Add missing whitespace to multiline strings (#13916)

src/transformers/models/speech_to_text/feature_extraction_speech_to_text.py

@@ -189,12 +189,12 @@ class Speech2TextFeatureExtractor(SequenceFeatureExtractor):
         if sampling_rate is not None:
             if sampling_rate != self.sampling_rate:
                 raise ValueError(
-                    f"The model corresponding to this feature extractor: {self} was trained using a sampling rate of {self.sampling_rate}."
+                    f"The model corresponding to this feature extractor: {self} was trained using a sampling rate of {self.sampling_rate}. "
                     f"Please make sure that the provided `raw_speech` input was sampled with {self.sampling_rate} and not {sampling_rate}."
                 )
         else:
             logger.warning(
-                "It is strongly recommended to pass the `sampling_rate` argument to this function."
+                "It is strongly recommended to pass the `sampling_rate` argument to this function. "
                 "Failing to do so can result in silent errors that might be hard to debug."
             )
 

src/transformers/models/squeezebert/modeling_squeezebert.py

@@ -300,7 +300,7 @@ class SqueezeBertEncoder(nn.Module):
         super().__init__()
 
         assert config.embedding_size == config.hidden_size, (
-            "If you want embedding_size != intermediate hidden_size,"
+            "If you want embedding_size != intermediate hidden_size, "
             "please insert a Conv1d layer to adjust the number of channels "
             "before the first SqueezeBertModule."
         )

src/transformers/models/tapas/modeling_tapas.py

@@ -54,8 +54,8 @@ if is_scatter_available():
         from torch_scatter import scatter
     except OSError:
         logger.error(
-            "TAPAS models are not usable since `torch_scatter` can't be loaded."
-            "It seems you have `torch_scatter` installed with the wrong CUDA version."
+            "TAPAS models are not usable since `torch_scatter` can't be loaded. "
+            "It seems you have `torch_scatter` installed with the wrong CUDA version. "
             "Please try to reinstall it following the instructions here: https://github.com/rusty1s/pytorch_scatter."
         )
 

src/transformers/models/tapas/tokenization_tapas.py

@@ -712,7 +712,7 @@ class TapasTokenizer(PreTrainedTokenizer):
 
         if return_offsets_mapping:
             raise NotImplementedError(
-                "return_offset_mapping is not available when using Python tokenizers."
+                "return_offset_mapping is not available when using Python tokenizers. "
                 "To use this feature, change your tokenizer to one deriving from "
                 "transformers.PreTrainedTokenizerFast."
             )
@@ -981,7 +981,7 @@ class TapasTokenizer(PreTrainedTokenizer):
 
         if return_offsets_mapping:
             raise NotImplementedError(
-                "return_offset_mapping is not available when using Python tokenizers."
+                "return_offset_mapping is not available when using Python tokenizers. "
                 "To use this feature, change your tokenizer to one deriving from "
                 "transformers.PreTrainedTokenizerFast."
             )
@@ -1159,7 +1159,7 @@ class TapasTokenizer(PreTrainedTokenizer):
 
         if max_length is not None and len(input_ids) > max_length:
             raise ValueError(
-                "Could not encode the query and table header given the maximum length. Encoding the query and table"
+                "Could not encode the query and table header given the maximum length. Encoding the query and table "
                 f"header results in a length of {len(input_ids)} which is higher than the max_length of {max_length}"
             )
 

src/transformers/models/transfo_xl/tokenization_transfo_xl.py

@@ -236,7 +236,7 @@ class TransfoXLTokenizer(PreTrainedTokenizer):
         except Exception as e:
             raise ValueError(
                 f"Unable to parse file {pretrained_vocab_file}. Unknown format. "
-                "If you tried to load a model saved through TransfoXLTokenizerFast,"
+                "If you tried to load a model saved through TransfoXLTokenizerFast, "
                 "please note they are not compatible."
             ) from e
 

src/transformers/models/visual_bert/modeling_visual_bert.py

@@ -174,7 +174,7 @@ class VisualBertEmbeddings(nn.Module):
                 if visual_position_embeddings.size(1) != visual_embeds.size(1):
                     if visual_position_embeddings.size(1) < visual_embeds.size(1):
                         raise ValueError(
-                            f"Visual position embeddings length: {visual_position_embeddings.size(1)}"
+                            f"Visual position embeddings length: {visual_position_embeddings.size(1)} "
                             f"should be the same as `visual_embeds` length: {visual_embeds.size(1)}"
                         )
                     visual_position_embeddings = visual_position_embeddings[:, : visual_embeds.size(1), :]
@@ -973,7 +973,7 @@ class VisualBertForPreTraining(VisualBertPreTrainedModel):
             total_size = attention_mask.size(-1) + visual_attention_mask.size(-1)
             if labels.size(-1) != total_size:
                 raise ValueError(
-                    f"The labels provided should have same sequence length as total attention mask."
+                    f"The labels provided should have same sequence length as total attention mask. "
                     f"Found labels with sequence length {labels.size(-1)}, expected {total_size}."
                 )
 
