tf-models-nightly 2.19.0.dev20250129__py2.py3-none-any.whl → 2.19.0.dev20250131__py2.py3-none-any.whl

official/common/distribute_utils_test.py

@@ -43,14 +43,14 @@ class DistributeUtilsTest(tf.test.TestCase):
 
   def test_one_device_strategy_cpu(self):
     ds = distribute_utils.get_distribution_strategy('one_device', num_gpus=0)
-    self.assertEquals(ds.num_replicas_in_sync, 1)
-    self.assertEquals(len(ds.extended.worker_devices), 1)
+    self.assertEqual(ds.num_replicas_in_sync, 1)
+    self.assertEqual(len(ds.extended.worker_devices), 1)
     self.assertIn('CPU', ds.extended.worker_devices[0])
 
   def test_one_device_strategy_gpu(self):
     ds = distribute_utils.get_distribution_strategy('one_device', num_gpus=1)
-    self.assertEquals(ds.num_replicas_in_sync, 1)
-    self.assertEquals(len(ds.extended.worker_devices), 1)
+    self.assertEqual(ds.num_replicas_in_sync, 1)
+    self.assertEqual(len(ds.extended.worker_devices), 1)
     self.assertIn('GPU', ds.extended.worker_devices[0])
 
   def test_mirrored_strategy(self):
@@ -58,8 +58,8 @@ class DistributeUtilsTest(tf.test.TestCase):
     _ = distribute_utils.get_distribution_strategy(num_gpus=0)
     # 5 GPUs.
     ds = distribute_utils.get_distribution_strategy(num_gpus=5)
-    self.assertEquals(ds.num_replicas_in_sync, 5)
-    self.assertEquals(len(ds.extended.worker_devices), 5)
+    self.assertEqual(ds.num_replicas_in_sync, 5)
+    self.assertEqual(len(ds.extended.worker_devices), 5)
     for device in ds.extended.worker_devices:
       self.assertIn('GPU', device)
 
@@ -105,12 +105,13 @@ class DistributeUtilsTest(tf.test.TestCase):
         ds, tf.distribute.TPUStrategy)
 
   def test_invalid_strategy(self):
-    with self.assertRaisesRegexp(
-        ValueError,
-        'distribution_strategy must be a string but got: False. If'):
+    with self.assertRaisesRegex(
+        ValueError, 'distribution_strategy must be a string but got: False. If'
+    ):
       distribute_utils.get_distribution_strategy(False)
-    with self.assertRaisesRegexp(
-        ValueError, 'distribution_strategy must be a string but got: 1'):
+    with self.assertRaisesRegex(
+        ValueError, 'distribution_strategy must be a string but got: 1'
+    ):
       distribute_utils.get_distribution_strategy(1)
 
   def test_get_strategy_scope(self):

official/projects/detr/tasks/detection.py

@@ -13,7 +13,6 @@
 # limitations under the License.
 
 """DETR detection task definition."""
-
 from typing import Optional
 
 from absl import logging
@@ -48,25 +47,21 @@ class DetectionTask(base_task.Task):
   def build_model(self):
     """Build DETR model."""
 
-    input_specs = tf_keras.layers.InputSpec(
-        shape=[None] + self._task_config.model.input_size
-    )
+    input_specs = tf_keras.layers.InputSpec(shape=[None] +
+                                            self._task_config.model.input_size)
 
     backbone = backbones.factory.build_backbone(
         input_specs=input_specs,
         backbone_config=self._task_config.model.backbone,
-        norm_activation_config=self._task_config.model.norm_activation,
-    )
-
-    model = detr.DETR(
-        backbone,
-        self._task_config.model.backbone_endpoint_name,
-        self._task_config.model.num_queries,
-        self._task_config.model.hidden_size,
-        self._task_config.model.num_classes,
-        self._task_config.model.num_encoder_layers,
-        self._task_config.model.num_decoder_layers,
-    )
+        norm_activation_config=self._task_config.model.norm_activation)
+
+    model = detr.DETR(backbone,
+                      self._task_config.model.backbone_endpoint_name,
+                      self._task_config.model.num_queries,
+                      self._task_config.model.hidden_size,
+                      self._task_config.model.num_classes,
+                      self._task_config.model.num_encoder_layers,
+                      self._task_config.model.num_decoder_layers)
     return model
 
   def initialize(self, model: tf_keras.Model):
@@ -89,13 +84,12 @@ class DetectionTask(base_task.Task):
       status = ckpt.restore(ckpt_dir_or_file)
       status.expect_partial().assert_existing_objects_matched()
 
