Hugging Face's logo

zwa73
/
SoulTide-AudioData-Model

Japanese
Model card Files
xet
Community
zwa73 commited on
Commit
1807e40
·
1 Parent(s): a6c05ee

sovits 3

Browse files
Files changed (24) hide show
  1. so-vits-svc-4.0/20230318-SilenusNS-44k/best/D_12800.pth +3 -0
  2. so-vits-svc-4.0/20230318-SilenusNS-44k/best/G_12800.pth +3 -0
  3. so-vits-svc-4.0/20230318-SilenusNS-44k/best/config.json +93 -0
  4. so-vits-svc-4.0/20230318-SilenusNS-44k/config.json +93 -0
  5. so-vits-svc-4.0/20230318-SilenusNS-44k/kmeans_10000.pt +3 -0
  6. so-vits-svc-4.0/20230318-SilenusNS-44k/pruned/P_G_12800.pth +3 -0
  7. so-vits-svc-4.0/20230318-SilenusNS-44k/pruned/config.json +93 -0
  8. so-vits-svc-4.0/20230318-SilenusNS-44k/train.log +498 -0
  9. so-vits-svc-4.0/20230324-Juewa-44k/best/D_12800.pth +3 -0
  10. so-vits-svc-4.0/20230324-Juewa-44k/best/G_12800.pth +3 -0
  11. so-vits-svc-4.0/20230324-Juewa-44k/best/config.json +93 -0
  12. so-vits-svc-4.0/20230324-Juewa-44k/config.json +93 -0
  13. so-vits-svc-4.0/20230324-Juewa-44k/kmeans_10000.pt +3 -0
  14. so-vits-svc-4.0/20230324-Juewa-44k/pruned/P_G_12800.pth +3 -0
  15. so-vits-svc-4.0/20230324-Juewa-44k/pruned/config.json +93 -0
  16. so-vits-svc-4.0/20230324-Juewa-44k/train.log +431 -0
  17. so-vits-svc-4.0/20230324-JuewaNS-44k/best/D_13600.pth +3 -0
  18. so-vits-svc-4.0/20230324-JuewaNS-44k/best/G_13600.pth +3 -0
  19. so-vits-svc-4.0/20230324-JuewaNS-44k/best/config.json +93 -0
  20. so-vits-svc-4.0/20230324-JuewaNS-44k/config.json +93 -0
  21. so-vits-svc-4.0/20230324-JuewaNS-44k/kmeans_10000.pt +3 -0
  22. so-vits-svc-4.0/20230324-JuewaNS-44k/pruned/P_G_13600.pth +3 -0
  23. so-vits-svc-4.0/20230324-JuewaNS-44k/pruned/config.json +93 -0
  24. so-vits-svc-4.0/20230324-JuewaNS-44k/train.log +772 -0
so-vits-svc-4.0/20230318-SilenusNS-44k/best/D_12800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d833c37330bba865519e043b3c8f652914dcc063e31f3e1df324dc95f7b1cbb
3
+ size 561098185
so-vits-svc-4.0/20230318-SilenusNS-44k/best/G_12800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12fb4e79b127698c14e337ea09c26b92c1447e1d963c6343220ab83052989cd6
3
+ size 542789405
so-vits-svc-4.0/20230318-SilenusNS-44k/best/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "SilenusNS": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230318-SilenusNS-44k/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "SilenusNS": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230318-SilenusNS-44k/kmeans_10000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32950ebf071b62f9bf6b6240c2c4a18f22535fa51430703104749b568da1b1ec
3
+ size 15446393
so-vits-svc-4.0/20230318-SilenusNS-44k/pruned/P_G_12800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:927581e85ab6d5485819b44b3dcb3c616ed104b15660ef91699c2458a9863116
3
+ size 180885829
so-vits-svc-4.0/20230318-SilenusNS-44k/pruned/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "SilenusNS": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230318-SilenusNS-44k/train.log ADDED
@@ -0,0 +1,498 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-03-18 15:25:05,061 44k INFO {'train': {'log_interval': 200, 'eval_interval': 800, 'seed': 1234, 'epochs': 10000, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 6, 'fp16_run': False, 'lr_decay': 0.999875, 'segment_size': 10240, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0, 'use_sr': True, 'max_speclen': 512, 'port': '8001', 'keep_ckpts': 99}, 'data': {'training_files': 'filelists/train.txt', 'validation_files': 'filelists/val.txt', 'max_wav_value': 32768.0, 'sampling_rate': 44100, 'filter_length': 2048, 'hop_length': 512, 'win_length': 2048, 'n_mel_channels': 80, 'mel_fmin': 0.0, 'mel_fmax': 22050}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.1, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [8, 8, 2, 2, 2], 'upsample_initial_channel': 512, 'upsample_kernel_sizes': [16, 16, 4, 4, 4], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 256, 'ssl_dim': 256, 'n_speakers': 200}, 'spk': {'SilenusNS': 0}, 'model_dir': './logs/44k'}
2
+ 2023-03-18 15:25:05,062 44k WARNING /root/so-vits-svc-4.0 is not a git repository, therefore hash value comparison will be ignored.
3
+ 2023-03-18 15:25:08,101 44k INFO emb_g.weight is not in the checkpoint
4
+ 2023-03-18 15:25:08,170 44k INFO Loaded checkpoint './logs/44k/G_0.pth' (iteration 0)
5
+ 2023-03-18 15:25:08,321 44k INFO Loaded checkpoint './logs/44k/D_0.pth' (iteration 0)
6
+ 2023-03-18 15:25:15,937 44k INFO Train Epoch: 1 [0%]
7
+ 2023-03-18 15:25:15,938 44k INFO Losses: [2.8153839111328125, 2.3640732765197754, 10.997515678405762, 34.97066116333008, 3.6714816093444824], step: 0, lr: 0.0001
8
+ 2023-03-18 15:25:20,732 44k INFO Saving model and optimizer state at iteration 1 to ./logs/44k/G_0.pth
9
+ 2023-03-18 15:25:21,865 44k INFO Saving model and optimizer state at iteration 1 to ./logs/44k/D_0.pth
10
+ 2023-03-18 15:25:42,275 44k INFO ====> Epoch: 1, cost 37.21 s
11
+ 2023-03-18 15:26:01,596 44k INFO ====> Epoch: 2, cost 19.32 s
12
+ 2023-03-18 15:26:20,960 44k INFO ====> Epoch: 3, cost 19.36 s
13
+ 2023-03-18 15:26:40,421 44k INFO ====> Epoch: 4, cost 19.46 s
14
+ 2023-03-18 15:26:59,757 44k INFO ====> Epoch: 5, cost 19.34 s
15
+ 2023-03-18 15:27:05,274 44k INFO Train Epoch: 6 [13%]
16
+ 2023-03-18 15:27:05,275 44k INFO Losses: [2.455998420715332, 2.1098105907440186, 8.791666030883789, 22.87582778930664, 1.493965744972229], step: 200, lr: 9.993751562304699e-05
17
+ 2023-03-18 15:27:20,129 44k INFO ====> Epoch: 6, cost 20.37 s
18
+ 2023-03-18 15:27:39,462 44k INFO ====> Epoch: 7, cost 19.33 s
19
+ 2023-03-18 15:27:58,896 44k INFO ====> Epoch: 8, cost 19.43 s
20
+ 2023-03-18 15:28:18,498 44k INFO ====> Epoch: 9, cost 19.60 s
21
+ 2023-03-18 15:28:38,294 44k INFO ====> Epoch: 10, cost 19.80 s
22
+ 2023-03-18 15:28:45,816 44k INFO Train Epoch: 11 [26%]
23
+ 2023-03-18 15:28:45,818 44k INFO Losses: [2.5894227027893066, 2.678623676300049, 10.583251953125, 23.435916900634766, 1.342075228691101], step: 400, lr: 9.987507028906759e-05
24
+ 2023-03-18 15:28:58,332 44k INFO ====> Epoch: 11, cost 20.04 s
25
+ 2023-03-18 15:29:18,589 44k INFO ====> Epoch: 12, cost 20.26 s
26
+ 2023-03-18 15:29:39,694 44k INFO ====> Epoch: 13, cost 21.11 s
27
+ 2023-03-18 15:29:59,110 44k INFO ====> Epoch: 14, cost 19.42 s
28
+ 2023-03-18 15:30:18,828 44k INFO ====> Epoch: 15, cost 19.72 s
29
+ 2023-03-18 15:30:28,571 44k INFO Train Epoch: 16 [38%]
30
+ 2023-03-18 15:30:28,572 44k INFO Losses: [2.3110649585723877, 2.5217154026031494, 8.599373817443848, 20.32848358154297, 1.1233292818069458], step: 600, lr: 9.981266397366609e-05
31
+ 2023-03-18 15:30:39,261 44k INFO ====> Epoch: 16, cost 20.43 s
32
+ 2023-03-18 15:30:58,613 44k INFO ====> Epoch: 17, cost 19.35 s
33
+ 2023-03-18 15:31:18,017 44k INFO ====> Epoch: 18, cost 19.40 s
34
+ 2023-03-18 15:31:37,554 44k INFO ====> Epoch: 19, cost 19.54 s
35
+ 2023-03-18 15:31:57,299 44k INFO ====> Epoch: 20, cost 19.74 s
36
+ 2023-03-18 15:32:09,588 44k INFO Train Epoch: 21 [51%]
37
+ 2023-03-18 15:32:09,589 44k INFO Losses: [2.4298205375671387, 2.607797145843506, 11.093040466308594, 22.630563735961914, 1.1880390644073486], step: 800, lr: 9.975029665246193e-05
38
+ 2023-03-18 15:32:13,823 44k INFO Saving model and optimizer state at iteration 21 to ./logs/44k/G_800.pth
39
+ 2023-03-18 15:32:14,917 44k INFO Saving model and optimizer state at iteration 21 to ./logs/44k/D_800.pth
40
+ 2023-03-18 15:32:23,333 44k INFO ====> Epoch: 21, cost 26.03 s
41
+ 2023-03-18 15:32:43,083 44k INFO ====> Epoch: 22, cost 19.75 s
42
+ 2023-03-18 15:33:03,071 44k INFO ====> Epoch: 23, cost 19.99 s
43
+ 2023-03-18 15:33:22,903 44k INFO ====> Epoch: 24, cost 19.83 s
44
+ 2023-03-18 15:33:42,382 44k INFO ====> Epoch: 25, cost 19.48 s
45
+ 2023-03-18 15:33:56,327 44k INFO Train Epoch: 26 [64%]
46
+ 2023-03-18 15:33:56,328 44k INFO Losses: [2.0924313068389893, 2.961833953857422, 7.720591068267822, 18.682811737060547, 0.679410457611084], step: 1000, lr: 9.968796830108985e-05
47
+ 2023-03-18 15:34:02,813 44k INFO ====> Epoch: 26, cost 20.43 s
48
+ 2023-03-18 15:34:22,323 44k INFO ====> Epoch: 27, cost 19.51 s
49
+ 2023-03-18 15:34:41,799 44k INFO ====> Epoch: 28, cost 19.48 s
50
+ 2023-03-18 15:35:01,307 44k INFO ====> Epoch: 29, cost 19.51 s
