Migrate linter from pylint to ruff (#1665)
This commit is contained in:
@@ -261,10 +261,7 @@ class AquilaModel(nn.Module):
|
||||
) -> torch.Tensor:
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states = layer(
|
||||
positions,
|
||||
|
||||
@@ -281,10 +281,7 @@ class BaiChuanModel(nn.Module):
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
residual = None
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states, residual = layer(
|
||||
positions,
|
||||
|
||||
@@ -256,10 +256,7 @@ class BloomModel(nn.Module):
|
||||
hidden_states = self.word_embeddings(input_ids)
|
||||
hidden_states = self.word_embeddings_layernorm(hidden_states)
|
||||
for i in range(len(self.h)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states = layer(
|
||||
position_ids,
|
||||
|
||||
@@ -269,10 +269,7 @@ class GLMTransformer(nn.Module):
|
||||
cache_events: Optional[List[torch.cuda.Event]],
|
||||
) -> torch.Tensor:
|
||||
for i in range(self.num_layers):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states = layer(
|
||||
hidden_states=hidden_states,
|
||||
|
||||
@@ -353,10 +353,7 @@ class FalconModel(nn.Module):
|
||||
) -> torch.Tensor:
|
||||
hidden_states = self.word_embeddings(input_ids)
|
||||
for i in range(len(self.h)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states = layer(
|
||||
positions,
|
||||
|
||||
@@ -206,10 +206,7 @@ class GPT2Model(nn.Module):
|
||||
hidden_states = inputs_embeds + position_embeds
|
||||
|
||||
for i in range(len(self.h)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states = layer(hidden_states, kv_caches[i], input_metadata,
|
||||
cache_event)
|
||||
|
||||
@@ -225,10 +225,7 @@ class GPTBigCodeModel(nn.Module):
|
||||
hidden_states = inputs_embeds + position_embeds
|
||||
|
||||
for i in range(len(self.h)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states = layer(hidden_states, kv_caches[i], input_metadata,
|
||||
cache_event)
|
||||
|
||||
@@ -147,10 +147,7 @@ class GPTJBlock(nn.Module):
|
||||
linear_method: Optional[LinearMethodBase] = None,
|
||||
):
|
||||
super().__init__()
|
||||
if config.n_inner is None:
|
||||
inner_dim = 4 * config.n_embd
|
||||
else:
|
||||
inner_dim = config.n_inner
|
||||
inner_dim = 4 * config.n_embd if config.n_inner is None else config.n_inner
|
||||
self.ln_1 = nn.LayerNorm(config.n_embd, eps=config.layer_norm_epsilon)
|
||||
self.attn = GPTJAttention(config, linear_method)
|
||||
self.mlp = GPTJMLP(inner_dim, config, linear_method)
|
||||
@@ -205,10 +202,7 @@ class GPTJModel(nn.Module):
|
||||
) -> torch.Tensor:
|
||||
hidden_states = self.wte(input_ids)
|
||||
for i in range(len(self.h)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states = layer(
|
||||
position_ids,
|
||||
|
||||
@@ -216,10 +216,7 @@ class GPTNeoXModel(nn.Module):
|
||||
) -> torch.Tensor:
|
||||
hidden_states = self.embed_in(input_ids)
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states = layer(
|
||||
position_ids,
|
||||
|
||||
@@ -213,10 +213,7 @@ class InternLMModel(nn.Module):
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
residual = None
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states, residual = layer(
|
||||
positions,
|
||||
|
||||
@@ -253,10 +253,7 @@ class LlamaModel(nn.Module):
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
residual = None
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states, residual = layer(
|
||||
positions,
|
||||
|
||||
@@ -248,10 +248,7 @@ class MistralModel(nn.Module):
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
residual = None
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states, residual = layer(
|
||||
positions,
|
||||
|
||||
@@ -203,10 +203,10 @@ class MPTModel(nn.Module):
|
||||
self.norm_f = nn.LayerNorm(config.d_model)
|
||||
if config.no_bias:
|
||||
for module in self.modules():
|
||||
if hasattr(module, "bias"):
|
||||
if isinstance(module.bias, nn.Parameter):
|
||||
# Remove the bias term in Linear and LayerNorm.
|
||||
module.register_parameter("bias", None)
|
||||
if hasattr(module, "bias") and isinstance(
|
||||
module.bias, nn.Parameter):
|
||||
# Remove the bias term in Linear and LayerNorm.
|
||||
module.register_parameter("bias", None)
|
||||
|
||||
def forward(
|
||||
self,
|
||||
@@ -218,10 +218,7 @@ class MPTModel(nn.Module):
|
||||
) -> torch.Tensor:
|
||||
hidden_states = self.wte(input_ids)
|
||||
for i in range(len(self.blocks)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
block = self.blocks[i]
|
||||
hidden_states = block(
|
||||
position_ids,
|
||||
|
||||
@@ -257,10 +257,7 @@ class OPTDecoder(nn.Module):
|
||||
hidden_states = inputs_embeds + pos_embeds
|
||||
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states = layer(hidden_states, kv_caches[i], input_metadata,
|
||||
cache_event)
|
||||
|
||||
@@ -258,10 +258,7 @@ class PhiModel(nn.Module):
|
||||
) -> SamplerOutput:
|
||||
hidden_states = self.embd(input_ids)
|
||||
for i in range(self.config.num_hidden_layers):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states = layer(
|
||||
positions,
|
||||
|
||||
@@ -213,10 +213,7 @@ class QWenModel(nn.Module):
|
||||
hidden_states = self.wte(input_ids)
|
||||
residual = None
|
||||
for i in range(len(self.h)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.h[i]
|
||||
hidden_states, residual = layer(
|
||||
positions,
|
||||
|
||||
@@ -249,10 +249,7 @@ class YiModel(nn.Module):
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
residual = None
|
||||
for i in range(len(self.layers)):
|
||||
if cache_events is None:
|
||||
cache_event = None
|
||||
else:
|
||||
cache_event = cache_events[i]
|
||||
cache_event = None if cache_events is None else cache_events[i]
|
||||
layer = self.layers[i]
|
||||
hidden_states, residual = layer(
|
||||
positions,
|
||||
|
||||
Reference in New Issue
Block a user