pfcfuse/logs/log_20241006_170059.log

2.4.1+cu121
True
Model: PFCFuse
Number of epochs: 60
Epoch gap: 40
Learning rate: 0.0001
Weight decay: 0
Batch size: 1
GPU number: 0
Coefficient of MSE loss VF: 1.0
Coefficient of MSE loss IF: 1.0
Coefficient of RMI loss VF: 1.0
Coefficient of RMI loss IF: 1.0
Coefficient of Cosine loss VF: 1.0
Coefficient of Cosine loss IF: 1.0
Coefficient of Decomposition loss: 2.0
Coefficient of Total Variation loss: 5.0
Clip gradient norm value: 0.01
Optimization step: 20
Optimization gamma: 0.5

[Epoch 0/60] [Batch 0/6487] [loss: 18.997314] ETA: 9 days, 21
[Epoch 0/60] [Batch 1000/6487] [loss: 0.168454] ETA: 9:49:11.56
[Epoch 0/60] [Batch 2000/6487] [loss: 0.350209] ETA: 9:50:02.48
[Epoch 0/60] [Batch 3000/6487] [loss: 0.069812] ETA: 9:50:01.39
[Epoch 0/60] [Batch 4000/6487] [loss: 0.065885] ETA: 10:07:51.9
[Epoch 0/60] [Batch 5000/6487] [loss: 0.245787] ETA: 9:54:20.43
[Epoch 0/60] [Batch 6000/6487] [loss: 0.042268] ETA: 10:03:01.5
[Epoch 1/60] [Batch 0/6487] [loss: 0.066391] ETA: 9:49:42.62
[Epoch 1/60] [Batch 1000/6487] [loss: 0.261340] ETA: 9:53:01.14
[Epoch 1/60] [Batch 2000/6487] [loss: 0.095140] ETA: 10:00:23.2
[Epoch 1/60] [Batch 3000/6487] [loss: 0.028199] ETA: 10:17:34.6
[Epoch 1/60] [Batch 4000/6487] [loss: 0.089190] ETA: 9:03:05.02
[Epoch 1/60] [Batch 5000/6487] [loss: 0.134494] ETA: 9:40:49.06
[Epoch 1/60] [Batch 6000/6487] [loss: 0.134911] ETA: 10:01:31.0
[Epoch 2/60] [Batch 0/6487] [loss: 0.105676] ETA: 9:39:54.85
[Epoch 2/60] [Batch 1000/6487] [loss: 0.048246] ETA: 9:53:22.48
[Epoch 2/60] [Batch 2000/6487] [loss: 0.028375] ETA: 9:40:39.29
[Epoch 2/60] [Batch 3000/6487] [loss: 0.034128] ETA: 9:56:36.89
[Epoch 2/60] [Batch 4000/6487] [loss: 0.022569] ETA: 9:42:22.52
[Epoch 2/60] [Batch 5000/6487] [loss: 0.055698] ETA: 9:37:40.30
[Epoch 2/60] [Batch 6000/6487] [loss: 0.056826] ETA: 9:44:44.57
[Epoch 3/60] [Batch 0/6487] [loss: 0.020840] ETA: 9:25:04.54
[Epoch 3/60] [Batch 1000/6487] [loss: 0.033522] ETA: 9:21:26.95
[Epoch 3/60] [Batch 2000/6487] [loss: 0.087376] ETA: 9:33:31.90
[Epoch 3/60] [Batch 3000/6487] [loss: 0.054073] ETA: 9:33:48.42
[Epoch 3/60] [Batch 4000/6487] [loss: 0.015202] ETA: 9:35:51.08
[Epoch 3/60] [Batch 5000/6487] [loss: 0.023986] ETA: 9:13:47.44
[Epoch 3/60] [Batch 6000/6487] [loss: 0.027252] ETA: 9:12:56.42
[Epoch 4/60] [Batch 0/6487] [loss: 0.028229] ETA: 9:07:31.99