51
+ 2023-03-18 15:35:20,945 44k INFO ====> Epoch: 30, cost 19.64 s
52
+ 2023-03-18 15:35:36,798 44k INFO Train Epoch: 31 [77%]
53
+ 2023-03-18 15:35:36,800 44k INFO Losses: [2.3155126571655273, 2.206496477127075, 11.064719200134277, 19.939838409423828, 0.7182945013046265], step: 1200, lr: 9.962567889519979e-05
54
+ 2023-03-18 15:35:41,510 44k INFO ====> Epoch: 31, cost 20.57 s
55
+ 2023-03-18 15:36:01,618 44k INFO ====> Epoch: 32, cost 20.11 s
56
+ 2023-03-18 15:36:21,079 44k INFO ====> Epoch: 33, cost 19.46 s
57
+ 2023-03-18 15:36:40,302 44k INFO ====> Epoch: 34, cost 19.22 s
58
+ 2023-03-18 15:36:59,771 44k INFO ====> Epoch: 35, cost 19.47 s
59
+ 2023-03-18 15:37:17,648 44k INFO Train Epoch: 36 [90%]
60
+ 2023-03-18 15:37:17,650 44k INFO Losses: [1.8354809284210205, 2.9983744621276855, 10.637933731079102, 18.277450561523438, 0.9634370803833008], step: 1400, lr: 9.956342841045691e-05
61
+ 2023-03-18 15:37:19,971 44k INFO ====> Epoch: 36, cost 20.20 s
62
+ 2023-03-18 15:37:39,555 44k INFO ====> Epoch: 37, cost 19.58 s
63
+ 2023-03-18 15:37:58,866 44k INFO ====> Epoch: 38, cost 19.31 s
64
+ 2023-03-18 15:38:18,557 44k INFO ====> Epoch: 39, cost 19.69 s
65
+ 2023-03-18 15:38:38,195 44k INFO ====> Epoch: 40, cost 19.64 s
66
+ 2023-03-18 15:38:57,921 44k INFO ====> Epoch: 41, cost 19.73 s
67
+ 2023-03-18 15:39:01,847 44k INFO Train Epoch: 42 [3%]
68
+ 2023-03-18 15:39:01,848 44k INFO Losses: [2.7251152992248535, 2.1925649642944336, 7.784726142883301, 18.160188674926758, 0.8011401295661926], step: 1600, lr: 9.948877917043875e-05
69
+ 2023-03-18 15:39:06,045 44k INFO Saving model and optimizer state at iteration 42 to ./logs/44k/G_1600.pth
70
+ 2023-03-18 15:39:07,090 44k INFO Saving model and optimizer state at iteration 42 to ./logs/44k/D_1600.pth
71
+ 2023-03-18 15:39:23,376 44k INFO ====> Epoch: 42, cost 25.45 s
72
+ 2023-03-18 15:39:43,065 44k INFO ====> Epoch: 43, cost 19.69 s
73
+ 2023-03-18 15:40:02,810 44k INFO ====> Epoch: 44, cost 19.75 s
74
+ 2023-03-18 15:40:22,260 44k INFO ====> Epoch: 45, cost 19.45 s
75
+ 2023-03-18 15:40:42,514 44k INFO ====> Epoch: 46, cost 20.25 s
76
+ 2023-03-18 15:40:48,648 44k INFO Train Epoch: 47 [15%]
77
+ 2023-03-18 15:40:48,649 44k INFO Losses: [2.5461223125457764, 2.1349644660949707, 6.2196221351623535, 15.844928741455078, 1.0983959436416626], step: 1800, lr: 9.942661422663591e-05
78
+ 2023-03-18 15:41:02,946 44k INFO ====> Epoch: 47, cost 20.43 s
79
+ 2023-03-18 15:41:23,064 44k INFO ====> Epoch: 48, cost 20.12 s
80
+ 2023-03-18 15:41:43,578 44k INFO ====> Epoch: 49, cost 20.51 s
81
+ 2023-03-18 15:42:03,255 44k INFO ====> Epoch: 50, cost 19.68 s
82
+ 2023-03-18 15:42:23,980 44k INFO ====> Epoch: 51, cost 20.72 s
83
+ 2023-03-18 15:42:32,232 44k INFO Train Epoch: 52 [28%]
84
+ 2023-03-18 15:42:32,234 44k INFO Losses: [2.4787557125091553, 2.4002110958099365, 7.5291852951049805, 17.007627487182617, 0.9123826026916504], step: 2000, lr: 9.936448812621091e-05
85
+ 2023-03-18 15:42:44,654 44k INFO ====> Epoch: 52, cost 20.67 s
86
+ 2023-03-18 15:43:04,778 44k INFO ====> Epoch: 53, cost 20.12 s
87
+ 2023-03-18 15:43:24,491 44k INFO ====> Epoch: 54, cost 19.71 s
88
+ 2023-03-18 15:43:44,206 44k INFO ====> Epoch: 55, cost 19.72 s
89
+ 2023-03-18 15:44:03,873 44k INFO ====> Epoch: 56, cost 19.67 s
90
+ 2023-03-18 15:44:14,036 44k INFO Train Epoch: 57 [41%]
91
+ 2023-03-18 15:44:14,038 44k INFO Losses: [2.3435280323028564, 2.4136173725128174, 8.905498504638672, 21.293319702148438, 0.8389438390731812], step: 2200, lr: 9.930240084489267e-05
92
+ 2023-03-18 15:44:24,123 44k INFO ====> Epoch: 57, cost 20.25 s
93
+ 2023-03-18 15:44:43,994 44k INFO ====> Epoch: 58, cost 19.87 s
94
+ 2023-03-18 15:45:05,109 44k INFO ====> Epoch: 59, cost 21.11 s
95
+ 2023-03-18 15:45:24,973 44k INFO ====> Epoch: 60, cost 19.86 s
96
+ 2023-03-18 15:45:45,042 44k INFO ====> Epoch: 61, cost 20.07 s
97
+ 2023-03-18 15:45:57,802 44k INFO Train Epoch: 62 [54%]
98
+ 2023-03-18 15:45:57,803 44k INFO Losses: [2.546645164489746, 2.279296636581421, 8.504755973815918, 22.139497756958008, 0.8844197392463684], step: 2400, lr: 9.924035235842533e-05
99
+ 2023-03-18 15:46:02,298 44k INFO Saving model and optimizer state at iteration 62 to ./logs/44k/G_2400.pth
100
+ 2023-03-18 15:46:03,340 44k INFO Saving model and optimizer state at iteration 62 to ./logs/44k/D_2400.pth
101
+ 2023-03-18 15:46:12,023 44k INFO ====> Epoch: 62, cost 26.98 s
102
+ 2023-03-18 15:46:31,951 44k INFO ====> Epoch: 63, cost 19.93 s
103
+ 2023-03-18 15:46:51,635 44k INFO ====> Epoch: 64, cost 19.68 s
104
+ 2023-03-18 15:47:11,463 44k INFO ====> Epoch: 65, cost 19.83 s
105
+ 2023-03-18 15:47:31,393 44k INFO ====> Epoch: 66, cost 19.93 s
106
+ 2023-03-18 15:47:45,850 44k INFO Train Epoch: 67 [67%]
107
+ 2023-03-18 15:47:45,852 44k INFO Losses: [2.351210594177246, 2.3612356185913086, 7.31257963180542, 14.846061706542969, 1.1602634191513062], step: 2600, lr: 9.917834264256819e-05
108
+ 2023-03-18 15:47:52,492 44k INFO ====> Epoch: 67, cost 21.10 s
109
+ 2023-03-18 15:48:12,681 44k INFO ====> Epoch: 68, cost 20.19 s
110
+ 2023-03-18 15:48:32,852 44k INFO ====> Epoch: 69, cost 20.17 s
111
+ 2023-03-18 15:48:53,823 44k INFO ====> Epoch: 70, cost 20.97 s
112
+ 2023-03-18 15:49:13,763 44k INFO ====> Epoch: 71, cost 19.94 s
113
+ 2023-03-18 15:49:30,248 44k INFO Train Epoch: 72 [79%]
114
+ 2023-03-18 15:49:30,250 44k INFO Losses: [2.2914111614227295, 2.856121778488159, 8.181981086730957, 13.763769149780273, 0.6541903614997864], step: 2800, lr: 9.911637167309565e-05
115
+ 2023-03-18 15:49:34,249 44k INFO ====> Epoch: 72, cost 20.49 s
116
+ 2023-03-18 15:49:54,134 44k INFO ====> Epoch: 73, cost 19.88 s
117
+ 2023-03-18 15:50:15,695 44k INFO ====> Epoch: 74, cost 21.56 s
118
+ 2023-03-18 15:50:35,823 44k INFO ====> Epoch: 75, cost 20.13 s
119
+ 2023-03-18 15:50:56,045 44k INFO ====> Epoch: 76, cost 20.22 s
120
+ 2023-03-18 15:51:14,689 44k INFO Train Epoch: 77 [92%]
121
+ 2023-03-18 15:51:14,691 44k INFO Losses: [2.3739912509918213, 2.6581947803497314, 9.7283935546875, 19.127307891845703, 0.7439061999320984], step: 3000, lr: 9.905443942579728e-05
122
+ 2023-03-18 15:51:16,643 44k INFO ====> Epoch: 77, cost 20.60 s
123
+ 2023-03-18 15:51:36,563 44k INFO ====> Epoch: 78, cost 19.92 s
124
+ 2023-03-18 15:51:56,622 44k INFO ====> Epoch: 79, cost 20.06 s
125
+ 2023-03-18 15:52:16,476 44k INFO ====> Epoch: 80, cost 19.85 s
126
+ 2023-03-18 15:52:36,368 44k INFO ====> Epoch: 81, cost 19.89 s
127
+ 2023-03-18 15:52:57,128 44k INFO ====> Epoch: 82, cost 20.76 s
128
+ 2023-03-18 15:53:01,575 44k INFO Train Epoch: 83 [5%]
129
+ 2023-03-18 15:53:01,576 44k INFO Losses: [2.575831413269043, 2.0763652324676514, 10.35012435913086, 21.57923698425293, 0.7908153533935547], step: 3200, lr: 9.89801718082432e-05
130
+ 2023-03-18 15:53:06,109 44k INFO Saving model and optimizer state at iteration 83 to ./logs/44k/G_3200.pth
131
+ 2023-03-18 15:53:07,151 44k INFO Saving model and optimizer state at iteration 83 to ./logs/44k/D_3200.pth
132
+ 2023-03-18 15:53:23,356 44k INFO ====> Epoch: 83, cost 26.23 s
133
+ 2023-03-18 15:53:43,274 44k INFO ====> Epoch: 84, cost 19.92 s
134
+ 2023-03-18 15:54:04,946 44k INFO ====> Epoch: 85, cost 21.67 s
135
+ 2023-03-18 15:54:24,908 44k INFO ====> Epoch: 86, cost 19.96 s
136
+ 2023-03-18 15:54:46,074 44k INFO ====> Epoch: 87, cost 21.17 s
137
+ 2023-03-18 15:54:52,624 44k INFO Train Epoch: 88 [18%]
138
+ 2023-03-18 15:54:52,625 44k INFO Losses: [2.676368236541748, 2.128420829772949, 6.567540645599365, 14.580389022827148, 0.793415904045105], step: 3400, lr: 9.891832466458178e-05
139
+ 2023-03-18 15:55:06,763 44k INFO ====> Epoch: 88, cost 20.69 s
140
+ 2023-03-18 15:55:27,296 44k INFO ====> Epoch: 89, cost 20.53 s
141
+ 2023-03-18 15:55:47,223 44k INFO ====> Epoch: 90, cost 19.93 s
142
+ 2023-03-18 15:56:07,621 44k INFO ====> Epoch: 91, cost 20.40 s
143
+ 2023-03-18 15:56:27,682 44k INFO ====> Epoch: 92, cost 20.06 s
144
+ 2023-03-18 15:56:36,318 44k INFO Train Epoch: 93 [31%]
145
+ 2023-03-18 15:56:36,320 44k INFO Losses: [2.152287006378174, 2.4968721866607666, 12.756311416625977, 22.887723922729492, 1.057271122932434], step: 3600, lr: 9.885651616572276e-05
146
+ 2023-03-18 15:56:48,140 44k INFO ====> Epoch: 93, cost 20.46 s
