Demosaic/experiment/RLAMBDA_DEMOSAIC20_R4_detach/log.txt

DataParallel(
  (module): RLAMBDANET(
    (sub_mean): MeanShift(3, 3, kernel_size=(1, 1), stride=(1, 1))
    (add_mean): MeanShift(3, 3, kernel_size=(1, 1), stride=(1, 1))
    (head): Sequential(
      (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    )
    (body): Sequential(
      (0): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (1): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (2): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (3): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (4): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (5): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (6): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (7): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (8): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (9): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (10): LambdaLayer(
        (to_q): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (to_k): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (to_v): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (norm_q): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (norm_v): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (pos_conv): Conv3d(4, 16, kernel_size=(1, 23, 23), stride=(1, 1, 1), padding=(0, 11, 11))
      )
      (11): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (12): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (13): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (14): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (15): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (16): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (17): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (18): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (19): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (20): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (21): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    )
    (tail): Sequential(
      (0): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    )
  )
)
[Epoch 1]	Learning rate: 1.00e-4
[1600/16000]	[L1: 2.9578]	51.6+0.5s
[3200/16000]	[L1: 2.0605]	47.0+0.1s
[4800/16000]	[L1: 1.5746]	49.1+0.0s
[6400/16000]	[L1: 1.3003]	49.3+0.0s
[8000/16000]	[L1: 1.1122]	50.0+0.0s
[9600/16000]	[L1: 0.9779]	49.1+0.0s
[11200/16000]	[L1: 0.8768]	50.7+0.1s
[12800/16000]	[L1: 0.7984]	51.7+0.1s
[14400/16000]	[L1: 0.7339]	51.7+0.1s
[16000/16000]	[L1: 0.6812]	51.6+0.0s

Evaluation:
DataParallel(
  (module): RLAMBDANET(
    (sub_mean): MeanShift(3, 3, kernel_size=(1, 1), stride=(1, 1))
    (add_mean): MeanShift(3, 3, kernel_size=(1, 1), stride=(1, 1))
    (head): Sequential(
      (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    )
    (body): Sequential(
      (0): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (1): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (2): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (3): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (4): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (5): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (6): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (7): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (8): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (9): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (10): LambdaLayer(
        (to_q): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (to_k): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (to_v): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (norm_q): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (norm_v): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (pos_conv): Conv3d(4, 16, kernel_size=(1, 23, 23), stride=(1, 1, 1), padding=(0, 11, 11))
      )
      (11): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (12): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (13): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (14): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (15): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (16): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (17): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (18): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (19): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (20): ResBlock(
        (body): Sequential(
          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): PReLU(num_parameters=1)
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (21): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    )
    (tail): Sequential(
      (0): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    )
  )
)
[Epoch 1]	Learning rate: 1.00e-4
[1600/16000]	[L1: 3.1531]	51.4+0.7s
[3200/16000]	[L1: 2.0836]	46.7+0.1s
[4800/16000]	[L1: 1.6088]	49.5+0.1s
[6400/16000]	[L1: 1.3442]	50.1+0.0s
[8000/16000]	[L1: 1.1582]	49.9+0.0s
[9600/16000]	[L1: 1.0251]	50.1+0.0s
[11200/16000]	[L1: 0.9277]	51.5+0.1s
[12800/16000]	[L1: 0.8480]	50.4+0.1s
[14400/16000]	[L1: 0.7845]	48.9+0.0s
[16000/16000]	[L1: 0.7314]	51.3+0.0s

Evaluation:
[DIV2K x1]	PSNR: 9.190 (Best: 9.190 @epoch 1)
Forward: 36.97s

Saving...
Total: 37.94s

[Epoch 2]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.2249]	51.2+0.9s
[3200/16000]	[L1: 0.2154]	50.0+0.0s
[4800/16000]	[L1: 0.2084]	49.8+0.0s
[6400/16000]	[L1: 0.2035]	50.9+0.0s
[8000/16000]	[L1: 0.1998]	50.0+0.0s
[9600/16000]	[L1: 0.1948]	50.5+0.0s
[11200/16000]	[L1: 0.1896]	49.9+0.0s
[12800/16000]	[L1: 0.1847]	50.0+0.0s
[14400/16000]	[L1: 0.1811]	49.3+0.0s
[16000/16000]	[L1: 0.1769]	50.1+0.0s