-    logging.info(
-        'Finished loading pretrained checkpoint from %s', ckpt_dir_or_file
-    )
+    logging.info('Finished loading pretrained checkpoint from %s',
+                 ckpt_dir_or_file)
 
-  def build_inputs(
-      self, params, input_context: Optional[tf.distribute.InputContext] = None
-  ):
+  def build_inputs(self,
+                   params,
+                   input_context: Optional[tf.distribute.InputContext] = None):
     """Build input dataset."""
     if isinstance(params, coco.COCODataConfig):
       dataset = coco.COCODataLoader(params).load(input_context)
@@ -106,17 +100,14 @@ class DetectionTask(base_task.Task):
         decoder_cfg = params.decoder.get()
         if params.decoder.type == 'simple_decoder':
           decoder = tf_example_decoder.TfExampleDecoder(
-              regenerate_source_id=decoder_cfg.regenerate_source_id
-          )
+              regenerate_source_id=decoder_cfg.regenerate_source_id)
         elif params.decoder.type == 'label_map_decoder':
           decoder = tf_example_label_map_decoder.TfExampleDecoderLabelMap(
               label_map=decoder_cfg.label_map,
-              regenerate_source_id=decoder_cfg.regenerate_source_id,
-          )
+              regenerate_source_id=decoder_cfg.regenerate_source_id)
         else:
-          raise ValueError(
-              'Unknown decoder type: {}!'.format(params.decoder.type)
-          )
+          raise ValueError('Unknown decoder type: {}!'.format(
+              params.decoder.type))
 
       parser = detr_input.Parser(
           class_offset=self._task_config.losses.class_offset,
@@ -127,8 +118,7 @@ class DetectionTask(base_task.Task):
           params,
           dataset_fn=dataset_fn.pick_dataset_fn(params.file_type),
           decoder_fn=decoder.decode,
-          parser_fn=parser.parse_fn(params.is_training),
-      )
+          parser_fn=parser.parse_fn(params.is_training))
       dataset = reader.read(input_context=input_context)
 
     return dataset
@@ -187,8 +177,7 @@ class DetectionTask(base_task.Task):
     box_targets = labels['boxes']
 
     cost = self._compute_cost(
-        cls_outputs, box_outputs, cls_targets, box_targets
-    )
+        cls_outputs, box_outputs, cls_targets, box_targets)
 
     _, indices = matchers.hungarian_matching(cost)
     indices = tf.stop_gradient(indices)
@@ -199,41 +188,31 @@ class DetectionTask(base_task.Task):
 
     background = tf.equal(cls_targets, 0)
     num_boxes = tf.reduce_sum(
-        tf.cast(tf.logical_not(background), tf.float32), axis=-1
-    )
+        tf.cast(tf.logical_not(background), tf.float32), axis=-1)
 
     # Down-weight background to account for class imbalance.
     xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
-        labels=cls_targets, logits=cls_assigned
-    )
+        labels=cls_targets, logits=cls_assigned)
     cls_loss = self._task_config.losses.lambda_cls * tf.where(
-        background,
-        self._task_config.losses.background_cls_weight * xentropy,
-        xentropy,
-    )
+        background, self._task_config.losses.background_cls_weight * xentropy,
+        xentropy)
     cls_weights = tf.where(
         background,
         self._task_config.losses.background_cls_weight * tf.ones_like(cls_loss),
-        tf.ones_like(cls_loss),
-    )
+        tf.ones_like(cls_loss))
 
     # Box loss is only calculated on non-background class.
     l_1 = tf.reduce_sum(tf.abs(box_assigned - box_targets), axis=-1)
     box_loss = self._task_config.losses.lambda_box * tf.where(
-        background, tf.zeros_like(l_1), l_1
-    )
+        background, tf.zeros_like(l_1), l_1)
 
     # Giou loss is only calculated on non-background class.
-    giou = tf.linalg.diag_part(
-        1.0
-        - box_ops.bbox_generalized_overlap(
-            box_ops.cycxhw_to_yxyx(box_assigned),
-            box_ops.cycxhw_to_yxyx(box_targets),
-        )
-    )
+    giou = tf.linalg.diag_part(1.0 - box_ops.bbox_generalized_overlap(
+        box_ops.cycxhw_to_yxyx(box_assigned),
+        box_ops.cycxhw_to_yxyx(box_targets)
+        ))
     giou_loss = self._task_config.losses.lambda_giou * tf.where(
-        background, tf.zeros_like(giou), giou
-    )
+        background, tf.zeros_like(giou), giou)
 
     # Consider doing all reduce once in train_step to speed up.
     num_boxes_per_replica = tf.reduce_sum(num_boxes)
@@ -241,11 +220,13 @@ class DetectionTask(base_task.Task):
     replica_context = tf.distribute.get_replica_context()
     num_boxes_sum, cls_weights_sum = replica_context.all_reduce(
         tf.distribute.ReduceOp.SUM,
-        [num_boxes_per_replica, cls_weights_per_replica],
-    )
-    cls_loss = tf.math.divide_no_nan(tf.reduce_sum(cls_loss), cls_weights_sum)
-    box_loss = tf.math.divide_no_nan(tf.reduce_sum(box_loss), num_boxes_sum)
-    giou_loss = tf.math.divide_no_nan(tf.reduce_sum(giou_loss), num_boxes_sum)
+        [num_boxes_per_replica, cls_weights_per_replica])
+    cls_loss = tf.math.divide_no_nan(
+        tf.reduce_sum(cls_loss), cls_weights_sum)
+    box_loss = tf.math.divide_no_nan(
+        tf.reduce_sum(box_loss), num_boxes_sum)
+    giou_loss = tf.math.divide_no_nan(
+        tf.reduce_sum(giou_loss), num_boxes_sum)
 
     aux_losses = tf.add_n(aux_losses) if aux_losses else 0.0
 
@@ -264,8 +245,7 @@ class DetectionTask(base_task.Task):
           annotation_file=self._task_config.annotation_file,
           include_mask=False,
           need_rescale_bboxes=True,
-          per_category_metrics=self._task_config.per_category_metrics,
-      )
+          per_category_metrics=self._task_config.per_category_metrics)
     return metrics
 
   def train_step(self, inputs, model, optimizer, metrics=None):
@@ -355,8 +335,7 @@ class DetectionTask(base_task.Task):
 
     outputs = model(features, training=False)[-1]
     loss, cls_loss, box_loss, giou_loss = self.build_losses(
-        outputs=outputs, labels=labels, aux_losses=model.losses
-    )
+        outputs=outputs, labels=labels, aux_losses=model.losses)
 
     # Multiply for logging.
     # Since we expect the gradient replica sum to happen in the optimizer,
@@ -374,33 +353,25 @@ class DetectionTask(base_task.Task):
     # This is for backward compatibility.
     if 'detection_boxes' not in outputs:
       detection_boxes = box_ops.cycxhw_to_yxyx(
-          outputs['box_outputs']
-      ) * tf.expand_dims(
-          tf.concat(
-              [
-                  labels['image_info'][:, 1:2, 0],
-                  labels['image_info'][:, 1:2, 1],
-                  labels['image_info'][:, 1:2, 0],
-                  labels['image_info'][:, 1:2, 1],
+          outputs['box_outputs']) * tf.expand_dims(
+              tf.concat([
+                  labels['image_info'][:, 1:2, 0], labels['image_info'][:, 1:2,
+                                                                        1],
+                  labels['image_info'][:, 1:2, 0], labels['image_info'][:, 1:2,
+                                                                        1]
               ],
-              axis=1,
-          ),
-          axis=1,
-      )
+                        axis=1),
+              axis=1)
     else:
       detection_boxes = outputs['detection_boxes']
 
-    if 'detection_scores' not in outputs:
-      detection_scores = tf.math.reduce_max(
-          tf.nn.softmax(outputs['cls_outputs'])[:, :, 1:], axis=-1
-      )
-    else:
-      detection_scores = outputs['detection_scores']
+    detection_scores = tf.math.reduce_max(
+        tf.nn.softmax(outputs['cls_outputs'])[:, :, 1:], axis=-1
+    ) if 'detection_scores' not in outputs else outputs['detection_scores']
 
     if 'detection_classes' not in outputs:
-      detection_classes = (
-          tf.math.argmax(outputs['cls_outputs'][:, :, 1:], axis=-1) + 1
-      )
+      detection_classes = tf.math.argmax(
+          outputs['cls_outputs'][:, :, 1:], axis=-1) + 1
     else:
       detection_classes = outputs['detection_classes']
 
@@ -408,12 +379,9 @@ class DetectionTask(base_task.Task):
       num_detections = tf.reduce_sum(
           tf.cast(
               tf.math.greater(
-                  tf.math.reduce_max(outputs['cls_outputs'], axis=-1), 0
-              ),
-              tf.int32,
-          ),
-          axis=-1,
-      )
+                  tf.math.reduce_max(outputs['cls_outputs'], axis=-1), 0),
+              tf.int32),
+          axis=-1)
     else:
       num_detections = outputs['num_detections']
 