147
+ 2023-03-18 15:57:07,967 44k INFO ====> Epoch: 94, cost 19.83 s
148
+ 2023-03-18 15:57:28,520 44k INFO ====> Epoch: 95, cost 20.55 s
149
+ 2023-03-18 15:57:48,340 44k INFO ====> Epoch: 96, cost 19.82 s
150
+ 2023-03-18 15:58:08,947 44k INFO ====> Epoch: 97, cost 20.61 s
151
+ 2023-03-18 15:58:19,669 44k INFO Train Epoch: 98 [44%]
152
+ 2023-03-18 15:58:19,670 44k INFO Losses: [2.427189350128174, 2.0381827354431152, 9.996284484863281, 19.298826217651367, 0.6840081810951233], step: 3800, lr: 9.879474628751914e-05
153
+ 2023-03-18 15:58:29,484 44k INFO ====> Epoch: 98, cost 20.54 s
154
+ 2023-03-18 15:58:49,521 44k INFO ====> Epoch: 99, cost 20.04 s
155
+ 2023-03-18 15:59:10,060 44k INFO ====> Epoch: 100, cost 20.54 s
156
+ 2023-03-18 15:59:30,092 44k INFO ====> Epoch: 101, cost 20.03 s
157
+ 2023-03-18 15:59:50,195 44k INFO ====> Epoch: 102, cost 20.10 s
158
+ 2023-03-18 16:00:03,365 44k INFO Train Epoch: 103 [56%]
159
+ 2023-03-18 16:00:03,367 44k INFO Losses: [2.5533149242401123, 2.544842004776001, 10.731807708740234, 20.337570190429688, 0.4659505784511566], step: 4000, lr: 9.873301500583906e-05
160
+ 2023-03-18 16:00:08,064 44k INFO Saving model and optimizer state at iteration 103 to ./logs/44k/G_4000.pth
161
+ 2023-03-18 16:00:09,144 44k INFO Saving model and optimizer state at iteration 103 to ./logs/44k/D_4000.pth
162
+ 2023-03-18 16:00:16,893 44k INFO ====> Epoch: 103, cost 26.70 s
163
+ 2023-03-18 16:00:36,789 44k INFO ====> Epoch: 104, cost 19.90 s
164
+ 2023-03-18 16:00:58,587 44k INFO ====> Epoch: 105, cost 21.80 s
165
+ 2023-03-18 16:01:18,557 44k INFO ====> Epoch: 106, cost 19.97 s
166
+ 2023-03-18 16:01:38,667 44k INFO ====> Epoch: 107, cost 20.11 s
167
+ 2023-03-18 16:01:53,610 44k INFO Train Epoch: 108 [69%]
168
+ 2023-03-18 16:01:53,612 44k INFO Losses: [2.335444211959839, 2.434490203857422, 8.708498001098633, 19.969039916992188, 0.9741002321243286], step: 4200, lr: 9.867132229656573e-05
169
+ 2023-03-18 16:01:59,163 44k INFO ====> Epoch: 108, cost 20.50 s
170
+ 2023-03-18 16:02:19,113 44k INFO ====> Epoch: 109, cost 19.95 s
171
+ 2023-03-18 16:02:39,324 44k INFO ====> Epoch: 110, cost 20.21 s
172
+ 2023-03-18 16:02:59,329 44k INFO ====> Epoch: 111, cost 20.00 s
173
+ 2023-03-18 16:03:20,073 44k INFO ====> Epoch: 112, cost 20.74 s
174
+ 2023-03-18 16:03:36,919 44k INFO Train Epoch: 113 [82%]
175
+ 2023-03-18 16:03:36,921 44k INFO Losses: [2.5057225227355957, 2.2557621002197266, 9.105669021606445, 17.700326919555664, 0.6261863112449646], step: 4400, lr: 9.86096681355974e-05
176
+ 2023-03-18 16:03:40,487 44k INFO ====> Epoch: 113, cost 20.41 s
177
+ 2023-03-18 16:04:00,296 44k INFO ====> Epoch: 114, cost 19.81 s
178
+ 2023-03-18 16:04:20,189 44k INFO ====> Epoch: 115, cost 19.89 s
179
+ 2023-03-18 16:04:39,901 44k INFO ====> Epoch: 116, cost 19.71 s
180
+ 2023-03-18 16:05:00,262 44k INFO ====> Epoch: 117, cost 20.36 s
181
+ 2023-03-18 16:05:19,262 44k INFO Train Epoch: 118 [95%]
182
+ 2023-03-18 16:05:19,263 44k INFO Losses: [2.282578229904175, 2.4207687377929688, 9.76828384399414, 21.194089889526367, 0.9633945822715759], step: 4600, lr: 9.854805249884741e-05
183
+ 2023-03-18 16:05:20,950 44k INFO ====> Epoch: 118, cost 20.69 s
184
+ 2023-03-18 16:05:41,036 44k INFO ====> Epoch: 119, cost 20.09 s
185
+ 2023-03-18 16:06:01,142 44k INFO ====> Epoch: 120, cost 20.11 s
186
+ 2023-03-18 16:06:21,123 44k INFO ====> Epoch: 121, cost 19.98 s
187
+ 2023-03-18 16:06:40,925 44k INFO ====> Epoch: 122, cost 19.80 s
188
+ 2023-03-18 16:07:00,860 44k INFO ====> Epoch: 123, cost 19.93 s
189
+ 2023-03-18 16:07:05,476 44k INFO Train Epoch: 124 [8%]
190
+ 2023-03-18 16:07:05,477 44k INFO Losses: [2.2480196952819824, 2.5063650608062744, 10.50140380859375, 22.77250099182129, 0.9855556488037109], step: 4800, lr: 9.847416455282387e-05
191
+ 2023-03-18 16:07:09,857 44k INFO Saving model and optimizer state at iteration 124 to ./logs/44k/G_4800.pth
192
+ 2023-03-18 16:07:11,047 44k INFO Saving model and optimizer state at iteration 124 to ./logs/44k/D_4800.pth
193
+ 2023-03-18 16:07:26,839 44k INFO ====> Epoch: 124, cost 25.98 s
194
+ 2023-03-18 16:07:48,107 44k INFO ====> Epoch: 125, cost 21.27 s
195
+ 2023-03-18 16:08:07,938 44k INFO ====> Epoch: 126, cost 19.83 s
196
+ 2023-03-18 16:08:27,940 44k INFO ====> Epoch: 127, cost 20.00 s
197
+ 2023-03-18 16:08:47,904 44k INFO ====> Epoch: 128, cost 19.96 s
198
+ 2023-03-18 16:08:54,670 44k INFO Train Epoch: 129 [21%]
199
+ 2023-03-18 16:08:54,672 44k INFO Losses: [2.0410420894622803, 3.1921937465667725, 13.629049301147461, 22.29718017578125, 0.7504812479019165], step: 5000, lr: 9.841263358464336e-05
200
+ 2023-03-18 16:09:08,185 44k INFO ====> Epoch: 129, cost 20.28 s
201
+ 2023-03-18 16:09:28,708 44k INFO ====> Epoch: 130, cost 20.52 s
202
+ 2023-03-18 16:09:49,065 44k INFO ====> Epoch: 131, cost 20.36 s
203
+ 2023-03-18 16:10:08,965 44k INFO ====> Epoch: 132, cost 19.90 s
204
+ 2023-03-18 16:10:28,600 44k INFO ====> Epoch: 133, cost 19.64 s
205
+ 2023-03-18 16:10:37,520 44k INFO Train Epoch: 134 [33%]
206
+ 2023-03-18 16:10:37,522 44k INFO Losses: [2.0373544692993164, 2.7676761150360107, 9.952956199645996, 19.683595657348633, 0.7579522728919983], step: 5200, lr: 9.835114106370493e-05
207
+ 2023-03-18 16:10:49,091 44k INFO ====> Epoch: 134, cost 20.49 s
208
+ 2023-03-18 16:11:09,151 44k INFO ====> Epoch: 135, cost 20.06 s
209
+ 2023-03-18 16:11:29,063 44k INFO ====> Epoch: 136, cost 19.91 s
210
+ 2023-03-18 16:11:48,924 44k INFO ====> Epoch: 137, cost 19.86 s
211
+ 2023-03-18 16:12:08,871 44k INFO ====> Epoch: 138, cost 19.95 s
212
+ 2023-03-18 16:12:19,890 44k INFO Train Epoch: 139 [46%]
213
+ 2023-03-18 16:12:19,891 44k INFO Losses: [2.2704925537109375, 2.290191650390625, 11.153946876525879, 20.080801010131836, 0.7212128043174744], step: 5400, lr: 9.828968696598508e-05
214
+ 2023-03-18 16:12:29,307 44k INFO ====> Epoch: 139, cost 20.44 s
215
+ 2023-03-18 16:12:49,225 44k INFO ====> Epoch: 140, cost 19.92 s
216
+ 2023-03-18 16:13:09,130 44k INFO ====> Epoch: 141, cost 19.90 s
217
+ 2023-03-18 16:13:29,144 44k INFO ====> Epoch: 142, cost 20.01 s
218
+ 2023-03-18 16:13:49,016 44k INFO ====> Epoch: 143, cost 19.87 s
219
+ 2023-03-18 16:14:02,148 44k INFO Train Epoch: 144 [59%]
220
+ 2023-03-18 16:14:02,149 44k INFO Losses: [1.7235863208770752, 3.459230899810791, 11.362791061401367, 15.820011138916016, 0.6718958020210266], step: 5600, lr: 9.822827126747529e-05
221
+ 2023-03-18 16:14:06,691 44k INFO Saving model and optimizer state at iteration 144 to ./logs/44k/G_5600.pth
222
+ 2023-03-18 16:14:07,736 44k INFO Saving model and optimizer state at iteration 144 to ./logs/44k/D_5600.pth
223
+ 2023-03-18 16:14:15,101 44k INFO ====> Epoch: 144, cost 26.08 s
224
+ 2023-03-18 16:14:35,033 44k INFO ====> Epoch: 145, cost 19.93 s
225
+ 2023-03-18 16:14:55,256 44k INFO ====> Epoch: 146, cost 20.22 s
226
+ 2023-03-18 16:15:16,052 44k INFO ====> Epoch: 147, cost 20.80 s
227
+ 2023-03-18 16:15:35,922 44k INFO ====> Epoch: 148, cost 19.87 s
228
+ 2023-03-18 16:15:51,015 44k INFO Train Epoch: 149 [72%]
229
+ 2023-03-18 16:15:51,017 44k INFO Losses: [1.8325955867767334, 2.861543655395508, 11.257867813110352, 22.430797576904297, 0.9396995306015015], step: 5800, lr: 9.816689394418209e-05
230
+ 2023-03-18 16:15:56,357 44k INFO ====> Epoch: 149, cost 20.44 s
231
+ 2023-03-18 16:16:16,123 44k INFO ====> Epoch: 150, cost 19.77 s
232
+ 2023-03-18 16:16:36,108 44k INFO ====> Epoch: 151, cost 19.99 s
233
+ 2023-03-18 16:16:56,012 44k INFO ====> Epoch: 152, cost 19.90 s
234
+ 2023-03-18 16:17:15,691 44k INFO ====> Epoch: 153, cost 19.68 s
235
+ 2023-03-18 16:17:32,924 44k INFO Train Epoch: 154 [85%]
236
+ 2023-03-18 16:17:32,925 44k INFO Losses: [2.4294025897979736, 2.6236627101898193, 8.439064979553223, 15.59575080871582, 0.777889609336853], step: 6000, lr: 9.810555497212693e-05
237
+ 2023-03-18 16:17:36,044 44k INFO ====> Epoch: 154, cost 20.35 s
238
+ 2023-03-18 16:17:58,583 44k INFO ====> Epoch: 155, cost 22.54 s
239
+ 2023-03-18 16:18:18,616 44k INFO ====> Epoch: 156, cost 20.03 s
240
+ 2023-03-18 16:18:38,531 44k INFO ====> Epoch: 157, cost 19.92 s
241
+ 2023-03-18 16:18:58,590 44k INFO ====> Epoch: 158, cost 20.06 s
242
+ 2023-03-18 16:19:18,098 44k INFO Train Epoch: 159 [97%]