Evaluation:
[DIV2K x1]	PSNR: 8.918 (Best: 9.190 @epoch 1)
Forward: 36.55s

Saving...
Total: 37.02s

[Epoch 3]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.1230]	51.5+0.8s
[3200/16000]	[L1: 0.1251]	51.2+0.1s
[4800/16000]	[L1: 0.1235]	49.6+0.0s
[6400/16000]	[L1: 0.1227]	49.6+0.0s
[8000/16000]	[L1: 0.1213]	50.1+0.0s
[9600/16000]	[L1: 0.1195]	49.7+0.0s
[11200/16000]	[L1: 0.1169]	49.7+0.0s
[12800/16000]	[L1: 0.1151]	50.2+0.0s
[14400/16000]	[L1: 0.1131]	50.1+0.0s
[16000/16000]	[L1: 0.1112]	49.4+0.0s

Evaluation:
[DIV2K x1]	PSNR: 8.799 (Best: 9.190 @epoch 1)
Forward: 36.60s

Saving...
Total: 37.17s

[Epoch 4]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0919]	51.4+0.8s
[3200/16000]	[L1: 0.0908]	50.5+0.1s
[4800/16000]	[L1: 0.0883]	50.0+0.0s
[6400/16000]	[L1: 0.0867]	50.1+0.0s
[8000/16000]	[L1: 0.0855]	49.8+0.0s
[9600/16000]	[L1: 0.0841]	49.0+0.0s
[11200/16000]	[L1: 0.0830]	49.4+0.0s
[12800/16000]	[L1: 0.0819]	51.5+0.1s
[14400/16000]	[L1: 0.0811]	49.5+0.0s
[16000/16000]	[L1: 0.0798]	49.4+0.0s

Evaluation:
[DIV2K x1]	PSNR: 8.825 (Best: 9.190 @epoch 1)
Forward: 36.71s

Saving...
Total: 37.21s

[Epoch 5]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0681]	50.4+0.7s
[3200/16000]	[L1: 0.0675]	50.9+0.1s
[4800/16000]	[L1: 0.0663]	50.1+0.0s
[6400/16000]	[L1: 0.0652]	50.2+0.0s
[8000/16000]	[L1: 0.0637]	48.8+0.0s
[9600/16000]	[L1: 0.0631]	50.6+0.1s
[11200/16000]	[L1: 0.0623]	51.2+0.1s
[12800/16000]	[L1: 0.0618]	49.4+0.0s
[14400/16000]	[L1: 0.0615]	49.4+0.0s
[16000/16000]	[L1: 0.0606]	48.8+0.0s

Evaluation:
[DIV2K x1]	PSNR: 8.654 (Best: 9.190 @epoch 1)
Forward: 36.51s

Saving...
Total: 37.05s

[Epoch 6]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0569]	51.1+0.8s
[3200/16000]	[L1: 0.0561]	51.2+0.1s
[4800/16000]	[L1: 0.0533]	49.4+0.0s
[6400/16000]	[L1: 0.0519]	50.4+0.0s
[8000/16000]	[L1: 0.0514]	49.7+0.0s
[9600/16000]	[L1: 0.0505]	50.5+0.0s
[11200/16000]	[L1: 0.0496]	49.3+0.0s
[12800/16000]	[L1: 0.0490]	49.5+0.0s
[14400/16000]	[L1: 0.0484]	49.9+0.0s
[16000/16000]	[L1: 0.0482]	49.7+0.0s

Evaluation:
[DIV2K x1]	PSNR: 8.628 (Best: 9.190 @epoch 1)
Forward: 36.45s

Saving...
Total: 37.06s

[Epoch 7]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0443]	49.7+0.9s
[3200/16000]	[L1: 0.0433]	49.4+0.0s
[4800/16000]	[L1: 0.0429]	50.7+0.1s
[6400/16000]	[L1: 0.0425]	49.8+0.0s
[8000/16000]	[L1: 0.0425]	48.9+0.0s
[9600/16000]	[L1: 0.0416]	50.4+0.0s
[11200/16000]	[L1: 0.0411]	49.2+0.0s
[12800/16000]	[L1: 0.0403]	51.3+0.0s
[14400/16000]	[L1: 0.0400]	49.7+0.0s
[16000/16000]	[L1: 0.0396]	50.1+0.0s

