pylint

Signed-off-by: n1ck-guo <heng.guo@intel.com>

intel_extension_for_transformers/transformers/kv_cache_compression/models/modeling_llama.py

@@ -50,7 +50,6 @@
     CausalLMOutputWithPast,
 )
 
-from intel_extension_for_transformers.transformers.modeling.modeling_gaudi import adapt_transformers_to_gaudi
 from ..prune import PruneConfig, H2OConfig
 
 logger = logging.get_logger(__name__)
@@ -920,12 +919,12 @@ def _update_causal_mask(
         # When output attentions is True, sdpa implementation's forward method calls the eager implementation's forward
         if self.config._attn_implementation == "sdpa" \
             and not using_static_cache and not output_attentions: # pylint: disable=E1101
-            if AttentionMaskConverter._ignore_causal_mask_sdpa(
+            if AttentionMaskConverter._ignore_causal_mask_sdpa( # pylint: disable=E1101
                 attention_mask,
                 inputs_embeds=input_tensor,
                 past_key_values_length=past_seen_tokens,
                 is_training=self.training,
-            ):  # pylint: disable=E1101
+            ):  
                 return None
 
         dtype, device = input_tensor.dtype, input_tensor.device

intel_extension_for_transformers/transformers/kv_cache_compression/prune/h2o.py

@@ -72,7 +72,11 @@ def get_hh_mask(heavy_budget_ratio, recent_budget_ratio, attn_weights, local=Tru
 
             zeros = torch.zeros_like(tmp_sum, dtype=torch.bool)
             mask_bottom = zeros.scatter(-1, tmp_topk, True).unsqueeze(2)
-            mask_bottom = mask_bottom.expand(mask_bottom.shape[0], mask_bottom.shape[1], attn_weights.shape[-2], mask_bottom.shape[-1])
+            mask_bottom = mask_bottom.expand(
+                mask_bottom.shape[0],
+                mask_bottom.shape[1],
+                attn_weights.shape[-2],
+                mask_bottom.shape[-1])
     else:
         mask_bottom = torch.zeros_like(attn_weights, dtype=torch.bool)