@@ -423,7 +391,7 @@ class DetectionTask(base_task.Task):
         'detection_classes': detection_classes,
         'num_detections': num_detections,
         'source_id': labels['id'],
-        'image_info': labels['image_info'],
+        'image_info': labels['image_info']
     }
 
     ground_truths = {
@@ -431,13 +399,13 @@ class DetectionTask(base_task.Task):
         'height': labels['image_info'][:, 0:1, 0],
         'width': labels['image_info'][:, 0:1, 1],
         'num_detections': tf.reduce_sum(
-            tf.cast(tf.math.greater(labels['classes'], 0), tf.int32), axis=-1
-        ),
+            tf.cast(tf.math.greater(labels['classes'], 0), tf.int32), axis=-1),
         'boxes': labels['gt_boxes'],
         'classes': labels['classes'],
-        'is_crowds': labels['is_crowd'],
+        'is_crowds': labels['is_crowd']
     }
-    logs.update({'predictions': predictions, 'ground_truths': ground_truths})
+    logs.update({'predictions': predictions,
+                 'ground_truths': ground_truths})
 
     all_losses = {
         'cls_loss': cls_loss,
@@ -457,8 +425,8 @@ class DetectionTask(base_task.Task):
       state = self.coco_metric
 
     state.update_state(
-        step_outputs['ground_truths'], step_outputs['predictions']
-    )
+        step_outputs['ground_truths'],
+        step_outputs['predictions'])
     return state
 
   def reduce_aggregated_logs(self, aggregated_logs, global_step=None):

official/recommendation/data_test.py

@@ -190,7 +190,7 @@ class BaseTest(tf.test.TestCase):
         train_examples[l].add((u_raw, i_raw))
         counts[(u_raw, i_raw)] += 1
 
-    self.assertRegexpMatches(md5.hexdigest(), END_TO_END_TRAIN_MD5)
+    self.assertRegex(md5.hexdigest(), END_TO_END_TRAIN_MD5)
 
     num_positives_seen = len(train_examples[True])
     self.assertEqual(producer._train_pos_users.shape[0], num_positives_seen)
@@ -254,7 +254,7 @@ class BaseTest(tf.test.TestCase):
           # from the negatives.
           assert (u_raw, i_raw) not in self.seen_pairs
 
-    self.assertRegexpMatches(md5.hexdigest(), END_TO_END_EVAL_MD5)
+    self.assertRegex(md5.hexdigest(), END_TO_END_EVAL_MD5)
 
   def _test_fresh_randomness(self, constructor_type):
     train_epochs = 5
@@ -300,7 +300,7 @@ class BaseTest(tf.test.TestCase):
         else:
           negative_counts[(u, i)] += 1
 
-    self.assertRegexpMatches(md5.hexdigest(), FRESH_RANDOMNESS_MD5)
+    self.assertRegex(md5.hexdigest(), FRESH_RANDOMNESS_MD5)
 
     # The positive examples should appear exactly once each epoch
     self.assertAllEqual(

official/utils/misc/model_helpers_test.py

@@ -82,7 +82,7 @@ class SyntheticDataTest(tf.test.TestCase):
       for n in range(5):
         inp, lab = sess.run((input_element, label_element))
         self.assertAllClose(inp, [123., 123., 123., 123., 123.])
-        self.assertEquals(lab, 456)
+        self.assertEqual(lab, 456)
 
   def test_generate_only_input_data(self):
     d = model_helpers.generate_synthetic_data(
@@ -111,7 +111,7 @@ class SyntheticDataTest(tf.test.TestCase):
     element = tf.compat.v1.data.make_one_shot_iterator(d).get_next()
     self.assertIn('a', element)
     self.assertIn('b', element)
-    self.assertEquals(len(element['b']), 2)
+    self.assertEqual(len(element['b']), 2)
     self.assertIn('c', element['b'])
     self.assertIn('d', element['b'])
     self.assertNotIn('c', element)

{tf_models_nightly-2.19.0.dev20250129.dist-info → tf_models_nightly-2.19.0.dev20250131.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: tf-models-nightly
-Version: 2.19.0.dev20250129
+Version: 2.19.0.dev20250131
 Summary: TensorFlow Official Models
 Home-page: https://github.com/tensorflow/models
 Author: Google Inc.