Evaluation:
[DIV2K x1]	PSNR: 9.004 (Best: 9.190 @epoch 1)
Forward: 36.78s

Saving...
Total: 37.24s

[Epoch 8]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0332]	50.7+0.7s
[3200/16000]	[L1: 0.0352]	49.9+0.0s
[4800/16000]	[L1: 0.0351]	49.4+0.0s
[6400/16000]	[L1: 0.0351]	49.9+0.0s
[8000/16000]	[L1: 0.0348]	50.1+0.0s
[9600/16000]	[L1: 0.0348]	50.0+0.0s
[11200/16000]	[L1: 0.0345]	51.5+0.1s
[12800/16000]	[L1: 0.0343]	49.8+0.0s
[14400/16000]	[L1: 0.0341]	49.8+0.0s
[16000/16000]	[L1: 0.0339]	49.2+0.0s

Evaluation:
[DIV2K x1]	PSNR: 9.533 (Best: 9.533 @epoch 8)
Forward: 36.63s

Saving...
Total: 37.26s

[Epoch 9]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0421]	51.3+0.8s
[3200/16000]	[L1: 0.0368]	50.5+0.1s
[4800/16000]	[L1: 0.0346]	50.5+0.0s
[6400/16000]	[L1: 0.0341]	49.8+0.0s
[8000/16000]	[L1: 0.0328]	50.5+0.1s
[9600/16000]	[L1: 0.0322]	48.7+0.0s
[11200/16000]	[L1: 0.0314]	51.1+0.0s
[12800/16000]	[L1: 0.0307]	50.9+0.1s
[14400/16000]	[L1: 0.0299]	50.8+0.1s
[16000/16000]	[L1: 0.0296]	50.9+0.0s

Evaluation:
[DIV2K x1]	PSNR: 9.870 (Best: 9.870 @epoch 9)
Forward: 36.44s

Saving...
Total: 37.06s

[Epoch 10]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0260]	50.7+0.9s
[3200/16000]	[L1: 0.0265]	50.2+0.1s
[4800/16000]	[L1: 0.0264]	49.9+0.0s
[6400/16000]	[L1: 0.0257]	49.5+0.0s
[8000/16000]	[L1: 0.0252]	49.2+0.0s
[9600/16000]	[L1: 0.0249]	50.0+0.0s
[11200/16000]	[L1: 0.0250]	49.6+0.0s
[12800/16000]	[L1: 0.0248]	50.4+0.0s
[14400/16000]	[L1: 0.0246]	51.4+0.1s
[16000/16000]	[L1: 0.0244]	49.9+0.0s

Evaluation:
[DIV2K x1]	PSNR: 10.933 (Best: 10.933 @epoch 10)
Forward: 36.68s

Saving...
Total: 37.29s

[Epoch 11]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0217]	51.0+0.9s
[3200/16000]	[L1: 0.0229]	50.5+0.1s
[4800/16000]	[L1: 0.0228]	49.0+0.1s
[6400/16000]	[L1: 0.0230]	49.8+0.0s
[8000/16000]	[L1: 0.0233]	49.5+0.0s
[9600/16000]	[L1: 0.0231]	49.3+0.0s
[11200/16000]	[L1: 0.0226]	49.7+0.0s
[12800/16000]	[L1: 0.0223]	50.7+0.1s
[14400/16000]	[L1: 0.0221]	51.0+0.0s
[16000/16000]	[L1: 0.0219]	50.1+0.0s

Evaluation:
[DIV2K x1]	PSNR: 12.240 (Best: 12.240 @epoch 11)
Forward: 36.70s

Saving...
Total: 37.18s

[Epoch 12]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0201]	51.4+0.8s
[3200/16000]	[L1: 0.0197]	49.1+0.0s
[4800/16000]	[L1: 0.0191]	49.1+0.0s
[6400/16000]	[L1: 0.0194]	50.3+0.1s
[8000/16000]	[L1: 0.0189]	51.3+0.1s
[9600/16000]	[L1: 0.0186]	51.1+0.1s
[11200/16000]	[L1: 0.0185]	50.2+0.1s
[12800/16000]	[L1: 0.0183]	50.3+0.0s
[14400/16000]	[L1: 0.0181]	49.3+0.0s
[16000/16000]	[L1: 0.0179]	50.0+0.0s

