Super tiny fix 12 typos about "with with" (#29926)

* with with

* style

docs/source/en/model_doc/fastspeech2_conformer.md

@@ -24,7 +24,7 @@ This model was contributed by [Connor Henderson](https://huggingface.co/connor-h
 
 
 ## 🤗 Model Architecture
-FastSpeech2's general structure with a Mel-spectrogram decoder was implemented, and the traditional transformer blocks were replaced with with conformer blocks as done in the ESPnet library.
+FastSpeech2's general structure with a Mel-spectrogram decoder was implemented, and the traditional transformer blocks were replaced with conformer blocks as done in the ESPnet library.
 
 #### FastSpeech2 Model Architecture
 ![FastSpeech2 Model Architecture](https://www.microsoft.com/en-us/research/uploads/prod/2021/04/fastspeech2-1.png)

src/transformers/models/align/modeling_align.py

@@ -400,7 +400,7 @@ class AlignVisionExpansionLayer(nn.Module):
         return hidden_states
 
 
-# Copied from transformers.models.efficientnet.modeling_efficientnet.EfficientNetDepthwiseLayer with with EfficientNet->AlignVision
+# Copied from transformers.models.efficientnet.modeling_efficientnet.EfficientNetDepthwiseLayer with EfficientNet->AlignVision
 class AlignVisionDepthwiseLayer(nn.Module):
     r"""
     This corresponds to the depthwise convolution phase of each block in the original implementation.
@@ -440,7 +440,7 @@ class AlignVisionDepthwiseLayer(nn.Module):
         return hidden_states
 
 
-# Copied from transformers.models.efficientnet.modeling_efficientnet.EfficientNetSqueezeExciteLayer with with EfficientNet->AlignVision
+# Copied from transformers.models.efficientnet.modeling_efficientnet.EfficientNetSqueezeExciteLayer with EfficientNet->AlignVision
 class AlignVisionSqueezeExciteLayer(nn.Module):
     r"""
     This corresponds to the Squeeze and Excitement phase of each block in the original implementation.

src/transformers/models/bart/modeling_bart.py

@@ -2093,7 +2093,7 @@ class BartDecoderWrapper(BartPreTrainedModel):
 
 @add_start_docstrings(
     """
-    BART decoder with with a language modeling head on top (linear layer with weights tied to the input embeddings).
+    BART decoder with a language modeling head on top (linear layer with weights tied to the input embeddings).
     """,
     BART_START_DOCSTRING,
 )

src/transformers/models/deprecated/open_llama/configuration_open_llama.py

@@ -154,8 +154,7 @@ class OpenLlamaConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/falcon/configuration_falcon.py

@@ -177,8 +177,7 @@ class FalconConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/fuyu/configuration_fuyu.py

@@ -199,8 +199,7 @@ class FuyuConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/gpt_neox/configuration_gpt_neox.py

@@ -167,8 +167,7 @@ class GPTNeoXConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/llama/configuration_llama.py

@@ -179,8 +179,7 @@ class LlamaConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/persimmon/configuration_persimmon.py

@@ -151,8 +151,7 @@ class PersimmonConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/phi/configuration_phi.py

@@ -179,8 +179,7 @@ class PhiConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/stablelm/configuration_stablelm.py

@@ -168,8 +168,7 @@ class StableLmConfig(PretrainedConfig):
 
         if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
             raise ValueError(
-                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                "`rope_scaling` must be a dictionary with two fields, `type` and `factor`, " f"got {self.rope_scaling}"
-                f"got {self.rope_scaling}"
             )
         rope_scaling_type = self.rope_scaling.get("type", None)
         rope_scaling_factor = self.rope_scaling.get("factor", None)

src/transformers/models/whisper/modeling_whisper.py

@@ -1861,7 +1861,7 @@ class WhisperDecoderWrapper(WhisperPreTrainedModel):
 
 @add_start_docstrings(
     """
-    Whisper decoder with with a language modeling head on top (linear layer with weights tied to the input embeddings).
+    Whisper decoder with a language modeling head on top (linear layer with weights tied to the input embeddings).
     """,
     WHISPER_START_DOCSTRING,
 )