243
+ 2023-03-18 16:19:18,100 44k INFO Losses: [2.6305150985717773, 2.01100754737854, 5.256586074829102, 12.10132884979248, 0.5776822566986084], step: 6200, lr: 9.804425432734629e-05
244
+ 2023-03-18 16:19:19,423 44k INFO ====> Epoch: 159, cost 20.83 s
245
+ 2023-03-18 16:19:39,921 44k INFO ====> Epoch: 160, cost 20.50 s
246
+ 2023-03-18 16:19:59,981 44k INFO ====> Epoch: 161, cost 20.06 s
247
+ 2023-03-18 16:20:20,378 44k INFO ====> Epoch: 162, cost 20.40 s
248
+ 2023-03-18 16:20:40,436 44k INFO ====> Epoch: 163, cost 20.06 s
249
+ 2023-03-18 16:21:00,451 44k INFO ====> Epoch: 164, cost 20.01 s
250
+ 2023-03-18 16:21:05,832 44k INFO Train Epoch: 165 [10%]
251
+ 2023-03-18 16:21:05,834 44k INFO Losses: [2.3310093879699707, 2.7704384326934814, 6.842113018035889, 14.216880798339844, 0.8058108687400818], step: 6400, lr: 9.797074411189339e-05
252
+ 2023-03-18 16:21:10,339 44k INFO Saving model and optimizer state at iteration 165 to ./logs/44k/G_6400.pth
253
+ 2023-03-18 16:21:11,548 44k INFO Saving model and optimizer state at iteration 165 to ./logs/44k/D_6400.pth
254
+ 2023-03-18 16:21:26,986 44k INFO ====> Epoch: 165, cost 26.54 s
255
+ 2023-03-18 16:21:47,179 44k INFO ====> Epoch: 166, cost 20.19 s
256
+ 2023-03-18 16:22:06,919 44k INFO ====> Epoch: 167, cost 19.74 s
257
+ 2023-03-18 16:22:26,894 44k INFO ====> Epoch: 168, cost 19.97 s
258
+ 2023-03-18 16:22:46,930 44k INFO ====> Epoch: 169, cost 20.04 s
259
+ 2023-03-18 16:22:54,400 44k INFO Train Epoch: 170 [23%]
260
+ 2023-03-18 16:22:54,401 44k INFO Losses: [2.2254819869995117, 2.3765923976898193, 11.166699409484863, 22.277254104614258, 0.7468198537826538], step: 6600, lr: 9.790952770283884e-05
261
+ 2023-03-18 16:23:07,540 44k INFO ====> Epoch: 170, cost 20.61 s
262
+ 2023-03-18 16:23:27,507 44k INFO ====> Epoch: 171, cost 19.97 s
263
+ 2023-03-18 16:23:47,511 44k INFO ====> Epoch: 172, cost 20.00 s
264
+ 2023-03-18 16:24:07,612 44k INFO ====> Epoch: 173, cost 20.10 s
265
+ 2023-03-18 16:24:27,302 44k INFO ====> Epoch: 174, cost 19.69 s
266
+ 2023-03-18 16:24:36,709 44k INFO Train Epoch: 175 [36%]
267
+ 2023-03-18 16:24:36,711 44k INFO Losses: [2.520063877105713, 2.7583086490631104, 10.505017280578613, 20.471426010131836, 1.0650105476379395], step: 6800, lr: 9.784834954447608e-05
268
+ 2023-03-18 16:24:47,651 44k INFO ====> Epoch: 175, cost 20.35 s
269
+ 2023-03-18 16:25:07,509 44k INFO ====> Epoch: 176, cost 19.86 s
270
+ 2023-03-18 16:25:27,155 44k INFO ====> Epoch: 177, cost 19.65 s
271
+ 2023-03-18 16:25:46,776 44k INFO ====> Epoch: 178, cost 19.62 s
272
+ 2023-03-18 16:26:06,771 44k INFO ====> Epoch: 179, cost 19.99 s
273
+ 2023-03-18 16:26:18,260 44k INFO Train Epoch: 180 [49%]
274
+ 2023-03-18 16:26:18,262 44k INFO Losses: [2.593710422515869, 2.4376070499420166, 9.219118118286133, 20.284957885742188, 0.6395339965820312], step: 7000, lr: 9.778720961290439e-05
275
+ 2023-03-18 16:26:27,317 44k INFO ====> Epoch: 180, cost 20.55 s
276
+ 2023-03-18 16:26:47,792 44k INFO ====> Epoch: 181, cost 20.47 s
277
+ 2023-03-18 16:27:07,625 44k INFO ====> Epoch: 182, cost 19.83 s
278
+ 2023-03-18 16:27:28,099 44k INFO ====> Epoch: 183, cost 20.47 s
279
+ 2023-03-18 16:27:48,074 44k INFO ====> Epoch: 184, cost 19.98 s
280
+ 2023-03-18 16:28:01,863 44k INFO Train Epoch: 185 [62%]
281
+ 2023-03-18 16:28:01,865 44k INFO Losses: [2.254882574081421, 2.5508973598480225, 11.759598731994629, 22.320451736450195, 0.728751003742218], step: 7200, lr: 9.772610788423802e-05
282
+ 2023-03-18 16:28:06,387 44k INFO Saving model and optimizer state at iteration 185 to ./logs/44k/G_7200.pth
283
+ 2023-03-18 16:28:07,442 44k INFO Saving model and optimizer state at iteration 185 to ./logs/44k/D_7200.pth
284
+ 2023-03-18 16:28:14,594 44k INFO ====> Epoch: 185, cost 26.52 s
285
+ 2023-03-18 16:28:34,485 44k INFO ====> Epoch: 186, cost 19.89 s
286
+ 2023-03-18 16:28:54,373 44k INFO ====> Epoch: 187, cost 19.89 s
287
+ 2023-03-18 16:29:14,629 44k INFO ====> Epoch: 188, cost 20.26 s
288
+ 2023-03-18 16:29:34,530 44k INFO ====> Epoch: 189, cost 19.90 s
289
+ 2023-03-18 16:29:50,193 44k INFO Train Epoch: 190 [74%]
290
+ 2023-03-18 16:29:50,195 44k INFO Losses: [2.3000588417053223, 2.6743597984313965, 12.745715141296387, 21.8736629486084, 1.0124669075012207], step: 7400, lr: 9.766504433460612e-05
291
+ 2023-03-18 16:29:55,076 44k INFO ====> Epoch: 190, cost 20.55 s
292
+ 2023-03-18 16:30:14,888 44k INFO ====> Epoch: 191, cost 19.81 s
293
+ 2023-03-18 16:30:34,745 44k INFO ====> Epoch: 192, cost 19.86 s
294
+ 2023-03-18 16:30:54,694 44k INFO ====> Epoch: 193, cost 19.95 s
295
+ 2023-03-18 16:31:14,670 44k INFO ====> Epoch: 194, cost 19.98 s
296
+ 2023-03-18 16:31:33,818 44k INFO Train Epoch: 195 [87%]
297
+ 2023-03-18 16:31:33,820 44k INFO Losses: [2.2755260467529297, 1.8814972639083862, 11.437822341918945, 18.62908172607422, 0.9606003761291504], step: 7600, lr: 9.760401894015275e-05
298
+ 2023-03-18 16:31:36,953 44k INFO ====> Epoch: 195, cost 22.28 s
299
+ 2023-03-18 16:31:56,908 44k INFO ====> Epoch: 196, cost 19.95 s
300
+ 2023-03-18 16:32:16,687 44k INFO ====> Epoch: 197, cost 19.78 s
301
+ 2023-03-18 16:32:36,602 44k INFO ====> Epoch: 198, cost 19.92 s
302
+ 2023-03-18 16:32:56,834 44k INFO ====> Epoch: 199, cost 20.23 s
303
+ 2023-03-18 16:33:17,030 44k INFO ====> Epoch: 200, cost 20.20 s
304
+ 2023-03-18 16:33:20,686 44k INFO Train Epoch: 201 [0%]
305
+ 2023-03-18 16:33:20,688 44k INFO Losses: [2.6007676124572754, 2.2041707038879395, 11.113945960998535, 23.517358779907227, 0.8698729276657104], step: 7800, lr: 9.753083879807726e-05
306
+ 2023-03-18 16:33:37,631 44k INFO ====> Epoch: 201, cost 20.60 s
307
+ 2023-03-18 16:33:57,697 44k INFO ====> Epoch: 202, cost 20.07 s
308
+ 2023-03-18 16:34:17,632 44k INFO ====> Epoch: 203, cost 19.93 s
309
+ 2023-03-18 16:34:37,812 44k INFO ====> Epoch: 204, cost 20.18 s
310
+ 2023-03-18 16:34:57,797 44k INFO ====> Epoch: 205, cost 19.99 s
311
+ 2023-03-18 16:35:03,451 44k INFO Train Epoch: 206 [13%]
312
+ 2023-03-18 16:35:03,453 44k INFO Losses: [2.4772579669952393, 2.2229604721069336, 12.207804679870605, 21.374996185302734, 0.4583211839199066], step: 8000, lr: 9.746989726111722e-05
313
+ 2023-03-18 16:35:08,033 44k INFO Saving model and optimizer state at iteration 206 to ./logs/44k/G_8000.pth
314
+ 2023-03-18 16:35:09,056 44k INFO Saving model and optimizer state at iteration 206 to ./logs/44k/D_8000.pth
315
+ 2023-03-18 16:35:24,005 44k INFO ====> Epoch: 206, cost 26.21 s
316
+ 2023-03-18 16:35:44,479 44k INFO ====> Epoch: 207, cost 20.47 s
317
+ 2023-03-18 16:36:04,435 44k INFO ====> Epoch: 208, cost 19.96 s
318
+ 2023-03-18 16:36:24,516 44k INFO ====> Epoch: 209, cost 20.08 s
319
+ 2023-03-18 16:36:44,785 44k INFO ====> Epoch: 210, cost 20.27 s
320
+ 2023-03-18 16:36:52,596 44k INFO Train Epoch: 211 [26%]
321
+ 2023-03-18 16:36:52,598 44k INFO Losses: [2.1331419944763184, 2.492473602294922, 9.553248405456543, 17.68283462524414, 0.8661839365959167], step: 8200, lr: 9.740899380309685e-05
322
+ 2023-03-18 16:37:05,613 44k INFO ====> Epoch: 211, cost 20.83 s
323
+ 2023-03-18 16:37:25,466 44k INFO ====> Epoch: 212, cost 19.85 s
324
+ 2023-03-18 16:37:45,586 44k INFO ====> Epoch: 213, cost 20.12 s
325
+ 2023-03-18 16:38:06,088 44k INFO ====> Epoch: 214, cost 20.50 s
326
+ 2023-03-18 16:38:25,877 44k INFO ====> Epoch: 215, cost 19.79 s
327
+ 2023-03-18 16:38:35,618 44k INFO Train Epoch: 216 [38%]
328
+ 2023-03-18 16:38:35,620 44k INFO Losses: [2.4721179008483887, 2.524754285812378, 8.135478973388672, 17.019596099853516, 0.7745293974876404], step: 8400, lr: 9.734812840022278e-05
329
+ 2023-03-18 16:38:46,258 44k INFO ====> Epoch: 216, cost 20.38 s
330
+ 2023-03-18 16:39:06,167 44k INFO ====> Epoch: 217, cost 19.91 s
331
+ 2023-03-18 16:39:26,299 44k INFO ====> Epoch: 218, cost 20.13 s
332
+ 2023-03-18 16:39:46,288 44k INFO ====> Epoch: 219, cost 19.99 s
333
+ 2023-03-18 16:40:06,382 44k INFO ====> Epoch: 220, cost 20.09 s
334
+ 2023-03-18 16:40:18,584 44k INFO Train Epoch: 221 [51%]
335
+ 2023-03-18 16:40:18,586 44k INFO Losses: [2.237769842147827, 3.1392822265625, 14.79629135131836, 22.576732635498047, 0.5992900729179382], step: 8600, lr: 9.728730102871649e-05
336
+ 2023-03-18 16:40:27,279 44k INFO ====> Epoch: 221, cost 20.90 s
