[Trainer] Undo #29896 (#30129)

* Undo

* Use tokenizer

* Undo data collator

docs/source/en/tasks/image_classification.md

@@ -322,7 +322,7 @@ At this point, only three steps remain:
 ...     data_collator=data_collator,
 ...     train_dataset=food["train"],
 ...     eval_dataset=food["test"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     compute_metrics=compute_metrics,
 ... )
 
@@ -418,7 +418,7 @@ and use the [PushToHubCallback](../main_classes/keras_callbacks#transformers.Pus
 >>> metric_callback = KerasMetricCallback(metric_fn=compute_metrics, eval_dataset=tf_eval_dataset)
 >>> push_to_hub_callback = PushToHubCallback(
 ...     output_dir="food_classifier",
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     save_strategy="no",
 ... )
 >>> callbacks = [metric_callback, push_to_hub_callback]

docs/source/en/tasks/object_detection.md

@@ -384,7 +384,7 @@ Finally, bring everything together, and call [`~transformers.Trainer.train`]:
 ...     args=training_args,
 ...     data_collator=collate_fn,
 ...     train_dataset=cppe5["train"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ... )
 
 >>> trainer.train()

docs/source/en/tasks/semantic_segmentation.md

@@ -642,7 +642,7 @@ and use the [`PushToHubCallback`] to upload the model:
 ...     metric_fn=compute_metrics, eval_dataset=tf_eval_dataset, batch_size=batch_size, label_cols=["labels"]
 ... )
 
->>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", image_processor=image_processor)
+>>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", tokenizer=image_processor)
 
 >>> callbacks = [metric_callback, push_to_hub_callback]
 ```

docs/source/en/tasks/video_classification.md

@@ -407,7 +407,7 @@ Then you just pass all of this along with the datasets to `Trainer`:
 ...     args,
 ...     train_dataset=train_dataset,
 ...     eval_dataset=val_dataset,
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     compute_metrics=compute_metrics,
 ...     data_collator=collate_fn,
 ... )

docs/source/es/tasks/image_classification.md

@@ -160,7 +160,7 @@ Al llegar a este punto, solo quedan tres pasos:
 ...     data_collator=data_collator,
 ...     train_dataset=food["train"],
 ...     eval_dataset=food["test"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ... )
 
 >>> trainer.train()

docs/source/ja/tasks/image_classification.md

@@ -328,7 +328,7 @@ food["test"].set_transform(preprocess_val)
 ...     data_collator=data_collator,
 ...     train_dataset=food["train"],
 ...     eval_dataset=food["test"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     compute_metrics=compute_metrics,
 ... )
 
@@ -426,7 +426,7 @@ Convert your datasets to the `tf.data.Dataset` format using the [`~datasets.Data
 >>> metric_callback = KerasMetricCallback(metric_fn=compute_metrics, eval_dataset=tf_eval_dataset)
 >>> push_to_hub_callback = PushToHubCallback(
 ...     output_dir="food_classifier",
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     save_strategy="no",
 ... )
 >>> callbacks = [metric_callback, push_to_hub_callback]

docs/source/ja/tasks/object_detection.md

@@ -376,7 +376,7 @@ DETR モデルをトレーニングできる「ラベル」。画像プロセッ
 ...     args=training_args,
 ...     data_collator=collate_fn,
 ...     train_dataset=cppe5["train"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ... )
 
 >>> trainer.train()

docs/source/ja/tasks/semantic_segmentation.md

@@ -434,7 +434,7 @@ TensorFlow でモデルを微調整するには、次の手順に従います。
 ...     metric_fn=compute_metrics, eval_dataset=tf_eval_dataset, batch_size=batch_size, label_cols=["labels"]
 ... )
 
->>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", image_processor=image_processor)
+>>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", tokenizer=image_processor)
 
 >>> callbacks = [metric_callback, push_to_hub_callback]
 ```

docs/source/ja/tasks/video_classification.md

@@ -414,7 +414,7 @@ def compute_metrics(eval_pred):
 ...     args,
 ...     train_dataset=train_dataset,
 ...     eval_dataset=val_dataset,
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     compute_metrics=compute_metrics,
 ...     data_collator=collate_fn,
 ... )

docs/source/ko/tasks/image_classification.md

@@ -321,7 +321,7 @@ food["test"].set_transform(preprocess_val)
 ...     data_collator=data_collator,
 ...     train_dataset=food["train"],
 ...     eval_dataset=food["test"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     compute_metrics=compute_metrics,
 ... )
 
@@ -417,7 +417,7 @@ TensorFlow에서 모델을 미세 조정하려면 다음 단계를 따르세요:
 >>> metric_callback = KerasMetricCallback(metric_fn=compute_metrics, eval_dataset=tf_eval_dataset)
 >>> push_to_hub_callback = PushToHubCallback(
 ...     output_dir="food_classifier",
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     save_strategy="no",
 ... )
 >>> callbacks = [metric_callback, push_to_hub_callback]

docs/source/ko/tasks/object_detection.md

@@ -366,7 +366,7 @@ DatasetDict({
 ...     args=training_args,
 ...     data_collator=collate_fn,
 ...     train_dataset=cppe5["train"],
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ... )
 
 >>> trainer.train()

docs/source/ko/tasks/semantic_segmentation.md

@@ -424,7 +424,7 @@ TensorFlow에서 모델을 미세 조정하려면 다음 단계를 따르세요:
 ...     metric_fn=compute_metrics, eval_dataset=tf_eval_dataset, batch_size=batch_size, label_cols=["labels"]
 ... )
 
->>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", image_processor=image_processor)
+>>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", tokenizer=image_processor)
 
 >>> callbacks = [metric_callback, push_to_hub_callback]
 ```

docs/source/ko/tasks/video_classification.md

@@ -411,7 +411,7 @@ def compute_metrics(eval_pred):
 ...     args,
 ...     train_dataset=train_dataset,
 ...     eval_dataset=val_dataset,
-...     image_processor=image_processor,
+...     tokenizer=image_processor,
 ...     compute_metrics=compute_metrics,
 ...     data_collator=collate_fn,
 ... )