jinx/ComfyUI
1
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2025-08-02 15:04:50 +08:00
Code Issues Activity

Lint all unused variables (#5989)

Browse Source 
* Enable F841

* Autofix

* Remove all unused variable assignment
This commit is contained in:
Chenlei Hu
2024-12-12 14:59:16 -08:00
committed by GitHub
parent fd5dfb812c
commit d9d7f3c619
29 changed files with 22 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
comfy/ldm/modules/diffusionmodules/model.py
View File

@@ -162,7 +162,6 @@ def slice_attention(q, k, v):
mem_free_total = model_management.get_free_memory(q.device)
gb = 1024 ** 3
tensor_size = q.shape[0] * q.shape[1] * k.shape[2] * q.element_size()
modifier = 3 if q.element_size() == 2 else 2.5
mem_required = tensor_size * modifier
@@ -218,7 +217,7 @@ def xformers_attention(q, k, v):
try:
out = xformers.ops.memory_efficient_attention(q, k, v, attn_bias=None)
out = out.transpose(1, 2).reshape(B, C, H, W)
except NotImplementedError as e:
except NotImplementedError:
out = slice_attention(q.view(B, -1, C), k.view(B, -1, C).transpose(1, 2), v.view(B, -1, C).transpose(1, 2)).reshape(B, C, H, W)
return out
@@ -233,7 +232,7 @@ def pytorch_attention(q, k, v):
try:
out = torch.nn.functional.scaled_dot_product_attention(q, k, v, attn_mask=None, dropout_p=0.0, is_causal=False)
out = out.transpose(2, 3).reshape(B, C, H, W)
except model_management.OOM_EXCEPTION as e:
except model_management.OOM_EXCEPTION:
logging.warning("scaled_dot_product_attention OOMed: switched to slice attention")
out = slice_attention(q.view(B, -1, C), k.view(B, -1, C).transpose(1, 2), v.view(B, -1, C).transpose(1, 2)).reshape(B, C, H, W)
return out
@@ -546,7 +545,6 @@ class Decoder(nn.Module):
attn_op=AttnBlock,
**ignorekwargs):
super().__init__()
if use_linear_attn: attn_type = "linear"
self.ch = ch
self.temb_ch = 0
self.num_resolutions = len(ch_mult)
@@ -556,8 +554,7 @@ class Decoder(nn.Module):
self.give_pre_end = give_pre_end
self.tanh_out = tanh_out
# compute in_ch_mult, block_in and curr_res at lowest res
in_ch_mult = (1,)+tuple(ch_mult)
# compute block_in and curr_res at lowest res
block_in = ch*ch_mult[self.num_resolutions-1]
curr_res = resolution // 2**(self.num_resolutions-1)
self.z_shape = (1,z_channels,curr_res,curr_res)