cptrans复现

models/roma_unsb_model.py

@@ -227,7 +227,7 @@ class RomaUnsbModel(BaseModel):
         
         # 指定需要打印的训练损失
         self.loss_names = ['G_GAN', 'D_ViT', 'G', 'global', 'spatial','ctn']
-        self.visual_names = ['real_A0', 'fake_B0_1','fake_B0', 'real_B0','real_A1', 'fake_B1_1', 'fake_B1', 'real_B1']
+        self.visual_names = ['real_A0', 'fake_B0', 'real_B0','real_A1',  'fake_B1', 'real_B1']
         self.atten_layers = [int(i) for i in self.opt.atten_layers.split(',')]
         
 
@@ -380,8 +380,8 @@ class RomaUnsbModel(BaseModel):
         """计算生成器的 GAN 损失"""
         if self.opt.lambda_ctn > 0.0:
             # 生成图像的CTN光流图
-            self.f_content0 = self.ctn(self.weight_fake0)
-            self.f_content1 = self.ctn(self.weight_fake1)
+            self.f_content0 = self.ctn(self.weight_fake0.detach())
+            self.f_content1 = self.ctn(self.weight_fake1.detach())
             
             # 变换后的图片
             self.warped_real_A0 = warp(self.real_A0, self.f_content0)
@@ -429,8 +429,8 @@ class RomaUnsbModel(BaseModel):
         n_layers = len(self.atten_layers)
         mutil_real_A0_tokens = self.mutil_real_A0_tokens
         mutil_real_A1_tokens = self.mutil_real_A1_tokens
-        mutil_fake_B0_tokens = self.mutil_fake_B0_tokens_list[-1]
-        mutil_fake_B1_tokens = self.mutil_fake_B1_tokens_list[-1]
+        mutil_fake_B0_tokens = self.mutil_fake_B0_tokens
+        mutil_fake_B1_tokens = self.mutil_fake_B1_tokens
 
             
         if self.opt.lambda_global > 0.0:

scripts/train.sh

@@ -17,7 +17,7 @@ python train.py \
     --lambda_global 6.0 \
     --gamma_stride 20 \
     --lr 0.000002 \
-    --gpu_id 1 \
+    --gpu_id 0 \
     --nce_idt False \
     --netF mlp_sample \
     --eta_ratio 0.4 \

scripts/traincp.sh

@@ -1,6 +1,6 @@
 python train.py \
     --dataroot /home/openxs/kunyu/datasets/InfraredCity-Lite/Double/Moitor \
-    --name cp_1 \
+    --name cp_2 \
     --dataset_mode unaligned_double \
     --display_env CP \
     --model roma_unsb \
@@ -9,9 +9,10 @@ python train.py \
     --lambda_global 6.0 \
     --lambda_spatial 6.0 \
     --gamma_stride 20 \
-    --lr 0.000001 \
-    --gpu_id 2 \
+    --lr 0.000002 \
+    --gpu_id 0 \
     --eta_ratio 0.4 \
     --n_epochs 100 \
     --n_epochs_decay 100 \
-# cp1 复现cptrans的效果
+# cp1 复现cptrans的效果 --lr 0.000001
+# cp2 修了一下cp1的代码，--lr 0.000002