123456/roma_unsb
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fixing CTN

Browse Source
This commit is contained in:
bishe 2025-02-23 19:06:35 +08:00
parent 1caa5f0625
commit 6705075876
3 changed files with 16 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
checkpoints/ROMA_UNSB_001/loss_log.txt
View File

@ -25,3 +25,11 @@
================ Training Loss (Sun Feb 23 18:39:21 2025) ================
================ Training Loss (Sun Feb 23 18:40:15 2025) ================
================ Training Loss (Sun Feb 23 18:41:15 2025) ================
================ Training Loss (Sun Feb 23 18:47:46 2025) ================
================ Training Loss (Sun Feb 23 18:48:36 2025) ================
================ Training Loss (Sun Feb 23 18:50:20 2025) ================
================ Training Loss (Sun Feb 23 18:51:50 2025) ================
================ Training Loss (Sun Feb 23 18:58:45 2025) ================
================ Training Loss (Sun Feb 23 18:59:52 2025) ================
================ Training Loss (Sun Feb 23 19:03:05 2025) ================
================ Training Loss (Sun Feb 23 19:03:57 2025) ================

BIN
models/__pycache__/roma_unsb_model.cpython-39.pyc
View File

Binary file not shown.

13
models/roma_unsb_model.py
View File

@ -166,6 +166,7 @@ class ContentAwareTemporalNorm(nn.Module):
Returns:
F_content: [B, 2, H, W] 生成的光流场(x/y方向位移)
"""
print(weight_map.shape)
B, _, H, W = weight_map.shape
# 1. 归一化权重图
@ -403,8 +404,8 @@ class RomaUnsbModel(BaseModel):
print(f'before resize: {self.real_A0.shape}')
real_A0 = self.resize(self.real_A0)
real_A1 = self.resize(self.real_A1)
real_B0 = self.resize(self.real_B0)
real_B1 = self.resize(self.real_B1)
real_B0 = self.resize(self.real_B0).requires_grad_(True)
real_B1 = self.resize(self.real_B1).requires_grad_(True)
# 使用VIT
print(f'before vit: {real_A0.shape}')
@ -526,12 +527,14 @@ class RomaUnsbModel(BaseModel):
setattr(self, "fake_"+str(t+1), Xt_1)
if self.opt.phase == 'train':
print(f'real_B0.shape = {real_B0.shape} fake_B0.shape = {self.fake_B0.shape}')
print(f"self.real_B0.requires_grad: {real_B0.requires_grad}")
# 真实图像的梯度
real_gradient = torch.autograd.grad(self.real_B0.sum(), self.real_B0, create_graph=True)[0]
real_gradient = torch.autograd.grad(real_B0.sum(), real_B0, create_graph=True)[0]
# 生成图像的梯度
fake_gradient = torch.autograd.grad(self.fake_B.sum(), self.fake_B, create_graph=True)[0]
fake_gradient = torch.autograd.grad(self.fake_B0.sum(), self.fake_B0, create_graph=True)[0]
# 梯度图
self.weight_real, self.weight_fake = self.cao.generate_weight_map(real_gradient, fake_gradient)
self.weight_real, self.weight_fake = self.cao.generate_weight_map(fake_gradient)
# 生成图像的CTN光流图
self.f_content = self.ctn(self.weight_fake)