{tf_models_nightly-2.19.0.dev20250129.dist-info → tf_models_nightly-2.19.0.dev20250131.dist-info}/RECORD

@@ -2,7 +2,7 @@ official/__init__.py,sha256=7oiypy0N82PDw9aSdcJBLVoGTd_oRSUOdvuJhMv4leQ,609
 official/common/__init__.py,sha256=Yqe5TjLrAR68lQ5G781dwlbbT6mELRByDipCtr4jQY4,610
 official/common/dataset_fn.py,sha256=MSRNhNGvMnZtjkuxFIZ90vGJgKlr2JeGaEYpz82Y3Ew,1858
 official/common/distribute_utils.py,sha256=dEciUuyBsxfv6T9rvPk_5zcvLyDFv0lPkhkFvyOGe_0,9575
-official/common/distribute_utils_test.py,sha256=ltR8wrIbfoX1miJElPQzZN2aioKX6Hx8-mfzoo-_WGc,4914
+official/common/distribute_utils_test.py,sha256=KSJ8zhiWrbmXzgoNC86gbwXGjnjoxtxou4S3_VFyilw,4908
 official/common/flags.py,sha256=rtCyljUH5M0p5a1CvBnj6tdubZe4XkwzqiOz1ML6F7I,5347
 official/common/registry_imports.py,sha256=ez-j0sOvVyYBHHGM5hNjTF0jqHUbHI0zv-4SHe_AZ4s,843
 official/common/streamz_counters.py,sha256=OtdlnI5aosH_DWP8jx2MGEpjs5AhyMztr5E5qhrC4nE,1057
@@ -555,7 +555,7 @@ official/projects/detr/serving/export_module.py,sha256=3m9LXFG8vVjHv5QZVf4lyrR7L
 official/projects/detr/serving/export_module_test.py,sha256=Y_1tGLg7rKQN7FgmE681m7QUbu2L3pa9pnNkqjQqaMQ,3497
 official/projects/detr/serving/export_saved_model.py,sha256=AuYQCrnWd7n3nupIgAYwtzLtWw0w4QmRZV8GtEOKxGk,3995
 official/projects/detr/tasks/__init__.py,sha256=7oiypy0N82PDw9aSdcJBLVoGTd_oRSUOdvuJhMv4leQ,609
-official/projects/detr/tasks/detection.py,sha256=UHBhtIQktgvg4MqRGHD3PEdQr7om7cjPdAbKQDrl2K8,16107
+official/projects/detr/tasks/detection.py,sha256=6eUHzRYKAsQUvKRoVTkXVk5nJr7_0mpIOfgagYu1qR8,16127
 official/projects/detr/tasks/detection_test.py,sha256=Ao5Taen9-QWMIa1ruiRDXY-ieAeluS8uXjMVF5lg3Lo,6864
 official/projects/maskconver/__init__.py,sha256=7oiypy0N82PDw9aSdcJBLVoGTd_oRSUOdvuJhMv4leQ,609
 official/projects/maskconver/train.py,sha256=XDW0ArwYLcJpT137J3QgBEaoZHb-GE3rGbnKNsSVynI,1397
@@ -902,7 +902,7 @@ official/recommendation/constants.py,sha256=gCMNNJHwX0fUEabCiTJa7tV7KHoXD5wc-jMI
 official/recommendation/create_ncf_data.py,sha256=_UEOulLVbZLY1MN8FukpD0hEN5n030CpeWpUu46gK5I,3999
 official/recommendation/data_pipeline.py,sha256=lE9dc5zLgwXN8wohNTGTClUvFytQSqbmU-Tv0jQ3PMU,37216
 official/recommendation/data_preprocessing.py,sha256=haYd4hoEgTLEfgTDzS037Mbk-IluQzr6d-4RrdyDsNQ,10318
-official/recommendation/data_test.py,sha256=N5w17RumG1YYoZm0obsQUDNIsF0ZfPWT-kfTgv_BgY0,12831
+official/recommendation/data_test.py,sha256=fkPAn16YK3z3UWQRDrKOlLSvVXn2sWKSCM7-R_cMm_s,12807
 official/recommendation/movielens.py,sha256=vET5x3zc8vW9ruZatMO3gRFUdPJjeDFJtHXfWr0bOAI,9725
 official/recommendation/ncf_common.py,sha256=r9P3HH4JQl4Tu3ae3RnNrdSa2ZcH7N7ZXyxCMhJwqY0,12288