Evaluation:
[DIV2K x1]	PSNR: 13.591 (Best: 13.591 @epoch 12)
Forward: 36.60s

Saving...
Total: 37.12s

[Epoch 13]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0164]	50.8+0.7s
[3200/16000]	[L1: 0.0163]	50.0+0.0s
[4800/16000]	[L1: 0.0160]	50.6+0.1s
[6400/16000]	[L1: 0.0163]	51.2+0.0s
[8000/16000]	[L1: 0.0166]	50.4+0.0s
[9600/16000]	[L1: 0.0167]	50.2+0.0s
[11200/16000]	[L1: 0.0167]	50.0+0.0s
[12800/16000]	[L1: 0.0165]	50.7+0.1s
[14400/16000]	[L1: 0.0164]	49.6+0.0s
[16000/16000]	[L1: 0.0164]	49.2+0.0s

Evaluation:
[DIV2K x1]	PSNR: 15.362 (Best: 15.362 @epoch 13)
Forward: 36.53s

Saving...
Total: 37.11s

[Epoch 14]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0146]	51.1+0.7s
[3200/16000]	[L1: 0.0151]	50.5+0.1s
[4800/16000]	[L1: 0.0158]	50.9+0.1s
[6400/16000]	[L1: 0.0154]	50.0+0.0s
[8000/16000]	[L1: 0.0152]	50.9+0.1s
[9600/16000]	[L1: 0.0150]	50.4+0.1s
[11200/16000]	[L1: 0.0149]	49.2+0.0s
[12800/16000]	[L1: 0.0147]	50.5+0.1s
[14400/16000]	[L1: 0.0146]	50.7+0.1s
[16000/16000]	[L1: 0.0146]	50.9+0.0s

Evaluation:
[DIV2K x1]	PSNR: 17.182 (Best: 17.182 @epoch 14)
Forward: 36.60s

Saving...
Total: 37.12s

[Epoch 15]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0128]	51.5+0.7s
[3200/16000]	[L1: 0.0132]	51.5+0.1s
[4800/16000]	[L1: 0.0133]	51.5+0.1s
[6400/16000]	[L1: 0.0133]	50.3+0.0s
[8000/16000]	[L1: 0.0132]	51.2+0.1s
[9600/16000]	[L1: 0.0131]	50.3+0.0s
[11200/16000]	[L1: 0.0130]	50.8+0.0s
[12800/16000]	[L1: 0.0130]	51.2+0.0s
[14400/16000]	[L1: 0.0130]	51.3+0.1s
[16000/16000]	[L1: 0.0129]	50.9+0.0s

Evaluation:
[DIV2K x1]	PSNR: 18.605 (Best: 18.605 @epoch 15)
Forward: 36.79s

Saving...
Total: 37.34s

[Epoch 16]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0129]	50.9+0.8s
[3200/16000]	[L1: 0.0132]	49.6+0.1s
[4800/16000]	[L1: 0.0129]	50.5+0.0s
[6400/16000]	[L1: 0.0130]	50.4+0.0s
[8000/16000]	[L1: 0.0128]	49.5+0.0s
[9600/16000]	[L1: 0.0127]	50.5+0.0s
[11200/16000]	[L1: 0.0126]	48.9+0.0s
[12800/16000]	[L1: 0.0126]	49.1+0.0s
[14400/16000]	[L1: 0.0124]	51.1+0.1s
[16000/16000]	[L1: 0.0123]	49.8+0.0s

Evaluation:
[DIV2K x1]	PSNR: 19.953 (Best: 19.953 @epoch 16)
Forward: 36.52s

Saving...
Total: 37.15s

[Epoch 17]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0124]	50.7+0.8s
[3200/16000]	[L1: 0.0120]	49.7+0.0s
[4800/16000]	[L1: 0.0118]	49.9+0.0s
[6400/16000]	[L1: 0.0117]	50.1+0.0s
[8000/16000]	[L1: 0.0117]	51.0+0.1s
[9600/16000]	[L1: 0.0117]	49.8+0.0s
[11200/16000]	[L1: 0.0118]	49.6+0.0s
[12800/16000]	[L1: 0.0119]	50.4+0.1s
[14400/16000]	[L1: 0.0118]	51.3+0.1s
[16000/16000]	[L1: 0.0117]	49.3+0.0s

Evaluation:
[DIV2K x1]	PSNR: 20.834 (Best: 20.834 @epoch 17)
Forward: 36.51s

Saving...
Total: 37.07s

[Epoch 18]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0106]	50.8+0.8s
[3200/16000]	[L1: 0.0110]	48.7+0.0s
[4800/16000]	[L1: 0.0108]	49.7+0.0s
[6400/16000]	[L1: 0.0108]	48.5+0.0s
[8000/16000]	[L1: 0.0108]	49.8+0.0s
[9600/16000]	[L1: 0.0110]	51.0+0.1s
[11200/16000]	[L1: 0.0110]	51.2+0.1s
[12800/16000]	[L1: 0.0109]	51.1+0.0s
[14400/16000]	[L1: 0.0109]	48.8+0.0s
[16000/16000]	[L1: 0.0109]	50.0+0.0s

