make fix-copies

src/transformers/models/whisper/generation_whisper.py

@@ -173,7 +173,9 @@ def _pad_to_max_length(
 
 
 class WhisperGenerationMixin:
-    def _extract_token_timestamps(self, generate_outputs, alignment_heads, time_precision=0.02, num_frames=None, num_input_ids=None):
+    def _extract_token_timestamps(
+        self, generate_outputs, alignment_heads, time_precision=0.02, num_frames=None, num_input_ids=None
+    ):
         """
         Calculates token-level timestamps using the encoder-decoder cross-attentions and dynamic time-warping (DTW) to
         map each output token to a position in the input audio. If `num_frames` is specified, the encoder-decoder
@@ -200,8 +202,8 @@ def _extract_token_timestamps(self, generate_outputs, alignment_heads, time_prec
             # since the beam search strategy chooses the most probable sequences at the end of the search.
             # In that case, the cross_attentions weights are too long and we have to make sure that they have the right output_length
             weight_length = (generate_outputs.beam_indices != -1).sum(-1).max()
-            
-            # beam search takes `decoder_input_ids` into account in the `beam_indices` length 
+
+            # beam search takes `decoder_input_ids` into account in the `beam_indices` length
             # but forgot to shift the beam_indices by the number of `decoder_input_ids`
             weight_length = weight_length if num_input_ids is None else weight_length + num_input_ids
             weights = weights[:, :, :weight_length]
@@ -223,7 +225,9 @@ def _extract_token_timestamps(self, generate_outputs, alignment_heads, time_prec
         # make sure timestamps are as long as weights
         input_length = weight_length or cross_attentions[0].shape[2]
         batch_size = generate_outputs.sequences.shape[0]
-        timestamps = torch.zeros((batch_size, input_length + 1), dtype=torch.float32, device=generate_outputs.sequences.device)
+        timestamps = torch.zeros(
+            (batch_size, input_length + 1), dtype=torch.float32, device=generate_outputs.sequences.device
+        )
 
         if num_frames is not None:
             # two cases:
@@ -953,7 +957,10 @@ def _postprocess_outputs(
         if return_token_timestamps and hasattr(generation_config, "alignment_heads"):
             num_frames = getattr(generation_config, "num_frames", None)
             seek_outputs["token_timestamps"] = self._extract_token_timestamps(
-                seek_outputs, generation_config.alignment_heads, num_frames=num_frames, num_input_ids=decoder_input_ids.shape[-1]
+                seek_outputs,
+                generation_config.alignment_heads,
+                num_frames=num_frames,
+                num_input_ids=decoder_input_ids.shape[-1],
             )
             seek_outputs["token_timestamps"] = seek_outputs["token_timestamps"][:, start_idx:]
 

src/transformers/models/whisper/modeling_whisper.py

@@ -475,7 +475,7 @@ def forward(
 
         causal_mask = attention_mask
         if attention_mask is not None:  # no matter the length, we just slice it
-            causal_mask = attention_mask[:, :key_states.shape[-2]]
+            causal_mask = attention_mask[:, : key_states.shape[-2]]
 
         # In PEFT, usually we cast the layer norms in float32 for training stability reasons
         # therefore the input hidden states gets silently casted in float32. Hence, we need