[Epoch 4/60] [Batch 1000/6487] [loss: 0.009431] ETA: 9:17:32.62
[Epoch 4/60] [Batch 2000/6487] [loss: 0.019674] ETA: 9:11:41.71
[Epoch 4/60] [Batch 3000/6487] [loss: 0.036560] ETA: 9:13:51.69
[Epoch 4/60] [Batch 4000/6487] [loss: 0.023415] ETA: 9:04:01.70
[Epoch 4/60] [Batch 5000/6487] [loss: 0.040478] ETA: 8:51:47.13
[Epoch 4/60] [Batch 6000/6487] [loss: 0.082253] ETA: 9:05:46.11
[Epoch 5/60] [Batch 0/6487] [loss: 0.071554] ETA: 9:08:54.46
[Epoch 5/60] [Batch 1000/6487] [loss: 0.049842] ETA: 8:58:09.51
[Epoch 5/60] [Batch 2000/6487] [loss: 0.010863] ETA: 8:56:36.05
[Epoch 5/60] [Batch 3000/6487] [loss: 0.043838] ETA: 9:04:35.75
[Epoch 5/60] [Batch 4000/6487] [loss: 0.108259] ETA: 8:57:50.84
[Epoch 5/60] [Batch 5000/6487] [loss: 0.024255] ETA: 8:58:37.17
[Epoch 5/60] [Batch 6000/6487] [loss: 0.040892] ETA: 8:55:42.00
[Epoch 6/60] [Batch 0/6487] [loss: 0.023312] ETA: 8:54:04.34
[Epoch 6/60] [Batch 1000/6487] [loss: 0.035162] ETA: 8:54:12.47
[Epoch 6/60] [Batch 2000/6487] [loss: 0.015531] ETA: 8:53:49.38
[Epoch 6/60] [Batch 3000/6487] [loss: 0.077984] ETA: 8:48:18.87
[Epoch 6/60] [Batch 4000/6487] [loss: 0.020643] ETA: 9:11:35.39
[Epoch 6/60] [Batch 5000/6487] [loss: 0.018760] ETA: 9:03:14.95
[Epoch 6/60] [Batch 6000/6487] [loss: 0.064921] ETA: 8:45:53.60
[Epoch 7/60] [Batch 0/6487] [loss: 0.020156] ETA: 8:57:09.24
[Epoch 7/60] [Batch 1000/6487] [loss: 0.023395] ETA: 8:24:13.53
[Epoch 7/60] [Batch 2000/6487] [loss: 0.029281] ETA: 8:43:02.96
[Epoch 7/60] [Batch 3000/6487] [loss: 0.034845] ETA: 8:42:22.83
[Epoch 7/60] [Batch 4000/6487] [loss: 0.003461] ETA: 8:38:22.20
[Epoch 7/60] [Batch 5000/6487] [loss: 0.017755] ETA: 8:34:32.13
[Epoch 7/60] [Batch 6000/6487] [loss: 0.029525] ETA: 8:56:35.38
[Epoch 8/60] [Batch 0/6487] [loss: 0.013173] ETA: 8:39:37.20
[Epoch 8/60] [Batch 1000/6487] [loss: 0.006751] ETA: 9:16:29.80
[Epoch 8/60] [Batch 2000/6487] [loss: 0.017919] ETA: 8:39:39.59
[Epoch 8/60] [Batch 3000/6487] [loss: 0.194062] ETA: 8:26:52.51
[Epoch 8/60] [Batch 4000/6487] [loss: 0.043372] ETA: 8:28:11.61
[Epoch 8/60] [Batch 5000/6487] [loss: 0.033920] ETA: 8:30:32.83
[Epoch 8/60] [Batch 6000/6487] [loss: 0.008135] ETA: 8:38:24.67
[Epoch 9/60] [Batch 0/6487] [loss: 0.008534] ETA: 8:22:37.70
[Epoch 9/60] [Batch 1000/64
修改代码结构，提高可读性和可维护性；调整训练输出频率。改进 self.enhancement_module 为 self.enhancement_module = WTConv2d(32, 32) 2024-10-09 11:35:06 +08:00			`2.4.1+cu121`
