Add many missing spaces in adjacent strings (#26751)

Add missing spaces in adjacent strings

examples/research_projects/xtreme-s/run_xtreme_s.py

@@ -116,8 +116,8 @@ class ModelArguments:
         default=0.05,
         metadata={
             "help": (
-                "Probability of each feature vector along the time axis to be chosen as the start of the vector"
-                "span to be masked. Approximately ``mask_time_prob * sequence_length // mask_time_length`` feature"
+                "Probability of each feature vector along the time axis to be chosen as the start of the vector "
+                "span to be masked. Approximately ``mask_time_prob * sequence_length // mask_time_length`` feature "
                 "vectors will be masked along the time axis."
             )
         },
@@ -455,7 +455,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
     )
     # Set the verbosity to info of the Transformers logger (on main process only):

examples/research_projects/zero-shot-distillation/distill_classifier.py

@@ -41,7 +41,7 @@ class TeacherModelArguments:
         default="This example is {}.",
         metadata={
             "help": (
-                "Template used to turn class names into mock hypotheses for teacher NLI model. Must include {{}}"
+                "Template used to turn class names into mock hypotheses for teacher NLI model. Must include {{}} "
                 "where class name is inserted."
             )
         },
@@ -53,7 +53,7 @@ class TeacherModelArguments:
         default=False,
         metadata={
             "help": (
-                "Allow multiple classes to be true rather than forcing them to sum to 1 (sometimes called"
+                "Allow multiple classes to be true rather than forcing them to sum to 1 (sometimes called "
                 "multi-class multi-label classification)."
             )
         },
@@ -98,7 +98,7 @@ class DistillTrainingArguments(TrainingArguments):
         default=True,
         metadata={
             "help": (
-                "Whether to evaluate the agreement of the final student predictions and the teacher predictions"
+                "Whether to evaluate the agreement of the final student predictions and the teacher predictions "
                 "after training."
             )
         },
@@ -107,7 +107,7 @@ class DistillTrainingArguments(TrainingArguments):
         default=0,
         metadata={
             "help": (
-                "Limit the total amount of checkpoints."
+                "Limit the total amount of checkpoints. "
                 "Deletes the older checkpoints in the output_dir. Default is 0 (no checkpoints)."
             )
         },

examples/tensorflow/contrastive-image-text/run_clip.py

@@ -113,7 +113,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -387,7 +387,7 @@ def main():
         )
     else:
         raise ValueError(
-            "You are instantiating a new tokenizer from scratch. This is not supported by this script."
+            "You are instantiating a new tokenizer from scratch. This is not supported by this script. "
             "You can do it from another script, save it, and load it from here, using --tokenizer_name."
         )
 

examples/tensorflow/image-classification/run_image_classification.py

@@ -178,7 +178,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },

examples/tensorflow/language-modeling/run_clm.py

@@ -133,7 +133,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -399,7 +399,7 @@ def main():
         )
     else:
         raise ValueError(
-            "You are instantiating a new tokenizer from scratch. This is not supported by this script."
+            "You are instantiating a new tokenizer from scratch. This is not supported by this script. "
             "You can do it from another script, save it, and load it from here, using --tokenizer_name."
         )
     # endregion
@@ -432,7 +432,7 @@ def main():
     else:
         if data_args.block_size > tokenizer.model_max_length:
             logger.warning(
-                f"The block_size passed ({data_args.block_size}) is larger than the maximum length for the model"
+                f"The block_size passed ({data_args.block_size}) is larger than the maximum length for the model "
                 f"({tokenizer.model_max_length}). Using block_size={tokenizer.model_max_length}."
             )
         block_size = min(data_args.block_size, tokenizer.model_max_length)

examples/tensorflow/language-modeling/run_mlm.py

@@ -131,7 +131,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -383,7 +383,7 @@ def main():
         )
     else:
         raise ValueError(
-            "You are instantiating a new tokenizer from scratch. This is not supported by this script."
+            "You are instantiating a new tokenizer from scratch. This is not supported by this script. "
             "You can do it from another script, save it, and load it from here, using --tokenizer_name."
         )
     # endregion
@@ -404,7 +404,7 @@ def main():
     else:
         if data_args.max_seq_length > tokenizer.model_max_length:
             logger.warning(
-                f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"
+                f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the "
                 f"model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}."
             )
         max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)

examples/tensorflow/multiple-choice/run_swag.py

@@ -167,7 +167,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -383,7 +383,7 @@ def main():
     else:
         if data_args.max_seq_length > tokenizer.model_max_length:
             logger.warning(
-                f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"
+                f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the "
                 f"model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}."
             )
         max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)

examples/tensorflow/question-answering/run_qa.py

@@ -98,7 +98,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -400,7 +400,7 @@ def main():
 
     if data_args.max_seq_length > tokenizer.model_max_length:
         logger.warning(
-            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"
+            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the "
             f"model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}."
         )
     max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)