337
+ 2023-03-18 16:40:49,262 44k INFO ====> Epoch: 222, cost 21.98 s
338
+ 2023-03-18 16:41:09,403 44k INFO ====> Epoch: 223, cost 20.14 s
339
+ 2023-03-18 16:41:29,335 44k INFO ====> Epoch: 224, cost 19.93 s
340
+ 2023-03-18 16:41:49,368 44k INFO ====> Epoch: 225, cost 20.03 s
341
+ 2023-03-18 16:42:03,483 44k INFO Train Epoch: 226 [64%]
342
+ 2023-03-18 16:42:03,484 44k INFO Losses: [2.1953744888305664, 2.8054661750793457, 12.387676239013672, 20.940820693969727, 0.895363450050354], step: 8800, lr: 9.722651166481428e-05
343
+ 2023-03-18 16:42:08,135 44k INFO Saving model and optimizer state at iteration 226 to ./logs/44k/G_8800.pth
344
+ 2023-03-18 16:42:09,165 44k INFO Saving model and optimizer state at iteration 226 to ./logs/44k/D_8800.pth
345
+ 2023-03-18 16:42:15,680 44k INFO ====> Epoch: 226, cost 26.31 s
346
+ 2023-03-18 16:42:35,788 44k INFO ====> Epoch: 227, cost 20.11 s
347
+ 2023-03-18 16:42:55,674 44k INFO ====> Epoch: 228, cost 19.89 s
348
+ 2023-03-18 16:43:15,864 44k INFO ====> Epoch: 229, cost 20.19 s
349
+ 2023-03-18 16:43:35,656 44k INFO ====> Epoch: 230, cost 19.79 s
350
+ 2023-03-18 16:43:51,625 44k INFO Train Epoch: 231 [77%]
351
+ 2023-03-18 16:43:51,626 44k INFO Losses: [2.2360494136810303, 2.6518003940582275, 10.69636344909668, 20.383787155151367, 0.5332014560699463], step: 9000, lr: 9.716576028476738e-05
352
+ 2023-03-18 16:43:56,079 44k INFO ====> Epoch: 231, cost 20.42 s
353
+ 2023-03-18 16:44:15,806 44k INFO ====> Epoch: 232, cost 19.73 s
354
+ 2023-03-18 16:44:35,609 44k INFO ====> Epoch: 233, cost 19.80 s
355
+ 2023-03-18 16:44:55,628 44k INFO ====> Epoch: 234, cost 20.02 s
356
+ 2023-03-18 16:45:15,712 44k INFO ====> Epoch: 235, cost 20.08 s
357
+ 2023-03-18 16:45:34,104 44k INFO Train Epoch: 236 [90%]
358
+ 2023-03-18 16:45:34,106 44k INFO Losses: [2.337740421295166, 2.3855791091918945, 10.388580322265625, 17.848812103271484, 0.6865084171295166], step: 9200, lr: 9.710504686484176e-05
359
+ 2023-03-18 16:45:36,420 44k INFO ====> Epoch: 236, cost 20.71 s
360
+ 2023-03-18 16:45:56,364 44k INFO ====> Epoch: 237, cost 19.94 s
361
+ 2023-03-18 16:46:16,415 44k INFO ====> Epoch: 238, cost 20.05 s
362
+ 2023-03-18 16:46:36,407 44k INFO ====> Epoch: 239, cost 19.99 s
363
+ 2023-03-18 16:46:56,603 44k INFO ====> Epoch: 240, cost 20.20 s
364
+ 2023-03-18 16:47:16,624 44k INFO ====> Epoch: 241, cost 20.02 s
365
+ 2023-03-18 16:47:20,599 44k INFO Train Epoch: 242 [3%]
366
+ 2023-03-18 16:47:20,601 44k INFO Losses: [2.1655261516571045, 2.7926673889160156, 13.443987846374512, 23.91498565673828, 0.7444508075714111], step: 9400, lr: 9.703224083489565e-05
367
+ 2023-03-18 16:47:37,437 44k INFO ====> Epoch: 242, cost 20.81 s
368
+ 2023-03-18 16:47:58,544 44k INFO ====> Epoch: 243, cost 21.11 s
369
+ 2023-03-18 16:48:18,497 44k INFO ====> Epoch: 244, cost 19.95 s
370
+ 2023-03-18 16:48:38,517 44k INFO ====> Epoch: 245, cost 20.02 s
371
+ 2023-03-18 16:48:58,373 44k INFO ====> Epoch: 246, cost 19.86 s
372
+ 2023-03-18 16:49:04,445 44k INFO Train Epoch: 247 [15%]
373
+ 2023-03-18 16:49:04,446 44k INFO Losses: [2.8257107734680176, 2.245692253112793, 6.386197090148926, 12.429339408874512, 0.5741127133369446], step: 9600, lr: 9.69716108437664e-05
374
+ 2023-03-18 16:49:08,880 44k INFO Saving model and optimizer state at iteration 247 to ./logs/44k/G_9600.pth
375
+ 2023-03-18 16:49:09,927 44k INFO Saving model and optimizer state at iteration 247 to ./logs/44k/D_9600.pth
376
+ 2023-03-18 16:49:24,184 44k INFO ====> Epoch: 247, cost 25.81 s
377
+ 2023-03-18 16:49:44,408 44k INFO ====> Epoch: 248, cost 20.22 s
378
+ 2023-03-18 16:50:04,256 44k INFO ====> Epoch: 249, cost 19.85 s
379
+ 2023-03-18 16:50:24,296 44k INFO ====> Epoch: 250, cost 20.04 s
380
+ 2023-03-18 16:50:44,262 44k INFO ====> Epoch: 251, cost 19.97 s
381
+ 2023-03-18 16:50:52,950 44k INFO Train Epoch: 252 [28%]
382
+ 2023-03-18 16:50:52,951 44k INFO Losses: [2.4544928073883057, 2.261460781097412, 7.994678974151611, 17.743366241455078, 0.5375484824180603], step: 9800, lr: 9.691101873690936e-05
383
+ 2023-03-18 16:51:05,304 44k INFO ====> Epoch: 252, cost 21.04 s
384
+ 2023-03-18 16:51:25,125 44k INFO ====> Epoch: 253, cost 19.82 s
385
+ 2023-03-18 16:51:44,876 44k INFO ====> Epoch: 254, cost 19.75 s
386
+ 2023-03-18 16:52:04,495 44k INFO ====> Epoch: 255, cost 19.62 s
387
+ 2023-03-18 16:52:24,331 44k INFO ====> Epoch: 256, cost 19.84 s
388
+ 2023-03-18 16:52:34,494 44k INFO Train Epoch: 257 [41%]
389
+ 2023-03-18 16:52:34,495 44k INFO Losses: [2.2319655418395996, 2.4747824668884277, 10.411455154418945, 17.785585403442383, 0.6818886995315552], step: 10000, lr: 9.685046449065278e-05
390
+ 2023-03-18 16:52:44,934 44k INFO ====> Epoch: 257, cost 20.60 s
391
+ 2023-03-18 16:53:05,038 44k INFO ====> Epoch: 258, cost 20.10 s
392
+ 2023-03-18 16:53:24,951 44k INFO ====> Epoch: 259, cost 19.91 s
393
+ 2023-03-18 16:53:44,824 44k INFO ====> Epoch: 260, cost 19.87 s
394
+ 2023-03-18 16:54:04,673 44k INFO ====> Epoch: 261, cost 19.85 s
395
+ 2023-03-18 16:54:17,273 44k INFO Train Epoch: 262 [54%]
396
+ 2023-03-18 16:54:17,275 44k INFO Losses: [2.368034839630127, 2.4957849979400635, 11.374425888061523, 21.15058135986328, 0.9322990775108337], step: 10200, lr: 9.678994808133967e-05
397
+ 2023-03-18 16:54:25,409 44k INFO ====> Epoch: 262, cost 20.74 s
398
+ 2023-03-18 16:54:45,150 44k INFO ====> Epoch: 263, cost 19.74 s
399
+ 2023-03-18 16:55:05,117 44k INFO ====> Epoch: 264, cost 19.97 s
400
+ 2023-03-18 16:55:25,454 44k INFO ====> Epoch: 265, cost 20.34 s
401
+ 2023-03-18 16:55:45,558 44k INFO ====> Epoch: 266, cost 20.10 s
402
+ 2023-03-18 16:56:00,233 44k INFO Train Epoch: 267 [67%]
403
+ 2023-03-18 16:56:00,234 44k INFO Losses: [2.170872926712036, 2.5498008728027344, 12.449470520019531, 20.475133895874023, 0.5462685823440552], step: 10400, lr: 9.67294694853279e-05
404
+ 2023-03-18 16:56:04,770 44k INFO Saving model and optimizer state at iteration 267 to ./logs/44k/G_10400.pth
405
+ 2023-03-18 16:56:05,822 44k INFO Saving model and optimizer state at iteration 267 to ./logs/44k/D_10400.pth
406
+ 2023-03-18 16:56:11,928 44k INFO ====> Epoch: 267, cost 26.37 s
407
+ 2023-03-18 16:56:31,938 44k INFO ====> Epoch: 268, cost 20.01 s
408
+ 2023-03-18 16:56:51,823 44k INFO ====> Epoch: 269, cost 19.89 s
409
+ 2023-03-18 16:57:11,802 44k INFO ====> Epoch: 270, cost 19.98 s
410
+ 2023-03-18 16:57:31,690 44k INFO ====> Epoch: 271, cost 19.89 s
411
+ 2023-03-18 16:57:48,179 44k INFO Train Epoch: 272 [79%]
412
+ 2023-03-18 16:57:48,181 44k INFO Losses: [2.1866960525512695, 2.445249319076538, 10.951682090759277, 19.355749130249023, 0.6413354873657227], step: 10600, lr: 9.666902867899003e-05
413
+ 2023-03-18 16:57:52,250 44k INFO ====> Epoch: 272, cost 20.56 s
414
+ 2023-03-18 16:58:12,989 44k INFO ====> Epoch: 273, cost 20.74 s
415
+ 2023-03-18 16:58:32,781 44k INFO ====> Epoch: 274, cost 19.79 s
416
+ 2023-03-18 16:58:52,693 44k INFO ====> Epoch: 275, cost 19.91 s
417
+ 2023-03-18 16:59:12,527 44k INFO ====> Epoch: 276, cost 19.83 s
418
+ 2023-03-18 16:59:31,113 44k INFO Train Epoch: 277 [92%]
419
+ 2023-03-18 16:59:31,115 44k INFO Losses: [2.256988525390625, 2.6625442504882812, 9.01707649230957, 18.851524353027344, 0.5929552316665649], step: 10800, lr: 9.660862563871342e-05
420
+ 2023-03-18 16:59:33,039 44k INFO ====> Epoch: 277, cost 20.51 s
421
+ 2023-03-18 16:59:53,278 44k INFO ====> Epoch: 278, cost 20.24 s
422
+ 2023-03-18 17:00:13,389 44k INFO ====> Epoch: 279, cost 20.11 s
423
+ 2023-03-18 17:00:33,239 44k INFO ====> Epoch: 280, cost 19.85 s
424
+ 2023-03-18 17:00:53,087 44k INFO ====> Epoch: 281, cost 19.85 s
425
+ 2023-03-18 17:01:14,621 44k INFO ====> Epoch: 282, cost 21.53 s
426
+ 2023-03-18 17:01:18,980 44k INFO Train Epoch: 283 [5%]
427
+ 2023-03-18 17:01:18,981 44k INFO Losses: [2.56038761138916, 2.2084388732910156, 9.195663452148438, 16.996200561523438, 0.7333958745002747], step: 11000, lr: 9.653619180835758e-05
428
+ 2023-03-18 17:01:35,126 44k INFO ====> Epoch: 283, cost 20.50 s
429
+ 2023-03-18 17:01:54,863 44k INFO ====> Epoch: 284, cost 19.74 s
430
+ 2023-03-18 17:02:14,868 44k INFO ====> Epoch: 285, cost 20.01 s
431
+ 2023-03-18 17:02:34,792 44k INFO ====> Epoch: 286, cost 19.92 s
432