			`True`
			`Model: PFCFuse`
			`Number of epochs: 60`
			`Epoch gap: 40`
			`Learning rate: 0.0001`
			`Weight decay: 0`
			`Batch size: 1`
			`GPU number: 0`
			`Coefficient of MSE loss VF: 1.0`
			`Coefficient of MSE loss IF: 1.0`
			`Coefficient of RMI loss VF: 1.0`
			`Coefficient of RMI loss IF: 1.0`
			`Coefficient of Cosine loss VF: 1.0`
			`Coefficient of Cosine loss IF: 1.0`
			`Coefficient of Decomposition loss: 2.0`
			`Coefficient of Total Variation loss: 5.0`
			`Clip gradient norm value: 0.01`
			`Optimization step: 20`
			`Optimization gamma: 0.5`
			[Epoch 0/60] [Batch 0/6487] [loss: 18.997314] ETA: 9 days, 21 [Epoch 0/60] [Batch 1000/6487] [loss: 0.168454] ETA: 9:49:11.56 [Epoch 0/60] [Batch 2000/6487] [loss: 0.350209] ETA: 9:50:02.48 [Epoch 0/60] [Batch 3000/6487] [loss: 0.069812] ETA: 9:50:01.39 [Epoch 0/60] [Batch 4000/6487] [loss: 0.065885] ETA: 10:07:51.9 [Epoch 0/60] [Batch 5000/6487] [loss: 0.245787] ETA: 9:54:20.43 [Epoch 0/60] [Batch 6000/6487] [loss: 0.042268] ETA: 10:03:01.5 [Epoch 1/60] [Batch 0/6487] [loss: 0.066391] ETA: 9:49:42.62 [Epoch 1/60] [Batch 1000/6487] [loss: 0.261340] ETA: 9:53:01.14 [Epoch 1/60] [Batch 2000/6487] [loss: 0.095140] ETA: 10:00:23.2 [Epoch 1/60] [Batch 3000/6487] [loss: 0.028199] ETA: 10:17:34.6 [Epoch 1/60] [Batch 4000/6487] [loss: 0.089190] ETA: 9:03:05.02 [Epoch 1/60] [Batch 5000/6487] [loss: 0.134494] ETA: 9:40:49.06 [Epoch 1/60] [Batch 6000/6487] [loss: 0.134911] ETA: 10:01:31.0 [Epoch 2/60] [Batch 0/6487] [loss: 0.105676] ETA: 9:39:54.85 [Epoch 2/60] [Batch 1000/6487] [loss: 0.048246] ETA: 9:53:22.48 [Epoch 2/60] [Batch 2000/6487] [loss: 0.028375] ETA: 9:40:39.29 [Epoch 2/60] [Batch 3000/6487] [loss: 0.034128] ETA: 9:56:36.89 [Epoch 2/60] [Batch 4000/6487] [loss: 0.022569] ETA: 9:42:22.52 [Epoch 2/60] [Batch 5000/6487] [loss: 0.055698] ETA: 9:37:40.30 [Epoch 2/60] [Batch 6000/6487] [loss: 0.056826] ETA: 9:44:44.57 [Epoch 3/60] [Batch 0/6487] [loss: 0.020840] ETA: 9:25:04.54 [Epoch 3/60] [Batch 1000/6487] [loss: 0.033522] ETA: 9:21:26.95 [Epoch 3/60] [Batch 2000/6487] [loss: 0.087376] ETA: 9:33:31.90 [Epoch 3/60] [Batch 3000/6487] [loss: 0.054073] ETA: 9:33:48.42 [Epoch 3/60] [Batch 4000/6487] [loss: 0.015202] ETA: 9:35:51.08 [Epoch 3/60] [Batch 5000/6487] [loss: 0.023986] ETA: 9:13:47.44 [Epoch 3/60] [Batch 