Add missing whitespace to multiline strings (#13916)

src/transformers/models/canine/modeling_canine.py

@@ -549,7 +549,7 @@ class CanineAttention(nn.Module):
         self.local = local
         if attend_from_chunk_width < attend_from_chunk_stride:
             raise ValueError(
-                "`attend_from_chunk_width` < `attend_from_chunk_stride`"
+                "`attend_from_chunk_width` < `attend_from_chunk_stride` "
                 "would cause sequence positions to get skipped."
             )
         if attend_to_chunk_width < attend_to_chunk_stride:

src/transformers/models/clip/feature_extraction_clip.py

@@ -129,7 +129,7 @@ class CLIPFeatureExtractor(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/cpm/tokenization_cpm.py

@@ -92,7 +92,7 @@ class CpmTokenizer(XLNetTokenizer):
             import jieba
         except ModuleNotFoundError as error:
             raise error.__class__(
-                "You need to install jieba to use CpmTokenizer or CpmTokenizerFast."
+                "You need to install jieba to use CpmTokenizer or CpmTokenizerFast. "
                 "See https://pypi.org/project/jieba/ for installation."
             )
         self.jieba = jieba

src/transformers/models/cpm/tokenization_cpm_fast.py

@@ -95,7 +95,7 @@ class CpmTokenizerFast(XLNetTokenizerFast):
             import jieba
         except ModuleNotFoundError as error:
             raise error.__class__(
-                "You need to install jieba to use CpmTokenizer or CpmTokenizerFast."
+                "You need to install jieba to use CpmTokenizer or CpmTokenizerFast. "
                 "See https://pypi.org/project/jieba/ for installation."
             )
         self.jieba = jieba

src/transformers/models/deit/feature_extraction_deit.py

@@ -132,7 +132,7 @@ class DeiTFeatureExtractor(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/detr/feature_extraction_detr.py

@@ -483,7 +483,7 @@ class DetrFeatureExtractor(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/electra/modeling_electra.py

@@ -799,7 +799,7 @@ ELECTRA_INPUTS_DOCSTRING = r"""
 @add_start_docstrings(
     "The bare Electra Model transformer outputting raw hidden-states without any specific head on top. Identical to "
     "the BERT model except that it uses an additional linear layer between the embedding layer and the encoder if the "
-    "hidden size and embedding size are different."
+    "hidden size and embedding size are different. "
     ""
     "Both the generator and discriminator checkpoints may be loaded into this model.",
     ELECTRA_START_DOCSTRING,

src/transformers/models/electra/modeling_tf_electra.py

@@ -719,7 +719,7 @@ ELECTRA_INPUTS_DOCSTRING = r"""
 @add_start_docstrings(
     "The bare Electra Model transformer outputting raw hidden-states without any specific head on top. Identical to "
     "the BERT model except that it uses an additional linear layer between the embedding layer and the encoder if the "
-    "hidden size and embedding size are different."
+    "hidden size and embedding size are different. "
     ""
     "Both the generator and discriminator checkpoints may be loaded into this model.",
     ELECTRA_START_DOCSTRING,

src/transformers/models/encoder_decoder/modeling_encoder_decoder.py

@@ -482,7 +482,7 @@ class EncoderDecoderModel(PreTrainedModel):
 
     def resize_token_embeddings(self, *args, **kwargs):
         raise NotImplementedError(
-            "Resizing the embedding layers via the EncoderDecoderModel directly is not supported."
+            "Resizing the embedding layers via the EncoderDecoderModel directly is not supported. "
             "Please use the respective methods of the wrapped objects (model.encoder.resize_token_embeddings(...) or model.decoder.resize_token_embeddings(...))"
         )
 

src/transformers/models/gpt_neo/configuration_gpt_neo.py

@@ -151,11 +151,11 @@ class GPTNeoConfig(PretrainedConfig):
 
         if len(self.attention_layers) != self.num_layers:
             raise ValueError(
-                "Configuration for convolutional module is incorrect."
-                "It is required that `len(config.attention_layers)` == `config.num_layers`"
-                f"but is `len(config.attention_layers) = {len(self.attention_layers)}`,"
-                f"`config.num_layers = {self.num_layers}`."
-                "`config.attention_layers` is prepared using `config.attention_types`."
+                "Configuration for convolutional module is incorrect. "
+                "It is required that `len(config.attention_layers)` == `config.num_layers` "
+                f"but is `len(config.attention_layers) = {len(self.attention_layers)}`, "
+                f"`config.num_layers = {self.num_layers}`. "
+                "`config.attention_layers` is prepared using `config.attention_types`. "
                 "Please verify the value of `config.attention_types` argument."
             )
 

src/transformers/models/hubert/configuration_hubert.py

@@ -211,9 +211,9 @@ class HubertConfig(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/layoutlmv2/feature_extraction_layoutlmv2.py

@@ -181,7 +181,7 @@ class LayoutLMv2FeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionM
 
         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), "
                 f"but is of type {type(images)}."
             )

src/transformers/models/layoutlmv2/tokenization_layoutlmv2.py

@@ -431,7 +431,7 @@ class LayoutLMv2Tokenizer(PreTrainedTokenizer):
                 raise ValueError("text input must of type `str` (single example) or `List[str]` (batch of examples). ")
             if not isinstance(text_pair, (list, tuple)):
                 raise ValueError(
-                    "words must of type `List[str]` (single pretokenized example),"
+                    "words must of type `List[str]` (single pretokenized example), "
                     "or `List[List[str]]` (batch of pretokenized examples)."
                 )
         else:
@@ -599,7 +599,7 @@ class LayoutLMv2Tokenizer(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."
             )
@@ -838,9 +838,9 @@ class LayoutLMv2Tokenizer(PreTrainedTokenizer):
     ) -> BatchEncoding:
         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"
             )
@@ -1158,7 +1158,7 @@ class LayoutLMv2Tokenizer(PreTrainedTokenizer):
                 labels = labels[:-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 first sequence has a length {len(ids)}. "
                     f"Please select another truncation strategy than {truncation_strategy}, "
                     f"for instance 'longest_first' or 'only_second'."
@@ -1172,7 +1172,7 @@ class LayoutLMv2Tokenizer(PreTrainedTokenizer):
                 pair_token_boxes = pair_token_boxes[:-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'."

src/transformers/models/layoutlmv2/tokenization_layoutlmv2_fast.py

@@ -235,7 +235,7 @@ class LayoutLMv2TokenizerFast(PreTrainedTokenizerFast):
                 raise ValueError("text input must of type `str` (single example) or `List[str]` (batch of examples). ")
             if not isinstance(text_pair, (list, tuple)):
                 raise ValueError(
-                    "words must of type `List[str]` (single pretokenized example),"
+                    "words must of type `List[str]` (single pretokenized example), "
                     "or `List[List[str]]` (batch of pretokenized examples)."
                 )
         else:

src/transformers/models/luke/tokenization_luke.py

@@ -519,9 +519,9 @@ class LukeTokenizer(RobertaTokenizer):
 
         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"
             )
@@ -683,7 +683,7 @@ class LukeTokenizer(RobertaTokenizer):
     ) -> BatchEncoding:
         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."
             )
@@ -1308,7 +1308,7 @@ class LukeTokenizer(RobertaTokenizer):
         # The model's main input name, usually `input_ids`, has be passed for padding
         if self.model_input_names[0] not in encoded_inputs:
             raise ValueError(
-                "You should supply an encoding or a list of encodings to this method"
+                "You should supply an encoding or a list of encodings to this method "
                 f"that includes {self.model_input_names[0]}, but you provided {list(encoded_inputs.keys())}"
             )
 

src/transformers/models/rag/retrieval_rag.py

@@ -122,7 +122,7 @@ class LegacyIndex(Index):
         except EnvironmentError:
             msg = (
                 f"Can't load '{archive_file}'. Make sure that:\n\n"
-                f"- '{index_path}' is a correct remote path to a directory containing a file named {filename}"
+                f"- '{index_path}' is a correct remote path to a directory containing a file named {filename}\n\n"
                 f"- or '{index_path}' is the correct path to a directory containing a file named {filename}.\n\n"
             )
             raise EnvironmentError(msg)