+ 2023-03-18 17:02:54,864 44k INFO ====> Epoch: 287, cost 20.07 s
433
+ 2023-03-18 17:03:01,838 44k INFO Train Epoch: 288 [18%]
434
+ 2023-03-18 17:03:01,839 44k INFO Losses: [2.502936363220215, 2.2773265838623047, 9.0471773147583, 18.72901725769043, 0.9384689331054688], step: 11200, lr: 9.647587177037196e-05
435
+ 2023-03-18 17:03:06,229 44k INFO Saving model and optimizer state at iteration 288 to ./logs/44k/G_11200.pth
436
+ 2023-03-18 17:03:07,422 44k INFO Saving model and optimizer state at iteration 288 to ./logs/44k/D_11200.pth
437
+ 2023-03-18 17:03:21,525 44k INFO ====> Epoch: 288, cost 26.66 s
438
+ 2023-03-18 17:03:41,370 44k INFO ====> Epoch: 289, cost 19.85 s
439
+ 2023-03-18 17:04:01,209 44k INFO ====> Epoch: 290, cost 19.84 s
440
+ 2023-03-18 17:04:22,790 44k INFO ====> Epoch: 291, cost 21.58 s
441
+ 2023-03-18 17:04:42,738 44k INFO ====> Epoch: 292, cost 19.95 s
442
+ 2023-03-18 17:04:51,194 44k INFO Train Epoch: 293 [31%]
443
+ 2023-03-18 17:04:51,196 44k INFO Losses: [2.4107394218444824, 2.5568552017211914, 11.272177696228027, 19.823131561279297, 0.7836748957633972], step: 11400, lr: 9.641558942298625e-05
444
+ 2023-03-18 17:05:03,095 44k INFO ====> Epoch: 293, cost 20.36 s
445
+ 2023-03-18 17:05:23,282 44k INFO ====> Epoch: 294, cost 20.19 s
446
+ 2023-03-18 17:05:43,171 44k INFO ====> Epoch: 295, cost 19.89 s
447
+ 2023-03-18 17:06:03,498 44k INFO ====> Epoch: 296, cost 20.33 s
448
+ 2023-03-18 17:06:23,305 44k INFO ====> Epoch: 297, cost 19.81 s
449
+ 2023-03-18 17:06:33,864 44k INFO Train Epoch: 298 [44%]
450
+ 2023-03-18 17:06:33,865 44k INFO Losses: [2.7009692192077637, 2.125077962875366, 7.26710319519043, 11.231626510620117, 0.8319841623306274], step: 11600, lr: 9.635534474264972e-05
451
+ 2023-03-18 17:06:43,833 44k INFO ====> Epoch: 298, cost 20.53 s
452
+ 2023-03-18 17:07:03,635 44k INFO ====> Epoch: 299, cost 19.80 s
453
+ 2023-03-18 17:07:23,150 44k INFO ====> Epoch: 300, cost 19.51 s
454
+ 2023-03-18 17:07:43,090 44k INFO ====> Epoch: 301, cost 19.94 s
455
+ 2023-03-18 17:08:03,333 44k INFO ====> Epoch: 302, cost 20.24 s
456
+ 2023-03-18 17:08:16,012 44k INFO Train Epoch: 303 [56%]
457
+ 2023-03-18 17:08:16,013 44k INFO Losses: [2.552933931350708, 2.472691059112549, 9.617679595947266, 20.058349609375, 0.4773886203765869], step: 11800, lr: 9.629513770582634e-05
458
+ 2023-03-18 17:08:23,517 44k INFO ====> Epoch: 303, cost 20.18 s
459
+ 2023-03-18 17:08:43,187 44k INFO ====> Epoch: 304, cost 19.67 s
460
+ 2023-03-18 17:09:03,569 44k INFO ====> Epoch: 305, cost 20.38 s
461
+ 2023-03-18 17:09:23,391 44k INFO ====> Epoch: 306, cost 19.82 s
462
+ 2023-03-18 17:09:42,960 44k INFO ====> Epoch: 307, cost 19.57 s
463
+ 2023-03-18 17:09:57,696 44k INFO Train Epoch: 308 [69%]
464
+ 2023-03-18 17:09:57,697 44k INFO Losses: [2.641923427581787, 2.117219924926758, 10.303349494934082, 20.380632400512695, 0.3930005133152008], step: 12000, lr: 9.62349682889948e-05
465
+ 2023-03-18 17:10:01,898 44k INFO Saving model and optimizer state at iteration 308 to ./logs/44k/G_12000.pth
466
+ 2023-03-18 17:10:02,834 44k INFO Saving model and optimizer state at iteration 308 to ./logs/44k/D_12000.pth
467
+ 2023-03-18 17:10:08,387 44k INFO ====> Epoch: 308, cost 25.43 s
468
+ 2023-03-18 17:10:28,754 44k INFO ====> Epoch: 309, cost 20.37 s
469
+ 2023-03-18 17:10:48,349 44k INFO ====> Epoch: 310, cost 19.60 s
470
+ 2023-03-18 17:11:09,263 44k INFO ====> Epoch: 311, cost 20.91 s
471
+ 2023-03-18 17:11:28,821 44k INFO ====> Epoch: 312, cost 19.56 s
472
+ 2023-03-18 17:11:45,379 44k INFO Train Epoch: 313 [82%]
473
+ 2023-03-18 17:11:45,380 44k INFO Losses: [2.7167956829071045, 2.2329747676849365, 7.7941179275512695, 16.260107040405273, 0.6568765044212341], step: 12200, lr: 9.617483646864849e-05
474
+ 2023-03-18 17:11:48,865 44k INFO ====> Epoch: 313, cost 20.04 s
475
+ 2023-03-18 17:12:08,448 44k INFO ====> Epoch: 314, cost 19.58 s
476
+ 2023-03-18 17:12:28,134 44k INFO ====> Epoch: 315, cost 19.69 s
477
+ 2023-03-18 17:12:47,471 44k INFO ====> Epoch: 316, cost 19.34 s
478
+ 2023-03-18 17:13:06,821 44k INFO ====> Epoch: 317, cost 19.35 s
479
+ 2023-03-18 17:13:25,502 44k INFO Train Epoch: 318 [95%]
480
+ 2023-03-18 17:13:25,503 44k INFO Losses: [2.3885555267333984, 2.3887124061584473, 7.18328857421875, 17.944881439208984, 0.5438012480735779], step: 12400, lr: 9.611474222129547e-05
481
+ 2023-03-18 17:13:27,095 44k INFO ====> Epoch: 318, cost 20.27 s
482
+ 2023-03-18 17:13:46,557 44k INFO ====> Epoch: 319, cost 19.46 s
483
+ 2023-03-18 17:14:06,371 44k INFO ====> Epoch: 320, cost 19.81 s
484
+ 2023-03-18 17:14:26,017 44k INFO ====> Epoch: 321, cost 19.65 s
485
+ 2023-03-18 17:14:45,545 44k INFO ====> Epoch: 322, cost 19.53 s
486
+ 2023-03-18 17:15:04,846 44k INFO ====> Epoch: 323, cost 19.30 s
487
+ 2023-03-18 17:15:09,498 44k INFO Train Epoch: 324 [8%]
488
+ 2023-03-18 17:15:09,499 44k INFO Losses: [2.1702542304992676, 2.652705430984497, 11.307723045349121, 21.379201889038086, 0.8918676972389221], step: 12600, lr: 9.604267868776807e-05
489
+ 2023-03-18 17:15:24,846 44k INFO ====> Epoch: 324, cost 20.00 s
490
+ 2023-03-18 17:15:44,240 44k INFO ====> Epoch: 325, cost 19.39 s
491
+ 2023-03-18 17:16:03,710 44k INFO ====> Epoch: 326, cost 19.47 s
492
+ 2023-03-18 17:16:23,110 44k INFO ====> Epoch: 327, cost 19.40 s
493
+ 2023-03-18 17:16:42,685 44k INFO ====> Epoch: 328, cost 19.58 s
494
+ 2023-03-18 17:16:49,386 44k INFO Train Epoch: 329 [21%]
495
+ 2023-03-18 17:16:49,387 44k INFO Losses: [2.395970344543457, 2.3684022426605225, 10.31136417388916, 17.88117790222168, 0.6932646036148071], step: 12800, lr: 9.5982667018381e-05
496
+ 2023-03-18 17:16:53,635 44k INFO Saving model and optimizer state at iteration 329 to ./logs/44k/G_12800.pth
497
+ 2023-03-18 17:16:54,607 44k INFO Saving model and optimizer state at iteration 329 to ./logs/44k/D_12800.pth
498
+ 2023-03-18 17:17:08,139 44k INFO ====> Epoch: 329, cost 25.45 s
so-vits-svc-4.0/20230324-Juewa-44k/best/D_12800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa3baa70695e03d963e3ac873c87760b0ca6d02b3b9e94c9716ffe258ec6130d
3
+ size 561098185
so-vits-svc-4.0/20230324-Juewa-44k/best/G_12800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5496c02bff57cd0614a70194b170c43767556242d7b0c8fba4d6ac437f7a9715
3
+ size 542789405
so-vits-svc-4.0/20230324-Juewa-44k/best/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "Juewa": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230324-Juewa-44k/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "Juewa": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230324-Juewa-44k/kmeans_10000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c8f7393e26f890587694355d6b45b6b2ee055ec679991d5f3f0f291832c9b87
3
+ size 15441785
so-vits-svc-4.0/20230324-Juewa-44k/pruned/P_G_12800.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2a99b10fe6dfda20fd5073d58b7ea71b06c2e648989e0614fe605869fafb0e5
3
+ size 180885829
so-vits-svc-4.0/20230324-Juewa-44k/pruned/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "Juewa": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230324-Juewa-44k/train.log ADDED
@@ -0,0 +1,431 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-03-24 16:10:32,809 44k INFO {'train': {'log_interval': 200, 'eval_interval': 800, 'seed': 1234, 'epochs': 10000, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 6, 'fp16_run': False, 'lr_decay': 0.999875, 'segment_size': 10240, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0, 'use_sr': True, 'max_speclen': 512, 'port': '8001', 'keep_ckpts': 99}, 'data': {'training_files': 'filelists/train.txt', 'validation_files': 'filelists/val.txt', 'max_wav_value': 32768.0, 'sampling_rate': 44100, 'filter_length': 2048, 'hop_length': 512, 'win_length': 2048, 'n_mel_channels': 80, 'mel_fmin': 0.0, 'mel_fmax': 22050}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.1, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [8, 8, 2, 2, 2], 'upsample_initial_channel': 512, 'upsample_kernel_sizes': [16, 16, 4, 4, 4], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 256, 'ssl_dim': 256, 'n_speakers': 200}, 'spk': {'Juewa': 0}, 'model_dir': './logs/44k'}
2
+ 2023-03-24 16:10:32,810 44k WARNING /root/so-vits-svc-4.0 is not a git repository, therefore hash value comparison will be ignored.