examples/tensorflow/summarization/run_summarization.py

@@ -120,7 +120,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -194,7 +194,7 @@ class DataTrainingArguments:
         metadata={
             "help": (
                 "The maximum total sequence length for validation target text after tokenization. Sequences longer "
-                "than this will be truncated, sequences shorter will be padded. Will default to `max_target_length`."
+                "than this will be truncated, sequences shorter will be padded. Will default to `max_target_length`. "
                 "This argument is also used to override the ``max_length`` param of ``model.generate``, which is used "
                 "during ``evaluate`` and ``predict``."
             )

examples/tensorflow/text-classification/run_glue.py

@@ -185,7 +185,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -353,7 +353,7 @@ def main():
 
     if data_args.max_seq_length > tokenizer.model_max_length:
         logger.warning(
-            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"
+            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the "
             f"model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}."
         )
     max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)

examples/tensorflow/text-classification/run_text_classification.py

@@ -101,7 +101,7 @@ class DataTrainingArguments:
         metadata={
             "help": (
                 "Whether to pad all samples to `max_seq_length`. "
-                "If False, will pad the samples dynamically when batching to the maximum length in the batch."
+                "If False, will pad the samples dynamically when batching to the maximum length in the batch. "
                 "Data will always be padded when using TPUs."
             )
         },
@@ -191,7 +191,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -360,7 +360,7 @@ def main():
 
     if data_args.max_seq_length > tokenizer.model_max_length:
         logger.warning(
-            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"
+            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the "
             f"model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}."
         )
     max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)

examples/tensorflow/token-classification/run_ner.py

@@ -96,7 +96,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -334,7 +334,7 @@ def main():
     tokenizer_name_or_path = model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path
     if not tokenizer_name_or_path:
         raise ValueError(
-            "You are instantiating a new tokenizer from scratch. This is not supported by this script."
+            "You are instantiating a new tokenizer from scratch. This is not supported by this script. "
             "You can do it from another script, save it, and load it from here, using --tokenizer_name."
         )
 

examples/tensorflow/translation/run_translation.py

@@ -114,7 +114,7 @@ class ModelArguments:
         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"
+                "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."
             )
         },
@@ -182,7 +182,7 @@ class DataTrainingArguments:
         metadata={
             "help": (
                 "The maximum total sequence length for validation target text after tokenization. Sequences longer "
-                "than this will be truncated, sequences shorter will be padded. Will default to `max_target_length`."
+                "than this will be truncated, sequences shorter will be padded. Will default to `max_target_length`. "
                 "This argument is also used to override the ``max_length`` param of ``model.generate``, which is used "
                 "during ``evaluate`` and ``predict``."
             )

src/transformers/commands/serving.py

@@ -131,7 +131,7 @@ class ServeCommand(BaseTransformersCLICommand):
         if not _serve_dependencies_installed:
             raise RuntimeError(
                 "Using serve command requires FastAPI and uvicorn. "
-                'Please install transformers with [serving]: pip install "transformers[serving]".'
+                'Please install transformers with [serving]: pip install "transformers[serving]". '
                 "Or install FastAPI and uvicorn separately."
             )
         else:

src/transformers/dynamic_module_utils.py

@@ -575,7 +575,7 @@ def custom_object_save(obj: Any, folder: Union[str, os.PathLike], config: Option
 
 def _raise_timeout_error(signum, frame):
     raise ValueError(
-        "Loading this model requires you to execute custom code contained in the model repository on your local"
+        "Loading this model requires you to execute custom code contained in the model repository on your local "
         "machine. Please set the option `trust_remote_code=True` to permit loading of this model."
     )
 
@@ -593,7 +593,7 @@ def resolve_trust_remote_code(trust_remote_code, model_name, has_local_code, has
                 signal.alarm(TIME_OUT_REMOTE_CODE)
                 while trust_remote_code is None:
                     answer = input(
-                        f"The repository for {model_name} contains custom code which must be executed to correctly"
+                        f"The repository for {model_name} contains custom code which must be executed to correctly "
                         f"load the model. You can inspect the repository content at https://hf.co/{model_name}.\n"
                         f"You can avoid this prompt in future by passing the argument `trust_remote_code=True`.\n\n"
                         f"Do you wish to run the custom code? [y/N] "
@@ -606,7 +606,7 @@ def resolve_trust_remote_code(trust_remote_code, model_name, has_local_code, has
             except Exception:
                 # OS which does not support signal.SIGALRM
                 raise ValueError(
-                    f"The repository for {model_name} contains custom code which must be executed to correctly"
+                    f"The repository for {model_name} contains custom code which must be executed to correctly "
                     f"load the model. You can inspect the repository content at https://hf.co/{model_name}.\n"
                     f"Please pass the argument `trust_remote_code=True` to allow custom code to be run."
                 )