@@ -986,7 +986,7 @@ class VisualBertForPreTraining(VisualBertPreTrainedModel):
             total_size = attention_mask.size(-1) + visual_attention_mask.size(-1)
             if labels.size(-1) != total_size:
                 raise ValueError(
-                    f"The labels provided should have same sequence length as total attention mask."
+                    f"The labels provided should have same sequence length as total attention mask. "
                     f"Found labels with sequence length {labels.size(-1)}, expected {total_size}."
                 )
 

src/transformers/models/vit/feature_extraction_vit.py

@@ -122,7 +122,7 @@ class ViTFeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionMixin):
 
         if not valid_images:
             raise ValueError(
-                "Images must of type `PIL.Image.Image`, `np.ndarray` or `torch.Tensor` (single example),"
+                "Images must of type `PIL.Image.Image`, `np.ndarray` or `torch.Tensor` (single example), "
                 "`List[PIL.Image.Image]`, `List[np.ndarray]` or `List[torch.Tensor]` (batch of examples)."
             )
 

src/transformers/models/wav2vec2/configuration_wav2vec2.py

@@ -237,9 +237,9 @@ class Wav2Vec2Config(PretrainedConfig):
             or (len(self.conv_dim) != self.num_feat_extract_layers)
         ):
             raise ValueError(
-                "Configuration for convolutional layers is incorrect."
-                "It is required that `len(config.conv_dim)` == `len(config.conv_stride)` == `len(config.conv_kernel)`,"
-                f"but is `len(config.conv_dim) = {len(self.conv_dim)}`, `len(config.conv_stride)"
+                "Configuration for convolutional layers is incorrect. "
+                "It is required that `len(config.conv_dim)` == `len(config.conv_stride)` == `len(config.conv_kernel)`, "
+                f"but is `len(config.conv_dim) = {len(self.conv_dim)}`, `len(config.conv_stride) "
                 f"= {len(self.conv_stride)}`, `len(config.conv_kernel) = {len(self.conv_kernel)}`."
             )
 

src/transformers/models/wav2vec2/feature_extraction_wav2vec2.py

@@ -170,12 +170,12 @@ class Wav2Vec2FeatureExtractor(SequenceFeatureExtractor):
         if sampling_rate is not None:
             if sampling_rate != self.sampling_rate:
                 raise ValueError(
-                    f"The model corresponding to this feature extractor: {self} was trained using a sampling rate of {self.sampling_rate}."
+                    f"The model corresponding to this feature extractor: {self} was trained using a sampling rate of {self.sampling_rate}. "
                     f"Please make sure that the provided `raw_speech` input was sampled with {self.sampling_rate} and not {sampling_rate}."
                 )
         else:
             logger.warning(
-                "It is strongly recommended to pass the ``sampling_rate`` argument to this function."
+                "It is strongly recommended to pass the ``sampling_rate`` argument to this function. "
                 "Failing to do so can result in silent errors that might be hard to debug."
             )
 

src/transformers/models/wav2vec2/modeling_wav2vec2.py

@@ -1421,7 +1421,7 @@ class Wav2Vec2ForCTC(Wav2Vec2PreTrainedModel):
             raise ValueError(
                 f"You are trying to instantiate {self.__class__} with a configuration that "
                 "does not define the vocabulary size of the language model head. Please "
-                "instantiate the model as follows: `Wav2Vec2ForCTC.from_pretrained(..., vocab_size=vocab_size)`."
+                "instantiate the model as follows: `Wav2Vec2ForCTC.from_pretrained(..., vocab_size=vocab_size)`. "
                 "or define `vocab_size` of your model's configuration."
             )
         self.lm_head = nn.Linear(config.hidden_size, config.vocab_size)

src/transformers/models/xlm_prophetnet/tokenization_xlm_prophetnet.py

@@ -152,7 +152,7 @@ class XLMProphetNetTokenizer(PreTrainedTokenizer):
             import sentencepiece as spm
         except ImportError:
             logger.warning(
-                "You need to install SentencePiece to use XLMRobertaTokenizer: https://github.com/google/sentencepiece"
+                "You need to install SentencePiece to use XLMRobertaTokenizer: https://github.com/google/sentencepiece "
                 "pip install sentencepiece"
             )
             raise
@@ -191,7 +191,7 @@ class XLMProphetNetTokenizer(PreTrainedTokenizer):
             import sentencepiece as spm
         except ImportError:
             logger.warning(
-                "You need to install SentencePiece to use XLMRobertaTokenizer: https://github.com/google/sentencepiece"
+                "You need to install SentencePiece to use XLMRobertaTokenizer: https://github.com/google/sentencepiece "
                 "pip install sentencepiece"
             )
             raise

