[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

Author: pre-commit-ci[bot]

transformer_engine/pytorch/attention/dot_product_attention/backends.py

@@ -169,6 +169,7 @@
     from flash_attn.cute.interface import flash_attn_varlen_func as flash_attn_varlen_func_v4
     from flash_attn.cute.interface import _flash_attn_fwd as _flash_attn_fwd_v4
     from flash_attn.cute.interface import _flash_attn_bwd as _flash_attn_bwd_v4
+
     # flash_attn_with_kvcache_v4 = None  # FA4 does not support kvcache yet
     fa_utils.set_flash_attention_4_params()
 

transformer_engine/pytorch/attention/dot_product_attention/utils.py

@@ -1128,8 +1128,9 @@ def _is_fa3_supported(num_heads, num_gqa_groups, head_dim_qk, head_dim_v, qkv_dt
             use_flash_attention_2 = False
     if use_flash_attention_3 and deterministic and FlashAttentionUtils.v3_is_installed:
         if head_dim_qk >= 256:
-            logger.debug("Disabling FlashAttention 3 for deterministic execution with "
-                         "head_dim_qk >= 256.")
+            logger.debug(
+                "Disabling FlashAttention 3 for deterministic execution with head_dim_qk >= 256."
+            )
             use_flash_attention_3 = False
     if use_fused_attention and deterministic:
         if fused_attention_backend == FusedAttnBackend["FP8"] and is_training: