6ca3CvNyJSWGDIb53YgI_Jw6s,1115
17
17
  official/core/export_base.py,sha256=V4BKNiBzzMcHTgjYTUXQgQwoZNZbCJbXWf53-_C_2ws,7025
18
18
  official/core/export_base_test.py,sha256=npqecqCdi2ovmuCWa3fBHCeST-oAvY4zgKg7BCEzWo8,4426
@@ -401,8 +401,8 @@ official/nlp/modeling/networks/encoder_scaffold.py,sha256=lHVTrsiXtS6KpodRinQ9MO
401
401
  official/nlp/modeling/networks/encoder_scaffold_test.py,sha256=ENzW6quIUE5Py-x6PtCbk0H4CaUZBwNb9HINTizV6tM,29056
402
402
  official/nlp/modeling/networks/fnet.py,sha256=fZ2xTvUilybR7BfiSBxj_5BRXe9qQwLo-BqWfyjRDxQ,14576
403
403
  official/nlp/modeling/networks/fnet_test.py,sha256=QuwBd5QsiCzTbqMkPsLSJ9g8QRWIGHQSpmAFCzTRc8Y,4543
404
- official/nlp/modeling/networks/funnel_transformer.py,sha256=1V7hkknqyJfHVItXRcnSXgxQyRcnHFTDrlNAeSsZnCo,23970
405
- official/nlp/modeling/networks/funnel_transformer_test.py,sha256=xa5gX6cfTwx0zTvJeA8i5jurCYEKv1Wpz2iEH84vIgE,17167
404
+ official/nlp/modeling/networks/funnel_transformer.py,sha256=uC0oqJxeqAyRWkX4h9nXIyyBYOXYzLloxIUDnkIdhr8,24117
405
+ official/nlp/modeling/networks/funnel_transformer_test.py,sha256=EchGCgslROR3qbpWGYmaATf4s9dQQ5lOWi5wz5TU1qI,17552
406
406
  official/nlp/modeling/networks/mobile_bert_encoder.py,sha256=swq2x5cpQUrh7yk77c7kGIzTX_xgG2ebGFQaqTU1vAM,7533
407
407
  official/nlp/modeling/networks/mobile_bert_encoder_test.py,sha256=Da8r3YPzs0G-YJF3ooQkhO7NFudL6hkAgl9cebpqHUs,7105
408
408
  official/nlp/modeling/networks/packed_sequence_embedding.py,sha256=4CGzp27Bw9Ia2cTCfjjK1Wqa_cLfmWFSiZfj4uGQJZ8,12800
@@ -1110,9 +1110,9 @@ tensorflow_models/__init__.py,sha256=Ciz_YBke6teb6y42QyQTUBDdXJAiV7Qdu1zOoZvYiKw
1110
1110
  tensorflow_models/tensorflow_models_test.py,sha256=Kz2y4V-rtBhZFFfKD2soCq52hviSfJVV1L2ztqS-9oM,1385
1111
1111
  tensorflow_models/nlp/__init__.py,sha256=3dULDpUBpDi9vljpXadq6oJrWH4y6z42Bz2d3hopYZw,807
1112
1112
  tensorflow_models/vision/__init__.py,sha256=4y77XkHaH8qLls3-6ta4tMp3Xj8CLbB0ihH91HsQ9z4,833
1113
- tf_models_nightly-2.13.1.dev20230811.dist-info/AUTHORS,sha256=1dG3fXVu9jlo7bul8xuix5F5vOnczMk7_yWn4y70uw0,337
1114
- tf_models_nightly-2.13.1.dev20230811.dist-info/LICENSE,sha256=WxeBS_DejPZQabxtfMOM_xn8qoZNJDQjrT7z2wG1I4U,11512
1115
- tf_models_nightly-2.13.1.dev20230811.dist-info/METADATA,sha256=-4PzH_YIWJQ7bNpCt3ehGSyCHkHPD37-D6GwGHkuz_s,1395
1116
- tf_models_nightly-2.13.1.dev20230811.dist-info/WHEEL,sha256=kGT74LWyRUZrL4VgLh6_g12IeVl_9u9ZVhadrgXZUEY,110
1117
- tf_models_nightly-2.13.1.dev20230811.dist-info/top_level.txt,sha256=gum2FfO5R4cvjl2-QtP-S1aNmsvIZaFFT6VFzU0f4-g,33
1118
- tf_models_nightly-2.13.1.dev20230811.dist-info/RECORD,,
1113
+ tf_models_nightly-2.13.1.dev20230813.dist-info/AUTHORS,sha256=1dG3fXVu9jlo7bul8xuix5F5vOnczMk7_yWn4y70uw0,337
1114
+ tf_models_nightly-2.13.1.dev20230813.dist-info/LICENSE,sha256=WxeBS_DejPZQabxtfMOM_xn8qoZNJDQjrT7z2wG1I4U,11512
1115
+ tf_models_nightly-2.13.1.dev20230813.dist-info/METADATA,sha256=OF5TjgUSsS-GtDUKpZDZDupLyW2k2YTh4Cai9xeELLo,1395
1116
+ tf_models_nightly-2.13.1.dev20230813.dist-info/WHEEL,sha256=kGT74LWyRUZrL4VgLh6_g12IeVl_9u9ZVhadrgXZUEY,110
1117
+ tf_models_nightly-2.13.1.dev20230813.dist-info/top_level.txt,sha256=gum2FfO5R4cvjl2-QtP-S1aNmsvIZaFFT6VFzU0f4-g,33
1118
+ tf_models_nightly-2.13.1.dev20230813.dist-info/RECORD,,