src/transformers/onnx/features.py

@@ -109,7 +109,7 @@ class FeaturesManager:
         task = FeaturesManager.feature_to_task(feature)
         if task not in FeaturesManager._TASKS_TO_AUTOMODELS:
             raise KeyError(
-                f"Unknown task: {feature}."
+                f"Unknown task: {feature}. "
                 f"Possible values are {list(FeaturesManager._TASKS_TO_AUTOMODELS.values())}"
             )
 

src/transformers/tokenization_utils.py

@@ -596,9 +596,9 @@ class PreTrainedTokenizer(PreTrainedTokenizerBase):
 
         if return_offsets_mapping:
             raise NotImplementedError(
-                "return_offset_mapping is not available when using Python tokenizers."
+                "return_offset_mapping is not available when using Python tokenizers. "
                 "To use this feature, change your tokenizer to one deriving from "
-                "transformers.PreTrainedTokenizerFast."
+                "transformers.PreTrainedTokenizerFast. "
                 "More information on available tokenizers at "
                 "https://github.com/huggingface/transformers/pull/2674"
             )
@@ -673,7 +673,7 @@ class PreTrainedTokenizer(PreTrainedTokenizerBase):
 
         if return_offsets_mapping:
             raise NotImplementedError(
-                "return_offset_mapping is not available when using Python tokenizers."
+                "return_offset_mapping is not available when using Python tokenizers. "
                 "To use this feature, change your tokenizer to one deriving from "
                 "transformers.PreTrainedTokenizerFast."
             )

src/transformers/tokenization_utils_base.py

@@ -3059,7 +3059,7 @@ class PreTrainedTokenizerBase(SpecialTokensMixin, PushToHubMixin):
                 pair_ids = pair_ids[:-num_tokens_to_remove]
             else:
                 logger.error(
-                    f"We need to remove {num_tokens_to_remove} to truncate the input"
+                    f"We need to remove {num_tokens_to_remove} to truncate the input "
                     f"but the second sequence has a length {len(pair_ids)}. "
                     f"Please select another truncation strategy than {truncation_strategy}, "
                     f"for instance 'longest_first' or 'only_first'."
@@ -3250,7 +3250,7 @@ class PreTrainedTokenizerBase(SpecialTokensMixin, PushToHubMixin):
         """
         assert already_has_special_tokens and token_ids_1 is None, (
             "You cannot use ``already_has_special_tokens=False`` with this tokenizer. "
-            "Please use a slow (full python) tokenizer to activate this argument."
+            "Please use a slow (full python) tokenizer to activate this argument. "
             "Or set `return_special_tokens_mask=True` when calling the encoding method "
             "to get the special tokens mask in any tokenizer. "
         )

src/transformers/trainer.py

@@ -385,7 +385,7 @@ class Trainer:
         self.optimizer, self.lr_scheduler = optimizers
         if model_init is not None and (self.optimizer is not None or self.lr_scheduler is not None):
             raise RuntimeError(
-                "Passing a `model_init` is incompatible with providing the `optimizers` argument."
+                "Passing a `model_init` is incompatible with providing the `optimizers` argument. "
                 "You should subclass `Trainer` and override the `create_optimizer_and_scheduler` method."
             )
         default_callbacks = DEFAULT_CALLBACKS + get_reporting_integration_callbacks(self.args.report_to)
@@ -1737,8 +1737,8 @@ class Trainer:
             if backend is None:
                 raise RuntimeError(
                     "At least one of optuna or ray should be installed. "
-                    "To install optuna run `pip install optuna`."
-                    "To install ray run `pip install ray[tune]`."
+                    "To install optuna run `pip install optuna`. "
+                    "To install ray run `pip install ray[tune]`. "
                     "To install sigopt run `pip install sigopt`."
                 )
         backend = HPSearchBackend(backend)

src/transformers/training_args.py

@@ -385,7 +385,7 @@ class TrainingArguments:
         default=False,
         metadata={
             "help": (
-                "Overwrite the content of the output directory."
+                "Overwrite the content of the output directory. "
                 "Use this to continue training if output_dir points to a checkpoint directory."
             )
         },
@@ -420,7 +420,7 @@ class TrainingArguments:
     per_gpu_eval_batch_size: Optional[int] = field(
         default=None,
         metadata={
-            "help": "Deprecated, the use of `--per_device_eval_batch_size` is preferred."
+            "help": "Deprecated, the use of `--per_device_eval_batch_size` is preferred. "
             "Batch size per GPU/TPU core/CPU for evaluation."
         },
     )
@@ -492,7 +492,7 @@ class TrainingArguments:
         default=None,
         metadata={
             "help": (
-                "Limit the total amount of checkpoints."
+                "Limit the total amount of checkpoints. "
                 "Deletes the older checkpoints in the output_dir. Default is unlimited checkpoints"
             )
         },
@@ -514,7 +514,7 @@ class TrainingArguments:
         default="O1",
         metadata={
             "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"
             )
         },