3
+ 2023-03-24 16:10:35,834 44k INFO emb_g.weight is not in the checkpoint
4
+ 2023-03-24 16:10:35,904 44k INFO Loaded checkpoint './logs/44k/G_0.pth' (iteration 0)
5
+ 2023-03-24 16:10:36,087 44k INFO Loaded checkpoint './logs/44k/D_0.pth' (iteration 0)
6
+ 2023-03-24 16:10:44,069 44k INFO Train Epoch: 1 [0%]
7
+ 2023-03-24 16:10:44,070 44k INFO Losses: [2.755509376525879, 2.2334935665130615, 12.809167861938477, 35.51326370239258, 3.423668384552002], step: 0, lr: 0.0001
8
+ 2023-03-24 16:10:49,834 44k INFO Saving model and optimizer state at iteration 1 to ./logs/44k/G_0.pth
9
+ 2023-03-24 16:10:51,452 44k INFO Saving model and optimizer state at iteration 1 to ./logs/44k/D_0.pth
10
+ 2023-03-24 16:11:16,686 44k INFO ====> Epoch: 1, cost 43.88 s
11
+ 2023-03-24 16:11:42,464 44k INFO ====> Epoch: 2, cost 25.78 s
12
+ 2023-03-24 16:12:07,775 44k INFO ====> Epoch: 3, cost 25.31 s
13
+ 2023-03-24 16:12:32,967 44k INFO ====> Epoch: 4, cost 25.19 s
14
+ 2023-03-24 16:12:36,634 44k INFO Train Epoch: 5 [0%]
15
+ 2023-03-24 16:12:36,636 44k INFO Losses: [2.4600982666015625, 2.0119576454162598, 11.58346939086914, 19.77667236328125, 1.1188390254974365], step: 200, lr: 9.995000937421877e-05
16
+ 2023-03-24 16:12:58,844 44k INFO ====> Epoch: 5, cost 25.88 s
17
+ 2023-03-24 16:13:23,698 44k INFO ====> Epoch: 6, cost 24.85 s
18
+ 2023-03-24 16:13:48,641 44k INFO ====> Epoch: 7, cost 24.94 s
19
+ 2023-03-24 16:14:13,609 44k INFO ====> Epoch: 8, cost 24.97 s
20
+ 2023-03-24 16:14:17,212 44k INFO Train Epoch: 9 [0%]
21
+ 2023-03-24 16:14:17,214 44k INFO Losses: [2.423004627227783, 2.078486919403076, 9.993337631225586, 23.189149856567383, 1.4135551452636719], step: 400, lr: 9.990004373906418e-05
22
+ 2023-03-24 16:14:39,134 44k INFO ====> Epoch: 9, cost 25.52 s
23
+ 2023-03-24 16:15:04,111 44k INFO ====> Epoch: 10, cost 24.98 s
24
+ 2023-03-24 16:15:28,810 44k INFO ====> Epoch: 11, cost 24.70 s
25
+ 2023-03-24 16:15:53,523 44k INFO ====> Epoch: 12, cost 24.71 s
26
+ 2023-03-24 16:15:57,344 44k INFO Train Epoch: 13 [0%]
27
+ 2023-03-24 16:15:57,345 44k INFO Losses: [2.2417097091674805, 2.8413851261138916, 12.686511039733887, 20.66866683959961, 1.215733289718628], step: 600, lr: 9.98501030820433e-05
28
+ 2023-03-24 16:16:19,204 44k INFO ====> Epoch: 13, cost 25.68 s
29
+ 2023-03-24 16:16:43,853 44k INFO ====> Epoch: 14, cost 24.65 s
30
+ 2023-03-24 16:17:08,953 44k INFO ====> Epoch: 15, cost 25.10 s
31
+ 2023-03-24 16:17:33,802 44k INFO ====> Epoch: 16, cost 24.85 s
32
+ 2023-03-24 16:17:37,469 44k INFO Train Epoch: 17 [0%]
33
+ 2023-03-24 16:17:37,471 44k INFO Losses: [2.2982616424560547, 2.2541720867156982, 8.421069145202637, 14.56866455078125, 1.1874828338623047], step: 800, lr: 9.980018739066937e-05
34
+ 2023-03-24 16:17:42,347 44k INFO Saving model and optimizer state at iteration 17 to ./logs/44k/G_800.pth
35
+ 2023-03-24 16:17:43,943 44k INFO Saving model and optimizer state at iteration 17 to ./logs/44k/D_800.pth
36
+ 2023-03-24 16:18:05,588 44k INFO ====> Epoch: 17, cost 31.79 s
37
+ 2023-03-24 16:18:30,523 44k INFO ====> Epoch: 18, cost 24.94 s
38
+ 2023-03-24 16:18:55,233 44k INFO ====> Epoch: 19, cost 24.71 s
39
+ 2023-03-24 16:19:19,823 44k INFO ====> Epoch: 20, cost 24.59 s
40
+ 2023-03-24 16:19:23,336 44k INFO Train Epoch: 21 [0%]
41
+ 2023-03-24 16:19:23,337 44k INFO Losses: [2.8217248916625977, 2.243264675140381, 8.612906455993652, 21.242691040039062, 1.4278812408447266], step: 1000, lr: 9.975029665246193e-05
42
+ 2023-03-24 16:19:44,860 44k INFO ====> Epoch: 21, cost 25.04 s
43
+ 2023-03-24 16:20:09,673 44k INFO ====> Epoch: 22, cost 24.81 s
44
+ 2023-03-24 16:20:34,476 44k INFO ====> Epoch: 23, cost 24.80 s
45
+ 2023-03-24 16:20:59,205 44k INFO ====> Epoch: 24, cost 24.73 s
46
+ 2023-03-24 16:21:03,136 44k INFO Train Epoch: 25 [0%]
47
+ 2023-03-24 16:21:03,137 44k INFO Losses: [2.3520963191986084, 2.346210479736328, 12.567983627319336, 20.541658401489258, 1.4853627681732178], step: 1200, lr: 9.970043085494672e-05
48
+ 2023-03-24 16:21:24,861 44k INFO ====> Epoch: 25, cost 25.66 s
49
+ 2023-03-24 16:21:49,922 44k INFO ====> Epoch: 26, cost 25.06 s
50
+ 2023-03-24 16:22:15,067 44k INFO ====> Epoch: 27, cost 25.15 s
51
+ 2023-03-24 16:22:40,244 44k INFO ====> Epoch: 28, cost 25.18 s
52
+ 2023-03-24 16:22:44,262 44k INFO Train Epoch: 29 [0%]
53
+ 2023-03-24 16:22:44,263 44k INFO Losses: [2.468628168106079, 2.170118808746338, 7.571216106414795, 20.927648544311523, 1.2605253458023071], step: 1400, lr: 9.965058998565574e-05
54
+ 2023-03-24 16:23:06,111 44k INFO ====> Epoch: 29, cost 25.87 s
55
+ 2023-03-24 16:23:31,293 44k INFO ====> Epoch: 30, cost 25.18 s
56
+ 2023-03-24 16:23:56,555 44k INFO ====> Epoch: 31, cost 25.26 s
57
+ 2023-03-24 16:24:21,726 44k INFO ====> Epoch: 32, cost 25.17 s
58
+ 2023-03-24 16:24:25,731 44k INFO Train Epoch: 33 [0%]
59
+ 2023-03-24 16:24:25,733 44k INFO Losses: [2.46799373626709, 2.3533313274383545, 11.058666229248047, 22.450557708740234, 1.4724372625350952], step: 1600, lr: 9.960077403212722e-05
60
+ 2023-03-24 16:24:30,814 44k INFO Saving model and optimizer state at iteration 33 to ./logs/44k/G_1600.pth
61
+ 2023-03-24 16:24:32,265 44k INFO Saving model and optimizer state at iteration 33 to ./logs/44k/D_1600.pth
62
+ 2023-03-24 16:24:54,314 44k INFO ====> Epoch: 33, cost 32.59 s
63
+ 2023-03-24 16:25:19,632 44k INFO ====> Epoch: 34, cost 25.32 s
64
+ 2023-03-24 16:25:45,327 44k INFO ====> Epoch: 35, cost 25.70 s
65
+ 2023-03-24 16:26:10,644 44k INFO ====> Epoch: 36, cost 25.32 s
66
+ 2023-03-24 16:26:14,616 44k INFO Train Epoch: 37 [0%]
67
+ 2023-03-24 16:26:14,618 44k INFO Losses: [2.637969970703125, 2.1250932216644287, 8.898857116699219, 19.450668334960938, 1.0567777156829834], step: 1800, lr: 9.95509829819056e-05
68
+ 2023-03-24 16:26:36,678 44k INFO ====> Epoch: 37, cost 26.03 s
69
+ 2023-03-24 16:27:01,814 44k INFO ====> Epoch: 38, cost 25.14 s
70
+ 2023-03-24 16:27:26,722 44k INFO ====> Epoch: 39, cost 24.91 s
71
+ 2023-03-24 16:27:51,749 44k INFO ====> Epoch: 40, cost 25.03 s
72
+ 2023-03-24 16:27:55,388 44k INFO Train Epoch: 41 [0%]
73
+ 2023-03-24 16:27:55,390 44k INFO Losses: [2.1182312965393066, 2.289332151412964, 11.968257904052734, 19.403751373291016, 1.1568026542663574], step: 2000, lr: 9.950121682254156e-05
74
+ 2023-03-24 16:28:17,070 44k INFO ====> Epoch: 41, cost 25.32 s
75
+ 2023-03-24 16:28:41,874 44k INFO ====> Epoch: 42, cost 24.80 s
76
+ 2023-03-24 16:29:06,865 44k INFO ====> Epoch: 43, cost 24.99 s
77
+ 2023-03-24 16:29:31,604 44k INFO ====> Epoch: 44, cost 24.74 s
78
+ 2023-03-24 16:29:35,467 44k INFO Train Epoch: 45 [0%]
79
+ 2023-03-24 16:29:35,468 44k INFO Losses: [2.7747840881347656, 2.2147929668426514, 8.382989883422852, 18.546131134033203, 1.1774940490722656], step: 2200, lr: 9.945147554159202e-05
80
+ 2023-03-24 16:29:56,996 44k INFO ====> Epoch: 45, cost 25.39 s
81
+ 2023-03-24 16:30:21,956 44k INFO ====> Epoch: 46, cost 24.96 s
82
