Allow `trust_remote_code` in example scripts (#25248)

* pytorch examples

* pytorch mim no trainer

* cookiecutter

* flax examples

* missed line in pytorch run_glue

* tensorflow examples

* tensorflow run_clip

* tensorflow run_mlm

* tensorflow run_ner

* tensorflow run_clm

* pytorch example from_configs

* pytorch no trainer examples

* Revert "tensorflow run_clip"

This reverts commit 261f86ac1f1c9e05dd3fd0291e1a1f8e573781d5.

* fix: duplicated argument

examples/tensorflow/language-modeling/run_mlm.py

@@ -126,6 +126,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
     def __post_init__(self):
         if self.config_overrides is not None and (self.config_name is not None or self.model_name_or_path is not None):
@@ -348,19 +358,25 @@ def main():
     # In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
     # download model & vocab.
     if checkpoint is not None:
-        config = AutoConfig.from_pretrained(checkpoint)
+        config = AutoConfig.from_pretrained(checkpoint, trust_remote_code=model_args.trust_remote_code)
     elif model_args.config_name:
-        config = AutoConfig.from_pretrained(model_args.config_name)
+        config = AutoConfig.from_pretrained(model_args.config_name, trust_remote_code=model_args.trust_remote_code)
     elif model_args.model_name_or_path:
-        config = AutoConfig.from_pretrained(model_args.model_name_or_path)
+        config = AutoConfig.from_pretrained(
+            model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code
+        )
     else:
         config = CONFIG_MAPPING[model_args.model_type]()
         logger.warning("You are instantiating a new config instance from scratch.")
 
     if model_args.tokenizer_name:
-        tokenizer = AutoTokenizer.from_pretrained(model_args.tokenizer_name)
+        tokenizer = AutoTokenizer.from_pretrained(
+            model_args.tokenizer_name, trust_remote_code=model_args.trust_remote_code
+        )
     elif model_args.model_name_or_path:
-        tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path)
+        tokenizer = AutoTokenizer.from_pretrained(
+            model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code
+        )
     else:
         raise ValueError(
             "You are instantiating a new tokenizer from scratch. This is not supported by this script."
@@ -495,12 +511,16 @@ def main():
     with training_args.strategy.scope():
         # region Prepare model
         if checkpoint is not None:
-            model = TFAutoModelForMaskedLM.from_pretrained(checkpoint, config=config)
+            model = TFAutoModelForMaskedLM.from_pretrained(
+                checkpoint, config=config, trust_remote_code=model_args.trust_remote_code
+            )
         elif model_args.model_name_or_path:
-            model = TFAutoModelForMaskedLM.from_pretrained(model_args.model_name_or_path, config=config)
+            model = TFAutoModelForMaskedLM.from_pretrained(
+                model_args.model_name_or_path, config=config, trust_remote_code=model_args.trust_remote_code
+            )
         else:
             logger.info("Training new model from scratch")
-            model = TFAutoModelForMaskedLM.from_config(config)
+            model = TFAutoModelForMaskedLM.from_config(config, trust_remote_code=model_args.trust_remote_code)
 
         # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch
         # on a small vocab and want a smaller embedding size, remove this test.

examples/tensorflow/multiple-choice/run_swag.py

@@ -162,6 +162,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 @dataclass
@@ -349,6 +359,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
@@ -356,6 +367,7 @@ def main():
         use_fast=model_args.use_fast_tokenizer,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     # endregion
 
@@ -442,6 +454,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
 
         num_replicas = training_args.strategy.num_replicas_in_sync

examples/tensorflow/question-answering/run_qa.py

@@ -93,6 +93,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 @dataclass
@@ -352,6 +362,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
@@ -359,6 +370,7 @@ def main():
         use_fast=True,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     # endregion
 
@@ -639,6 +651,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
         if training_args.do_train:
             training_dataset = model.prepare_tf_dataset(

examples/tensorflow/summarization/run_summarization.py

@@ -115,6 +115,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 @dataclass
@@ -402,6 +412,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
@@ -409,6 +420,7 @@ def main():
         use_fast=model_args.use_fast_tokenizer,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
 
     prefix = data_args.source_prefix if data_args.source_prefix is not None else ""
@@ -527,6 +539,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
 
         # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch

examples/tensorflow/text-classification/run_glue.py

@@ -180,6 +180,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 # endregion
@@ -298,6 +308,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
@@ -305,6 +316,7 @@ def main():
         use_fast=model_args.use_fast_tokenizer,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     # endregion
 
@@ -388,6 +400,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
         # endregion
 

examples/tensorflow/text-classification/run_text_classification.py

@@ -186,6 +186,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 # endregion
@@ -315,6 +325,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
     else:
         config = AutoConfig.from_pretrained(
@@ -322,12 +333,14 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     # endregion
 