6000/6487] [loss: 0.027252] ETA: 9:12:56.42 [Epoch 4/60] [Batch 0/6487] [loss: 0.028229] ETA: 9:07:31.99 [Epoch 4/60] [Batch 1000/6487] [loss: 0.009431] ETA: 9:17:32.62 [Epoch 4/60] [Batch 2000/6487] [loss: 0.019674] ETA: 9:11:41.71 [Epoch 4/60] [Batch 3000/6487] [loss: 0.036560] ETA: 9:13:51.69 [Epoch 4/60] [Batch 4000/6487] [loss: 0.023415] ETA: 9:04:01.70 [Epoch 4/60] [Batch 5000/6487] [loss: 0.040478] ETA: 8:51:47.13 [Epoch 4/60] [Batch 6000/6487] [loss: 0.082253] ETA: 9:05:46.11 [Epoch 5/60] [Batch 0/6487] [loss: 0.071554] ETA: 9:08:54.46 [Epoch 5/60] [Batch 1000/6487] [loss: 0.049842] ETA: 8:58:09.51 [Epoch 5/60] [Batch 2000/6487] [loss: 0.010863] ETA: 8:56:36.05 [Epoch 5/60] [Batch 3000/6487] [loss: 0.043838] ETA: 9:04:35.75 [Epoch 5/60] [Batch 4000/6487] [loss: 0.108259] ETA: 8:57:50.84 [Epoch 5/60] [Batch 5000/6487] [loss: 0.024255] ETA: 8:58:37.17 [Epoch 5/60] [Batch 6000/6487] [loss: 0.040892] ETA: 8:55:42.00 [Epoch 6/60] [Batch 0/6487] [loss: 0.023312] ETA: 8:54:04.34 [Epoch 6/60] [Batch 1000/6487] [loss: 0.035162] ETA: 8:54:12.47 [Epoch 6/60] [Batch 2000/6487] [loss: 0.015531] ETA: 8:53:49.38 [Epoch 6/60] [Batch 3000/6487] [loss: 0.077984] ETA: 8:48:18.87 [Epoch 6/60] [Batch 4000/6487] [loss: 0.020643] ETA: 9:11:35.39 [Epoch 6/60] [Batch 5000/6487] [loss: 0.018760] ETA: 9:03:14.95 [Epoch 6/60] [Batch 6000/6487] [loss: 0.064921] ETA: 8:45:53.60 [Epoch 7/60] [Batch 0/6487] [loss: 0.020156] ETA: 8:57:09.24 [Epoch 7/60] [Batch 1000/6487] [loss: 0.023395] ETA: 8:24:13.53 [Epoch 7/60] [Batch 2000/6487] [loss: 0.029281] ETA: 8:43:02.96 [Epoch 7/60] [Batch 3000/6487] [loss: 0.034845] ETA: 8:42:22.83 [Epoch 7/60] [Batch 4000/6487] [loss: 0.003461] ETA: 8:38:22.20 [Epoch 7/60] [Batch 5000/6487] [loss: 0.017755] ETA: 8:34:32.13 [Epoch 7/60] [Batch 6000/6487] [loss: 0.029525] ETA: 8:56:35.38 [Epoch 8/60] [Batch 0/6487] [loss: 0.013173] ETA: 8:39:37.20 [Epoch 8/60] [Batch 1000/6487] [loss: 0.006751] ETA: 9:16:29.80 [Epoch 8/60] [Batch 2000/6487] [loss: 0.017919] ETA: 8:39:39.59 [Epoch 8/60] [Batch 3000/6487] [loss: 0.194062] ETA: 8:26:52.51 [Epoch 8/60] [Batch 4000/6487] [loss: 0.043372] ETA: 8:28:11.61 [Epoch 8/60] [Batch 5000/6487] [loss: 0.033920] ETA: 8:30:32.83 [Epoch 8/60] [Batch 6000/6487] [loss: 0.008135] ETA: 8:38:24.67 [Epoch 9/60] [Batch 0/6487] [loss: 0.008534] ETA: 8:22:37.70 [Epoch 9/60] [Batch 1000/64