 official/recommendation/ncf_input_pipeline.py,sha256=y2DCc5IVBK3LBk9yUVWfwb2dz-DCt2T0suASZI02UxU,6974
@@ -988,7 +988,7 @@ official/utils/flags/flags_test.py,sha256=UFGiGEjyR1wCvbqjdCxhzjGWm0xLrc4kn-zkaE
 official/utils/misc/__init__.py,sha256=7oiypy0N82PDw9aSdcJBLVoGTd_oRSUOdvuJhMv4leQ,609
 official/utils/misc/keras_utils.py,sha256=AxsatfCV3Desrp48_WGxoQsfMO0uA9bcEXFXa7djRqs,7793
 official/utils/misc/model_helpers.py,sha256=E7AHQ-3ppDrtmKUFKAvaEnBAZI2TW3p8Hg8-I4Rvp3w,3370
-official/utils/misc/model_helpers_test.py,sha256=JyH9vycF4nX6PGYuK-IZe53ApvAc6MkVj7p7m8F3ckw,4559
+official/utils/misc/model_helpers_test.py,sha256=40paIH7IzeFa1OCSXUSmfMDAJYTYj3C2BDAR1t1b9fU,4557
 official/utils/testing/__init__.py,sha256=7oiypy0N82PDw9aSdcJBLVoGTd_oRSUOdvuJhMv4leQ,609
 official/utils/testing/integration.py,sha256=qo00sSG1fC65wI5nVZ6qm2Iek8CBShpMODgt2IdjPsE,2220
 official/utils/testing/mock_task.py,sha256=Lag4UZpJlhHu0nZy2HG-7Rcnkrg0Xv6mW5zj18_a7Ek,3312
@@ -1248,9 +1248,9 @@ tensorflow_models/tensorflow_models_test.py,sha256=nc6A9K53OGqF25xN5St8EiWvdVbda
 tensorflow_models/nlp/__init__.py,sha256=4tA5Pf4qaFwT-fIFOpX7x7FHJpnyJT-5UgOeFYTyMlc,807
 tensorflow_models/uplift/__init__.py,sha256=mqfa55gweOdpKoaQyid4A_4u7xw__FcQeSIF0k_pYmI,999
 tensorflow_models/vision/__init__.py,sha256=zBorY_v5xva1uI-qxhZO3Qh-Dii-Suq6wEYh6hKHDfc,833
-tf_models_nightly-2.19.0.dev20250129.dist-info/AUTHORS,sha256=1dG3fXVu9jlo7bul8xuix5F5vOnczMk7_yWn4y70uw0,337
-tf_models_nightly-2.19.0.dev20250129.dist-info/LICENSE,sha256=WxeBS_DejPZQabxtfMOM_xn8qoZNJDQjrT7z2wG1I4U,11512
-tf_models_nightly-2.19.0.dev20250129.dist-info/METADATA,sha256=95ZUPaAhvpDnBFQr2f7G5M6ehijusBpEd-EThxNmCIE,1432
-tf_models_nightly-2.19.0.dev20250129.dist-info/WHEEL,sha256=kGT74LWyRUZrL4VgLh6_g12IeVl_9u9ZVhadrgXZUEY,110
-tf_models_nightly-2.19.0.dev20250129.dist-info/top_level.txt,sha256=gum2FfO5R4cvjl2-QtP-S1aNmsvIZaFFT6VFzU0f4-g,33
-tf_models_nightly-2.19.0.dev20250129.dist-info/RECORD,,
+tf_models_nightly-2.19.0.dev20250131.dist-info/AUTHORS,sha256=1dG3fXVu9jlo7bul8xuix5F5vOnczMk7_yWn4y70uw0,337
+tf_models_nightly-2.19.0.dev20250131.dist-info/LICENSE,sha256=WxeBS_DejPZQabxtfMOM_xn8qoZNJDQjrT7z2wG1I4U,11512
+tf_models_nightly-2.19.0.dev20250131.dist-info/METADATA,sha256=8DTSuiOFEreTUsiFrAK1BugRzgjHpDPVUHLYrgK8UWA,1432
+tf_models_nightly-2.19.0.dev20250131.dist-info/WHEEL,sha256=kGT74LWyRUZrL4VgLh6_g12IeVl_9u9ZVhadrgXZUEY,110
+tf_models_nightly-2.19.0.dev20250131.dist-info/top_level.txt,sha256=gum2FfO5R4cvjl2-QtP-S1aNmsvIZaFFT6VFzU0f4-g,33
+tf_models_nightly-2.19.0.dev20250131.dist-info/RECORD,,