Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

add *Swin MLP*: a hierarchical fully MLP architecture using shifted windows. #90

Closed
wants to merge 10 commits into from
Closed

add *Swin MLP*: a hierarchical fully MLP architecture using shifted windows. #90

Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
e39adac
add swin mlp
zeliu98 Jul 2, 2021
e0149d1
update swin_mlp_tiny_patch4_hd24_window8_256
zeliu98 Jul 2, 2021
336677d
add swin_tiny_patch4_hd24_window8_256.yaml
zeliu98 Jul 3, 2021
3b7b6a9
update
zeliu98 Jul 3, 2021
5343bc1
update
zeliu98 Jul 3, 2021
ddb0e50
update
zeliu98 Jul 3, 2021
bdb3f11
update
zeliu98 Jul 3, 2021
01f0c55
update config name
zeliu98 Jul 3, 2021
6875292
add swin_mlp_base_patch4_window7_224.yaml
zeliu98 Jul 3, 2021
97337da
add Swin MLP: a hierarchical fully MLP architecture using shifted win…
zeliu98 Jul 3, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
update
  • Loading branch information
@zeliu98
zeliu98 committed Jul 3, 2021
commit ddb0e50235038d1b6cfd2f174066ec6908ed2eb9
11 changes: 6 additions & 5 deletions models/swin_mlp.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,6 @@
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.checkpoint as checkpoint
import numpy as np
from timm.models.layers import DropPath, to_2tuple, trunc_normal_


Expand Down Expand Up @@ -167,12 +166,14 @@ def extra_repr(self) -> str:
def flops(self):
flops = 0
H, W = self.input_resolution
Hp = int(np.ceil(H / self.window_size)) * self.window_size
Wp = int(np.ceil(W / self.window_size)) * self.window_size
# norm1
flops += self.dim * H * W
# W-MSA/SW-MSA
nW = Hp * Wp / self.window_size / self.window_size

# Window/Shifted-Window Spatial MLP
if self.shift_size > 0:
nW = (H / self.window_size + 1) * (W / self.window_size + 1)
else:
nW = H * W / self.window_size / self.window_size
flops += nW * self.dim * (self.window_size * self.window_size) * (self.window_size * self.window_size)
# mlp
flops += 2 * H * W * self.dim * self.dim * self.mlp_ratio
Expand Down