Add many missing spaces in adjacent strings (#26751)

Add missing spaces in adjacent strings

examples/research_projects/mm-imdb/run_mmimdb.py

@@ -426,7 +426,7 @@ def main():
         type=str,
         default="O1",
         help=(
-            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. "
             "See details at https://nvidia.github.io/apex/amp.html"
         ),
     )

examples/research_projects/movement-pruning/bertarize.py

@@ -112,8 +112,8 @@ if __name__ == "__main__":
         type=float,
         required=False,
         help=(
-            "For `magnitude` and `topK`, it is the level of remaining weights (in %) in the fine-pruned model."
-            "For `sigmoied_threshold`, it is the threshold \tau against which the (sigmoied) scores are compared."
+            "For `magnitude` and `topK`, it is the level of remaining weights (in %) in the fine-pruned model. "
+            "For `sigmoied_threshold`, it is the threshold \tau against which the (sigmoied) scores are compared. "
             "Not needed for `l0`"
         ),
     )

examples/research_projects/movement-pruning/counts_parameters.py

@@ -79,8 +79,8 @@ if __name__ == "__main__":
         type=float,
         required=False,
         help=(
-            "For `topK`, it is the level of remaining weights (in %) in the fine-pruned model."
-            "For `sigmoied_threshold`, it is the threshold \tau against which the (sigmoied) scores are compared."
+            "For `topK`, it is the level of remaining weights (in %) in the fine-pruned model. "
+            "For `sigmoied_threshold`, it is the threshold \tau against which the (sigmoied) scores are compared. "
             "Not needed for `l0`"
         ),
     )

examples/research_projects/movement-pruning/masked_run_glue.py

@@ -671,7 +671,7 @@ def main():
         default=1,
         type=int,
         help=(
-            "Run `initial_warmup` * `warmup_steps` steps of threshold warmup during which threshold stays"
+            "Run `initial_warmup` * `warmup_steps` steps of threshold warmup during which threshold stays "
             "at its `initial_threshold` value (sparsity schedule)."
         ),
     )
@@ -680,7 +680,7 @@ def main():
         default=2,
         type=int,
         help=(
-            "Run `final_warmup` * `warmup_steps` steps of threshold cool-down during which threshold stays"
+            "Run `final_warmup` * `warmup_steps` steps of threshold cool-down during which threshold stays "
             "at its final_threshold value (sparsity schedule)."
         ),
     )
@@ -799,7 +799,7 @@ def main():
         type=str,
         default="O1",
         help=(
-            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. "
             "See details at https://nvidia.github.io/apex/amp.html"
         ),
     )

examples/research_projects/movement-pruning/masked_run_squad.py

@@ -789,7 +789,7 @@ def main():
         default=1,
         type=int,
         help=(
-            "Run `initial_warmup` * `warmup_steps` steps of threshold warmup during which threshold stays"
+            "Run `initial_warmup` * `warmup_steps` steps of threshold warmup during which threshold stays "
             "at its `initial_threshold` value (sparsity schedule)."
         ),
     )
@@ -798,7 +798,7 @@ def main():
         default=2,
         type=int,
         help=(
-            "Run `final_warmup` * `warmup_steps` steps of threshold cool-down during which threshold stays"
+            "Run `final_warmup` * `warmup_steps` steps of threshold cool-down during which threshold stays "
             "at its final_threshold value (sparsity schedule)."
         ),
     )
@@ -946,7 +946,7 @@ def main():
         type=str,
         default="O1",
         help=(
-            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. "
             "See details at https://nvidia.github.io/apex/amp.html"
         ),
     )

examples/research_projects/performer/run_mlm_performer.py

@@ -466,7 +466,7 @@ if __name__ == "__main__":
         and not training_args.overwrite_output_dir
     ):
         raise ValueError(
-            f"Output directory ({training_args.output_dir}) already exists and is not empty."
+            f"Output directory ({training_args.output_dir}) already exists and is not empty. "
             "Use --overwrite_output_dir to overcome."
         )
 
@@ -558,7 +558,7 @@ if __name__ == "__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/research_projects/pplm/run_pplm_discrim_train.py

@@ -490,8 +490,8 @@ if __name__ == "__main__":
         default="SST",
         choices=("SST", "clickbait", "toxic", "generic"),
         help=(
-            "dataset to train the discriminator on."
-            "In case of generic, the dataset is expected"
+            "dataset to train the discriminator on. "
+            "In case of generic, the dataset is expected "
             "to be a TSBV file with structure: class \\t text"
         ),
     )

examples/research_projects/quantization-qdqbert/evaluate-hf-trt-qa.py

@@ -153,7 +153,7 @@ if args.tokenizer_name:
     tokenizer = AutoTokenizer.from_pretrained(args.tokenizer_name, use_fast=True)
 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."
     )
 