src/transformers/models/roformer/tokenization_roformer.py

@@ -161,7 +161,7 @@ class RoFormerTokenizer(PreTrainedTokenizer):
             import rjieba
         except ImportError:
             raise ImportError(
-                "You need to install rjieba to use RoFormerTokenizer."
+                "You need to install rjieba to use RoFormerTokenizer. "
                 "See https://pypi.org/project/rjieba/ for installation."
             )
         self.jieba = rjieba

src/transformers/models/roformer/tokenization_utils.py

@@ -32,7 +32,7 @@ class JiebaPreTokenizer:
             import rjieba
         except ImportError:
             raise ImportError(
-                "You need to install rjieba to use RoFormerTokenizer."
+                "You need to install rjieba to use RoFormerTokenizer. "
                 "See https://pypi.org/project/rjieba/ for installation."
             )
         self.jieba = rjieba

src/transformers/models/speech_encoder_decoder/modeling_speech_encoder_decoder.py

@@ -352,7 +352,7 @@ class SpeechEncoderDecoderModel(PreTrainedModel):
                 decoder_config = AutoConfig.from_pretrained(decoder_pretrained_model_name_or_path)
                 if decoder_config.is_decoder is False or decoder_config.add_cross_attention is False:
                     logger.info(
-                        f"Initializing {decoder_pretrained_model_name_or_path} as a decoder model."
+                        f"Initializing {decoder_pretrained_model_name_or_path} as a decoder model. "
                         "Cross attention layers are added to {decoder_pretrained_model_name_or_path} "
                         "and randomly initialized if {decoder_pretrained_model_name_or_path}'s architecture allows for cross attention layers."
                     )
@@ -363,9 +363,9 @@ class SpeechEncoderDecoderModel(PreTrainedModel):
 
             if kwargs_decoder["config"].is_decoder is False or kwargs_decoder["config"].add_cross_attention is False:
                 logger.warning(
-                    f"Decoder model {decoder_pretrained_model_name_or_path} is not initialized as a decoder."
+                    f"Decoder model {decoder_pretrained_model_name_or_path} is not initialized as a decoder. "
                     f"In order to initialize {decoder_pretrained_model_name_or_path} as a decoder, "
-                    "make sure that the attributes `is_decoder` and `add_cross_attention` of `decoder_config`"
+                    "make sure that the attributes `is_decoder` and `add_cross_attention` of `decoder_config` "
                     "passed to `.from_encoder_decoder_pretrained(...)` are set to `True` or do not pass a `decoder_config` to `.from_encoder_decoder_pretrained(...)`"
                 )
 
@@ -513,7 +513,7 @@ class SpeechEncoderDecoderModel(PreTrainedModel):
 
     def resize_token_embeddings(self, *args, **kwargs):
         raise NotImplementedError(
-            "Resizing the embedding layers via the SpeechEncoderDecoderModel directly is not supported."
+            "Resizing the embedding layers via the SpeechEncoderDecoderModel directly is not supported. "
             "Please use the respective methods of the wrapped decoder object (model.decoder.resize_token_embeddings(...))"
         )
 

src/transformers/models/speech_to_text/configuration_speech_to_text.py

@@ -175,9 +175,9 @@ class Speech2TextConfig(PretrainedConfig):
 
         if len(self.conv_kernel_sizes) != self.num_conv_layers:
             raise ValueError(
-                "Configuration for convolutional module is incorrect."
-                "It is required that `len(config.conv_kernel_sizes)` == `config.num_conv_layers`"
-                f"but is `len(config.conv_kernel_sizes) = {len(self.conv_kernel_sizes)}`,"
+                "Configuration for convolutional module is incorrect. "
+                "It is required that `len(config.conv_kernel_sizes)` == `config.num_conv_layers` "
+                f"but is `len(config.conv_kernel_sizes) = {len(self.conv_kernel_sizes)}`, "
                 f"`config.num_conv_layers = {self.num_conv_layers}`."
             )