TF: add test for PushToHubCallback (#20231)

* test hub tf callback

* create repo before cloning it

src/transformers/keras_callbacks.py

@@ -9,7 +9,7 @@ import tensorflow as tf
 from packaging.version import parse
 from tensorflow.keras.callbacks import Callback
 
-from huggingface_hub import Repository
+from huggingface_hub import Repository, create_repo
 
 from . import IntervalStrategy, PreTrainedTokenizerBase
 from .modelcard import TrainingSummary
@@ -339,11 +339,13 @@ class PushToHubCallback(Callback):
 
         self.output_dir = output_dir
         self.hub_model_id = hub_model_id
+        create_repo(self.hub_model_id, exist_ok=True)
         self.repo = Repository(
             str(self.output_dir),
             clone_from=self.hub_model_id,
             use_auth_token=hub_token if hub_token else True,
         )
+
         self.tokenizer = tokenizer
         self.last_job = None
         self.checkpoint = checkpoint
@@ -394,15 +396,20 @@ class PushToHubCallback(Callback):
             )
 
     def on_train_end(self, logs=None):
+        # Makes sure the latest version of the model is uploaded
         if self.last_job is not None and not self.last_job.is_done:
-            self.last_job._process.terminate()  # Gotta go fast
+            logging.info("Pushing the last epoch to the Hub, this may take a while...")
             while not self.last_job.is_done:
                 sleep(1)
+        else:
             self.model.save_pretrained(self.output_dir)
             if self.tokenizer is not None:
                 self.tokenizer.save_pretrained(self.output_dir)
             train_summary = TrainingSummary.from_keras(
-            model=self.model, model_name=self.hub_model_id, keras_history=self.training_history, **self.model_card_args
+                model=self.model,
+                model_name=self.hub_model_id,
+                keras_history=self.training_history,
+                **self.model_card_args,
             )
             model_card = train_summary.to_model_card()
             with (self.output_dir / "README.md").open("w") as f:

src/transformers/models/blenderbot/tokenization_blenderbot.py

@@ -399,6 +399,7 @@ class BlenderbotTokenizer(PreTrainedTokenizer):
         Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
         adding special tokens. A Blenderbot sequence has the following format:
         - single sequence: ` X </s>`
+
         Args:
             token_ids_0 (`List[int]`):
                 List of IDs to which the special tokens will be added

src/transformers/models/blenderbot/tokenization_blenderbot_fast.py

@@ -284,6 +284,7 @@ class BlenderbotTokenizerFast(PreTrainedTokenizerFast):
         Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
         adding special tokens. A Blenderbot sequence has the following format:
         - single sequence: ` X </s>`
+
         Args:
             token_ids_0 (`List[int]`):
                 List of IDs to which the special tokens will be added

src/transformers/models/markuplm/tokenization_markuplm.py

@@ -428,6 +428,7 @@ class MarkupLMTokenizer(PreTrainedTokenizer):
         adding special tokens. A RoBERTa sequence has the following format:
         - single sequence: `<s> X </s>`
         - pair of sequences: `<s> A </s></s> B </s>`
+
         Args:
             token_ids_0 (`List[int]`):
                 List of IDs to which the special tokens will be added.

src/transformers/models/markuplm/tokenization_markuplm_fast.py

@@ -883,6 +883,7 @@ class MarkupLMTokenizerFast(PreTrainedTokenizerFast):
         adding special tokens. A RoBERTa sequence has the following format:
         - single sequence: `<s> X </s>`
         - pair of sequences: `<s> A </s></s> B </s>`
+
         Args:
             token_ids_0 (`List[int]`):
                 List of IDs to which the special tokens will be added.

src/transformers/models/tapex/tokenization_tapex.py

@@ -342,6 +342,7 @@ class TapexTokenizer(PreTrainedTokenizer):
         adding special tokens. A TAPEX sequence has the following format:
         - single sequence: `<s> X </s>`
         - pair of sequences: `<s> A </s></s> B </s>`
+
         Args:
             token_ids_0 (`List[int]`):
                 List of IDs to which the special tokens will be added.