@@ -416,6 +429,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
         # endregion
 

examples/tensorflow/token-classification/run_ner.py

@@ -91,6 +91,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 @dataclass
@@ -304,9 +314,17 @@ def main():
     # In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
     # download model & vocab.
     if model_args.config_name:
-        config = AutoConfig.from_pretrained(model_args.config_name, num_labels=num_labels)
+        config = AutoConfig.from_pretrained(
+            model_args.config_name,
+            num_labels=num_labels,
+            trust_remote_code=model_args.trust_remote_code,
+        )
     elif model_args.model_name_or_path:
-        config = AutoConfig.from_pretrained(model_args.model_name_or_path, num_labels=num_labels)
+        config = AutoConfig.from_pretrained(
+            model_args.model_name_or_path,
+            num_labels=num_labels,
+            trust_remote_code=model_args.trust_remote_code,
+        )
     else:
         config = CONFIG_MAPPING[model_args.model_type]()
         logger.warning("You are instantiating a new config instance from scratch.")
@@ -319,9 +337,18 @@ def main():
         )
 
     if config.model_type in {"gpt2", "roberta"}:
-        tokenizer = AutoTokenizer.from_pretrained(tokenizer_name_or_path, use_fast=True, add_prefix_space=True)
+        tokenizer = AutoTokenizer.from_pretrained(
+            tokenizer_name_or_path,
+            use_fast=True,
+            add_prefix_space=True,
+            trust_remote_code=model_args.trust_remote_code,
+        )
     else:
-        tokenizer = AutoTokenizer.from_pretrained(tokenizer_name_or_path, use_fast=True)
+        tokenizer = AutoTokenizer.from_pretrained(
+            tokenizer_name_or_path,
+            use_fast=True,
+            trust_remote_code=model_args.trust_remote_code,
+        )
     # endregion
 
     # region Preprocessing the raw datasets
@@ -392,10 +419,13 @@ def main():
             model = TFAutoModelForTokenClassification.from_pretrained(
                 model_args.model_name_or_path,
                 config=config,
+                trust_remote_code=model_args.trust_remote_code,
             )
         else:
             logger.info("Training new model from scratch")
-            model = TFAutoModelForTokenClassification.from_config(config)
+            model = TFAutoModelForTokenClassification.from_config(
+                config, trust_remote_code=model_args.trust_remote_code
+            )
 
         # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch
         # on a small vocab and want a smaller embedding size, remove this test.

examples/tensorflow/translation/run_translation.py

@@ -109,6 +109,16 @@ class ModelArguments:
             "help": "The `use_auth_token` argument is deprecated and will be removed in v4.34. Please use `token`."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 
 
 @dataclass
@@ -366,6 +376,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
@@ -373,6 +384,7 @@ def main():
         use_fast=model_args.use_fast_tokenizer,
         revision=model_args.model_revision,
         token=model_args.token,
+        trust_remote_code=model_args.trust_remote_code,
     )
 
     prefix = data_args.source_prefix if data_args.source_prefix is not None else ""
@@ -480,6 +492,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=model_args.token,
+            trust_remote_code=model_args.trust_remote_code,
         )
 
         # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch

templates/adding_a_new_example_script/{{cookiecutter.directory_name}}/run_{{cookiecutter.example_shortcut}}.py

@@ -122,6 +122,16 @@ class ModelArguments:
             "with private models)."
         },
     )
+    trust_remote_code: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether or not to allow for custom models defined on the Hub in their own modeling files. This option"
+                "should only be set to `True` for repositories you trust and in which you have read the code, as it will"
+                "execute code present on the Hub on your local machine."
+            )
+        },
+    )
 {% endif %}
 
 
@@ -290,6 +300,7 @@ def main():
         "cache_dir": model_args.cache_dir,
         "revision": model_args.model_revision,
         "token": True if model_args.token else None,
+        "trust_remote_code": model_args.trust_remote_code,
     }
     if model_args.config_name:
         config = AutoConfig.from_pretrained(model_args.config_name, **config_kwargs)
@@ -304,6 +315,7 @@ def main():
         "use_fast": model_args.use_fast_tokenizer,
         "revision": model_args.model_revision,
         "token": True if model_args.token else None,
+        "trust_remote_code": model_args.trust_remote_code,
     }
     if model_args.tokenizer_name:
         tokenizer = AutoTokenizer.from_pretrained(model_args.tokenizer_name, **tokenizer_kwargs)
@@ -323,6 +335,7 @@ def main():
             cache_dir=model_args.cache_dir,
             revision=model_args.model_revision,
             token=True if model_args.token else None,
+            trust_remote_code=model_args.trust_remote_code,
         )
     else:
         logger.info("Training new model from scratch")
@@ -337,6 +350,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=True if model_args.token else None,
+        trust_remote_code=model_args.trust_remote_code,
     )
     tokenizer = AutoTokenizer.from_pretrained(
         model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
@@ -344,6 +358,7 @@ def main():
         use_fast=model_args.use_fast_tokenizer,
         revision=model_args.model_revision,
         token=True if model_args.token else None,
+        trust_remote_code=model_args.trust_remote_code,
     )
     model = AutoModelForSequenceClassification.from_pretrained(
         model_args.model_name_or_path,
@@ -352,6 +367,7 @@ def main():
         cache_dir=model_args.cache_dir,
         revision=model_args.model_revision,
         token=True if model_args.token else None,
+        trust_remote_code=model_args.trust_remote_code,
     )
 {% endif %}