src/transformers/generation/tf_utils.py

@@ -1198,7 +1198,7 @@ class TFGenerationMixin:
         inputs_kwarg = model_kwargs.pop(input_name, None)
         if inputs_kwarg is not None and inputs is not None:
             raise ValueError(
-                f"`inputs`: {inputs}` were passed alongside {input_name} which is not allowed."
+                f"`inputs`: {inputs}` were passed alongside {input_name} which is not allowed. "
                 f"Make sure to either pass {inputs} or {input_name}=..."
             )
         elif inputs_kwarg is not None:

src/transformers/generation/utils.py

@@ -544,7 +544,7 @@ class GenerationMixin:
         inputs_kwarg = model_kwargs.pop(input_name, None)
         if inputs_kwarg is not None and inputs is not None:
             raise ValueError(
-                f"`inputs`: {inputs}` were passed alongside {input_name} which is not allowed."
+                f"`inputs`: {inputs}` were passed alongside {input_name} which is not allowed. "
                 f"Make sure to either pass {inputs} or {input_name}=..."
             )
         elif inputs_kwarg is not None:
@@ -1816,7 +1816,7 @@ class GenerationMixin:
 
                 def typeerror():
                     raise ValueError(
-                        "`force_words_ids` has to either be a `List[List[List[int]]]` or `List[List[int]]`"
+                        "`force_words_ids` has to either be a `List[List[List[int]]]` or `List[List[int]]` "
                         f"of positive integers, but is {generation_config.force_words_ids}."
                     )
 

src/transformers/integrations/integration_utils.py

@@ -1312,7 +1312,7 @@ class NeptuneCallback(TrainerCallback):
                 target_path = consistent_checkpoint_path
             except IOError as e:
                 logger.warning(
-                    "NeptuneCallback was unable to made a copy of checkpoint due to I/O exception: '{}'."
+                    "NeptuneCallback was unable to made a copy of checkpoint due to I/O exception: '{}'. "
                     "Could fail trying to upload.".format(e)
                 )
 

src/transformers/modeling_tf_utils.py

@@ -850,7 +850,7 @@ def load_tf_shard(model, model_layer_map, resolved_archive_file, ignore_mismatch
             raise OSError(
                 f"Unable to load weights from TF checkpoint file for '{resolved_archive_file}' "
                 f"at '{resolved_archive_file}'. "
-                "If you tried to load a TF model from a sharded checkpoint, you should try converting the model"
+                "If you tried to load a TF model from a sharded checkpoint, you should try converting the model "
                 "by loading it in pytorch and saving it localy. A convertion script should be realeased soon."
             )
 

src/transformers/modeling_utils.py

@@ -2635,8 +2635,8 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
                 else:
                     raise RuntimeError("No GPU found. A GPU is needed for quantization.")
                 logger.info(
-                    "The device_map was not initialized."
-                    "Setting device_map to {'':torch.cuda.current_device()}."
+                    "The device_map was not initialized. "
+                    "Setting device_map to {'':torch.cuda.current_device()}. "
                     "If you want to use the model for inference, please set device_map ='auto' "
                 )
                 if low_cpu_mem_usage is None:
@@ -2762,8 +2762,8 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
                     else:
                         raise RuntimeError("No GPU found. A GPU is needed for quantization.")
                     logger.info(
-                        "The device_map was not initialized."
-                        "Setting device_map to {'':torch.cuda.current_device()}."
+                        "The device_map was not initialized. "
+                        "Setting device_map to {'':torch.cuda.current_device()}. "
                         "If you want to use the model for inference, please set device_map ='auto' "
                     )
                     if low_cpu_mem_usage is None:
@@ -3149,7 +3149,7 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
 
         if load_in_8bit and torch_dtype is None:
             logger.warning(
-                "You are loading your model in 8bit but you did not specify a `torch_dtype` attribute."
+                "You are loading your model in 8bit but you did not specify a `torch_dtype` attribute. "
                 "All non-linear modules will be loaded in full precision."
                 " If you want to load the other modules in other precision, please specify a `torch_dtype` attribute."
             )
@@ -3193,8 +3193,8 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
                 else:
                     raise ValueError(
                         "You are using `device_map='auto'` on a 4bit loaded version of the model. To automatically compute"
-                        " the appropriate device map, you should upgrade your `accelerate` library,"
-                        "`pip install --upgrade accelerate` or install it from source to support fp4 auto device map"
+                        " the appropriate device map, you should upgrade your `accelerate` library, "
+                        "`pip install --upgrade accelerate` or install it from source to support fp4 auto device map "
                         "calculation. You may encounter unexpected behavior, or pass your own device map"
                     )
             elif load_in_8bit:
@@ -3202,7 +3202,7 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
 
             if model._no_split_modules is None:
                 raise ValueError(
-                    f"{model.__class__.__name__} does not support `device_map='{device_map}'`. To implement support, the model"
+                    f"{model.__class__.__name__} does not support `device_map='{device_map}'`. To implement support, the model "
                     "class needs to implement the `_no_split_modules` attribute."
                 )
             no_split_modules = model._no_split_modules