tests/test_modeling_tf_common.py

@@ -78,9 +78,11 @@ if is_tf_available():
         TF_MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING,
         TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING,
         BertConfig,
+        PushToHubCallback,
         RagRetriever,
         TFAutoModel,
         TFAutoModelForSequenceClassification,
+        TFBertForMaskedLM,
         TFBertModel,
         TFRagModel,
         TFSharedEmbeddings,
@@ -2359,6 +2361,11 @@ class TFModelPushToHubTester(unittest.TestCase):
         except HTTPError:
             pass
 
+        try:
+            delete_repo(token=cls._token, repo_id="test-model-tf-callback")
+        except HTTPError:
+            pass
+
         try:
             delete_repo(token=cls._token, repo_id="valid_org/test-model-tf-org")
         except HTTPError:
@@ -2378,13 +2385,14 @@ class TFModelPushToHubTester(unittest.TestCase):
             model.push_to_hub("test-model-tf", use_auth_token=self._token)
         logging.set_verbosity_warning()
         # Check the model card was created and uploaded.
-        self.assertIn("Uploading README.md to __DUMMY_TRANSFORMERS_USER__/test-model-tf", cl.out)
+        self.assertIn("Uploading the following files to __DUMMY_TRANSFORMERS_USER__/test-model-tf", cl.out)
 
         new_model = TFBertModel.from_pretrained(f"{USER}/test-model-tf")
         models_equal = True
         for p1, p2 in zip(model.weights, new_model.weights):
-            if tf.math.reduce_sum(tf.math.abs(p1 - p2)) > 0:
+            if not tf.math.reduce_all(p1 == p2):
                 models_equal = False
+                break
         self.assertTrue(models_equal)
 
         # Reset repo
@@ -2397,8 +2405,32 @@ class TFModelPushToHubTester(unittest.TestCase):
         new_model = TFBertModel.from_pretrained(f"{USER}/test-model-tf")
         models_equal = True
         for p1, p2 in zip(model.weights, new_model.weights):
-            if tf.math.reduce_sum(tf.math.abs(p1 - p2)) > 0:
+            if not tf.math.reduce_all(p1 == p2):
                 models_equal = False
+                break
+        self.assertTrue(models_equal)
+
+    def test_push_to_hub_callback(self):
+        config = BertConfig(
+            vocab_size=99, hidden_size=32, num_hidden_layers=5, num_attention_heads=4, intermediate_size=37
+        )
+        model = TFBertForMaskedLM(config)
+        model.compile()
+
+        with tempfile.TemporaryDirectory() as tmp_dir:
+            push_to_hub_callback = PushToHubCallback(
+                output_dir=tmp_dir,
+                hub_model_id="test-model-tf-callback",
+                hub_token=self._token,
+            )
+            model.fit(model.dummy_inputs, model.dummy_inputs, epochs=1, callbacks=[push_to_hub_callback])
+
+        new_model = TFBertForMaskedLM.from_pretrained(f"{USER}/test-model-tf-callback")
+        models_equal = True
+        for p1, p2 in zip(model.weights, new_model.weights):
+            if not tf.math.reduce_all(p1 == p2):
+                models_equal = False
+                break
         self.assertTrue(models_equal)
 
     def test_push_to_hub_in_organization(self):
@@ -2414,8 +2446,9 @@ class TFModelPushToHubTester(unittest.TestCase):
         new_model = TFBertModel.from_pretrained("valid_org/test-model-tf-org")
         models_equal = True
         for p1, p2 in zip(model.weights, new_model.weights):
-            if tf.math.reduce_sum(tf.math.abs(p1 - p2)) > 0:
+            if not tf.math.reduce_all(p1 == p2):
                 models_equal = False
+                break
         self.assertTrue(models_equal)
 
         # Reset repo
@@ -2430,6 +2463,7 @@ class TFModelPushToHubTester(unittest.TestCase):
         new_model = TFBertModel.from_pretrained("valid_org/test-model-tf-org")
         models_equal = True
         for p1, p2 in zip(model.weights, new_model.weights):
-            if tf.math.reduce_sum(tf.math.abs(p1 - p2)) > 0:
+            if not tf.math.reduce_all(p1 == p2):
                 models_equal = False
+                break
         self.assertTrue(models_equal)