Evaluation:
[DIV2K x1]	PSNR: 21.466 (Best: 21.466 @epoch 18)
Forward: 36.63s

Saving...
Total: 37.17s

[Epoch 19]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0107]	50.9+0.9s
[3200/16000]	[L1: 0.0107]	49.2+0.0s
[4800/16000]	[L1: 0.0108]	49.0+0.0s
[6400/16000]	[L1: 0.0109]	49.7+0.0s
[8000/16000]	[L1: 0.0108]	51.0+0.1s
[9600/16000]	[L1: 0.0107]	49.9+0.0s
[11200/16000]	[L1: 0.0107]	49.6+0.0s
[12800/16000]	[L1: 0.0106]	50.0+0.0s
[14400/16000]	[L1: 0.0105]	49.5+0.0s
[16000/16000]	[L1: 0.0105]	49.1+0.0s

Evaluation:
[DIV2K x1]	PSNR: 21.952 (Best: 21.952 @epoch 19)
Forward: 36.57s

Saving...
Total: 37.13s

[Epoch 20]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0104]	51.2+0.7s
[3200/16000]	[L1: 0.0104]	50.8+0.1s
[4800/16000]	[L1: 0.0104]	51.1+0.1s
[6400/16000]	[L1: 0.0103]	50.6+0.0s
[8000/16000]	[L1: 0.0103]	50.3+0.0s
[9600/16000]	[L1: 0.0104]	50.5+0.0s
[11200/16000]	[L1: 0.0104]	50.5+0.0s
[12800/16000]	[L1: 0.0103]	51.5+0.1s
[14400/16000]	[L1: 0.0103]	49.3+0.0s
[16000/16000]	[L1: 0.0103]	51.4+0.0s

Evaluation:
[DIV2K x1]	PSNR: 22.624 (Best: 22.624 @epoch 20)
Forward: 36.49s

Saving...
Total: 37.04s

[Epoch 21]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0098]	50.9+0.7s
[3200/16000]	[L1: 0.0097]	49.0+0.0s
[4800/16000]	[L1: 0.0097]	51.0+0.1s
[6400/16000]	[L1: 0.0097]	49.9+0.0s
[8000/16000]	[L1: 0.0096]	49.5+0.0s
[9600/16000]	[L1: 0.0096]	48.7+0.0s
[11200/16000]	[L1: 0.0096]	48.8+0.0s
[12800/16000]	[L1: 0.0096]	48.8+0.0s
[14400/16000]	[L1: 0.0096]	50.4+0.0s
[16000/16000]	[L1: 0.0096]	50.7+0.0s

Evaluation:
[DIV2K x1]	PSNR: 22.887 (Best: 22.887 @epoch 21)
Forward: 36.52s

Saving...
Total: 37.15s

[Epoch 22]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0096]	50.9+0.9s
[3200/16000]	[L1: 0.0096]	50.8+0.1s
[4800/16000]	[L1: 0.0095]	51.3+0.1s
[6400/16000]	[L1: 0.0097]	51.1+0.1s
[8000/16000]	[L1: 0.0096]	51.1+0.1s
[9600/16000]	[L1: 0.0095]	49.8+0.0s
[11200/16000]	[L1: 0.0095]	48.9+0.0s
[12800/16000]	[L1: 0.0094]	49.0+0.0s
[14400/16000]	[L1: 0.0094]	50.2+0.0s
[16000/16000]	[L1: 0.0094]	49.0+0.0s

Evaluation:
[DIV2K x1]	PSNR: 23.476 (Best: 23.476 @epoch 22)
Forward: 36.75s

Saving...
Total: 37.30s

[Epoch 23]	Learning rate: 1.00e-4
[1600/16000]	[L1: 0.0096]	50.2+0.7s
[3200/16000]	[L1: 0.1840]	50.5+0.1s
[4800/16000]	[L1: 0.1394]	50.3+0.0s
[6400/16000]	[L1: 0.1132]	50.1+0.1s
[8000/16000]	[L1: 0.0961]	50.7+0.1s
[9600/16000]	[L1: 0.0840]	50.4+0.0s
[11200/16000]	[L1: 0.0751]	50.7+0.0s