@@ -288,7 +288,7 @@ pad_on_right = tokenizer.padding_side == "right"
 
 if args.max_seq_length > tokenizer.model_max_length:
     logger.warning(
-        f"The max_seq_length passed ({args.max_seq_length}) is larger than the maximum length for the"
+        f"The max_seq_length passed ({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}."
     )
 

examples/research_projects/quantization-qdqbert/run_quant_qa.py

@@ -365,7 +365,7 @@ 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/research_projects/rag-end2end-retriever/finetune_rag.py

@@ -680,7 +680,7 @@ class GenerativeQAModule(BaseTransformer):
             type=int,
             default=1,
             help=(
-                "The number of retrieval actors to use when Ray is selected"
+                "The number of retrieval actors to use when Ray is selected "
                 "for the distributed retriever. Has no effect when "
                 "distributed_retriever is set to pytorch."
             ),
@@ -719,7 +719,7 @@ def main(args=None, model=None) -> GenerativeQAModule:
             ray.init(address=args.ray_address, namespace="rag")
         except (ConnectionError, ValueError):
             logger.warning(
-                "Connection to Ray cluster failed. Make sure a Ray"
+                "Connection to Ray cluster failed. Make sure a Ray "
                 "cluster is running by either using Ray's cluster "
                 "launcher (`ray up`) or by manually starting Ray on "
                 "each node via `ray start --head` for the head node "

examples/research_projects/rag-end2end-retriever/lightning_base.py

@@ -333,7 +333,7 @@ def add_generic_args(parser, root_dir) -> None:
         type=str,
         default="O2",
         help=(
-            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. "
             "See details at https://nvidia.github.io/apex/amp.html"
         ),
     )

examples/research_projects/rag/finetune_rag.py

@@ -525,7 +525,7 @@ class GenerativeQAModule(BaseTransformer):
             type=int,
             default=1,
             help=(
-                "The number of retrieval actors to use when Ray is selected"
+                "The number of retrieval actors to use when Ray is selected "
                 "for the distributed retriever. Has no effect when "
                 "distributed_retriever is set to pytorch."
             ),
@@ -552,7 +552,7 @@ def main(args=None, model=None) -> GenerativeQAModule:
             ray.init(address=args.ray_address, namespace="rag")
         except (ConnectionError, ValueError):
             logger.warning(
-                "Connection to Ray cluster failed. Make sure a Ray"
+                "Connection to Ray cluster failed. Make sure a Ray "
                 "cluster is running by either using Ray's cluster "
                 "launcher (`ray up`) or by manually starting Ray on "
                 "each node via `ray start --head` for the head node "

examples/research_projects/rag/lightning_base.py

@@ -322,7 +322,7 @@ def add_generic_args(parser, root_dir) -> None:
         type=str,
         default="O2",
         help=(
-            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. "
             "See details at https://nvidia.github.io/apex/amp.html"
         ),
     )

examples/research_projects/robust-speech-event/run_speech_recognition_ctc_bnb.py

@@ -104,8 +104,8 @@ 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."
             )
         },
@@ -399,7 +399,7 @@ 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/robust-speech-event/run_speech_recognition_ctc_streaming.py

@@ -103,8 +103,8 @@ 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."
             )
         },
@@ -354,7 +354,7 @@ 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/seq2seq-distillation/lightning_base.py

@@ -313,7 +313,7 @@ def add_generic_args(parser, root_dir) -> None:
         type=str,
         default="O2",
         help=(
-            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
+            "For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. "
             "See details at https://nvidia.github.io/apex/amp.html"
         ),
     )

examples/research_projects/tapex/run_tabfact_with_tapex.py

@@ -325,7 +325,7 @@ 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/research_projects/tapex/run_wikisql_with_tapex.py

@@ -170,7 +170,7 @@ 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``."
             )
@@ -379,7 +379,7 @@ def main():
 
     if training_args.label_smoothing_factor > 0 and not hasattr(model, "prepare_decoder_input_ids_from_labels"):
         logger.warning(
-            "label_smoothing is enabled but the `prepare_decoder_input_ids_from_labels` method is not defined for"
+            "label_smoothing is enabled but the `prepare_decoder_input_ids_from_labels` method is not defined for "
             f"`{model.__class__.__name__}`. This will lead to loss being calculated twice and will take up more memory"
         )
 

examples/research_projects/tapex/run_wikitablequestions_with_tapex.py

@@ -168,7 +168,7 @@ 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``."
             )
@@ -377,7 +377,7 @@ def main():
 
     if training_args.label_smoothing_factor > 0 and not hasattr(model, "prepare_decoder_input_ids_from_labels"):
         logger.warning(
-            "label_smoothing is enabled but the `prepare_decoder_input_ids_from_labels` method is not defined for"
+            "label_smoothing is enabled but the `prepare_decoder_input_ids_from_labels` method is not defined for "
             f"`{model.__class__.__name__}`. This will lead to loss being calculated twice and will take up more memory"
         )
 

examples/research_projects/wav2vec2/run_common_voice.py

@@ -80,8 +80,8 @@ class ModelArguments:
         default=0.05,
         metadata={
             "help": (
-                "Propability 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"
+                "Propability 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. This is only relevant if ``apply_spec_augment is True``."
             )
         },