+ 2023-03-24 16:30:46,873 44k INFO ====> Epoch: 47, cost 24.92 s
83
+ 2023-03-24 16:31:11,765 44k INFO ====> Epoch: 48, cost 24.89 s
84
+ 2023-03-24 16:31:15,566 44k INFO Train Epoch: 49 [0%]
85
+ 2023-03-24 16:31:15,567 44k INFO Losses: [2.438877582550049, 2.4737958908081055, 10.301446914672852, 21.869646072387695, 1.067798137664795], step: 2400, lr: 9.940175912662009e-05
86
+ 2023-03-24 16:31:20,492 44k INFO Saving model and optimizer state at iteration 49 to ./logs/44k/G_2400.pth
87
+ 2023-03-24 16:31:21,957 44k INFO Saving model and optimizer state at iteration 49 to ./logs/44k/D_2400.pth
88
+ 2023-03-24 16:31:43,508 44k INFO ====> Epoch: 49, cost 31.74 s
89
+ 2023-03-24 16:32:08,428 44k INFO ====> Epoch: 50, cost 24.92 s
90
+ 2023-03-24 16:32:33,099 44k INFO ====> Epoch: 51, cost 24.67 s
91
+ 2023-03-24 16:32:57,933 44k INFO ====> Epoch: 52, cost 24.83 s
92
+ 2023-03-24 16:33:01,799 44k INFO Train Epoch: 53 [0%]
93
+ 2023-03-24 16:33:01,801 44k INFO Losses: [2.3304405212402344, 2.397515058517456, 13.642292976379395, 21.823060989379883, 1.1119345426559448], step: 2600, lr: 9.935206756519513e-05
94
+ 2023-03-24 16:33:23,375 44k INFO ====> Epoch: 53, cost 25.44 s
95
+ 2023-03-24 16:33:49,208 44k INFO ====> Epoch: 54, cost 25.83 s
96
+ 2023-03-24 16:34:14,334 44k INFO ====> Epoch: 55, cost 25.13 s
97
+ 2023-03-24 16:34:39,142 44k INFO ====> Epoch: 56, cost 24.81 s
98
+ 2023-03-24 16:34:43,036 44k INFO Train Epoch: 57 [0%]
99
+ 2023-03-24 16:34:43,038 44k INFO Losses: [2.4579014778137207, 2.3200392723083496, 8.339593887329102, 20.653188705444336, 1.1132451295852661], step: 2800, lr: 9.930240084489267e-05
100
+ 2023-03-24 16:35:04,540 44k INFO ====> Epoch: 57, cost 25.40 s
101
+ 2023-03-24 16:35:29,399 44k INFO ====> Epoch: 58, cost 24.86 s
102
+ 2023-03-24 16:35:54,290 44k INFO ====> Epoch: 59, cost 24.89 s
103
+ 2023-03-24 16:36:19,055 44k INFO ====> Epoch: 60, cost 24.77 s
104
+ 2023-03-24 16:36:22,860 44k INFO Train Epoch: 61 [0%]
105
+ 2023-03-24 16:36:22,862 44k INFO Losses: [2.3698558807373047, 2.2452638149261475, 9.925782203674316, 20.825319290161133, 0.6393007040023804], step: 3000, lr: 9.92527589532945e-05
106
+ 2023-03-24 16:36:44,782 44k INFO ====> Epoch: 61, cost 25.73 s
107
+ 2023-03-24 16:37:09,557 44k INFO ====> Epoch: 62, cost 24.78 s
108
+ 2023-03-24 16:37:34,361 44k INFO ====> Epoch: 63, cost 24.80 s
109
+ 2023-03-24 16:37:59,183 44k INFO ====> Epoch: 64, cost 24.82 s
110
+ 2023-03-24 16:38:02,824 44k INFO Train Epoch: 65 [0%]
111
+ 2023-03-24 16:38:02,826 44k INFO Losses: [2.4456441402435303, 2.655974864959717, 11.801567077636719, 19.65692901611328, 0.9295780062675476], step: 3200, lr: 9.92031418779886e-05
112
+ 2023-03-24 16:38:07,725 44k INFO Saving model and optimizer state at iteration 65 to ./logs/44k/G_3200.pth
113
+ 2023-03-24 16:38:09,193 44k INFO Saving model and optimizer state at iteration 65 to ./logs/44k/D_3200.pth
114
+ 2023-03-24 16:38:30,997 44k INFO ====> Epoch: 65, cost 31.81 s
115
+ 2023-03-24 16:38:56,138 44k INFO ====> Epoch: 66, cost 25.14 s
116
+ 2023-03-24 16:39:21,052 44k INFO ====> Epoch: 67, cost 24.91 s
117
+ 2023-03-24 16:39:46,032 44k INFO ====> Epoch: 68, cost 24.98 s
118
+ 2023-03-24 16:39:49,805 44k INFO Train Epoch: 69 [0%]
119
+ 2023-03-24 16:39:49,807 44k INFO Losses: [2.511517286300659, 2.0972821712493896, 9.085113525390625, 18.71324348449707, 1.1218843460083008], step: 3400, lr: 9.915354960656915e-05
120
+ 2023-03-24 16:40:11,500 44k INFO ====> Epoch: 69, cost 25.47 s
121
+ 2023-03-24 16:40:36,372 44k INFO ====> Epoch: 70, cost 24.87 s
122
+ 2023-03-24 16:41:01,150 44k INFO ====> Epoch: 71, cost 24.78 s
123
+ 2023-03-24 16:41:25,926 44k INFO ====> Epoch: 72, cost 24.78 s
124
+ 2023-03-24 16:41:29,560 44k INFO Train Epoch: 73 [0%]
125
+ 2023-03-24 16:41:29,561 44k INFO Losses: [2.4871113300323486, 2.247950553894043, 9.168888092041016, 21.552160263061523, 0.9532811641693115], step: 3600, lr: 9.910398212663652e-05
126
+ 2023-03-24 16:41:51,435 44k INFO ====> Epoch: 73, cost 25.51 s
127
+ 2023-03-24 16:42:16,376 44k INFO ====> Epoch: 74, cost 24.94 s
128
+ 2023-03-24 16:42:41,110 44k INFO ====> Epoch: 75, cost 24.73 s
129
+ 2023-03-24 16:43:06,273 44k INFO ====> Epoch: 76, cost 25.16 s
130
+ 2023-03-24 16:43:09,995 44k INFO Train Epoch: 77 [0%]
131
+ 2023-03-24 16:43:09,996 44k INFO Losses: [3.0498757362365723, 2.0931332111358643, 9.740159034729004, 20.8754940032959, 0.8113635182380676], step: 3800, lr: 9.905443942579728e-05
132
+ 2023-03-24 16:43:31,648 44k INFO ====> Epoch: 77, cost 25.38 s
133
+ 2023-03-24 16:43:56,531 44k INFO ====> Epoch: 78, cost 24.88 s
134
+ 2023-03-24 16:44:21,788 44k INFO ====> Epoch: 79, cost 25.26 s
135
+ 2023-03-24 16:44:46,863 44k INFO ====> Epoch: 80, cost 25.07 s
136
+ 2023-03-24 16:44:50,520 44k INFO Train Epoch: 81 [0%]
137
+ 2023-03-24 16:44:50,521 44k INFO Losses: [2.3789913654327393, 2.4813644886016846, 10.093775749206543, 17.8828182220459, 0.9886296987533569], step: 4000, lr: 9.900492149166423e-05
138
+ 2023-03-24 16:44:55,336 44k INFO Saving model and optimizer state at iteration 81 to ./logs/44k/G_4000.pth
139
+ 2023-03-24 16:44:56,786 44k INFO Saving model and optimizer state at iteration 81 to ./logs/44k/D_4000.pth
140
+ 2023-03-24 16:45:18,665 44k INFO ====> Epoch: 81, cost 31.80 s
141
+ 2023-03-24 16:45:43,655 44k INFO ====> Epoch: 82, cost 24.99 s
142
+ 2023-03-24 16:46:08,720 44k INFO ====> Epoch: 83, cost 25.06 s
143
+ 2023-03-24 16:46:33,650 44k INFO ====> Epoch: 84, cost 24.93 s
144
+ 2023-03-24 16:46:37,358 44k INFO Train Epoch: 85 [0%]
145
+ 2023-03-24 16:46:37,360 44k INFO Losses: [2.3212366104125977, 1.990512490272522, 8.910666465759277, 21.229427337646484, 0.4932689964771271], step: 4200, lr: 9.895542831185631e-05
146
+ 2023-03-24 16:46:59,141 44k INFO ====> Epoch: 85, cost 25.49 s
147
+ 2023-03-24 16:47:24,077 44k INFO ====> Epoch: 86, cost 24.94 s
148
+ 2023-03-24 16:47:49,217 44k INFO ====> Epoch: 87, cost 25.14 s
149
+ 2023-03-24 16:48:14,071 44k INFO ====> Epoch: 88, cost 24.85 s
150
+ 2023-03-24 16:48:18,000 44k INFO Train Epoch: 89 [0%]
151
+ 2023-03-24 16:48:18,002 44k INFO Losses: [2.222428560256958, 2.3397562503814697, 15.815887451171875, 22.820709228515625, 0.802300751209259], step: 4400, lr: 9.89059598739987e-05
152
+ 2023-03-24 16:48:40,075 44k INFO ====> Epoch: 89, cost 26.00 s
153
+ 2023-03-24 16:49:05,197 44k INFO ====> Epoch: 90, cost 25.12 s
154
+ 2023-03-24 16:49:30,107 44k INFO ====> Epoch: 91, cost 24.91 s
155
+ 2023-03-24 16:49:55,009 44k INFO ====> Epoch: 92, cost 24.90 s
156
+ 2023-03-24 16:49:58,820 44k INFO Train Epoch: 93 [0%]
157
+ 2023-03-24 16:49:58,822 44k INFO Losses: [2.3582277297973633, 2.0633132457733154, 10.956375122070312, 19.99727439880371, 1.0134141445159912], step: 4600, lr: 9.885651616572276e-05
158
+ 2023-03-24 16:50:20,550 44k INFO ====> Epoch: 93, cost 25.54 s
159
+ 2023-03-24 16:50:45,626 44k INFO ====> Epoch: 94, cost 25.08 s
160
+ 2023-03-24 16:51:10,825 44k INFO ====> Epoch: 95, cost 25.20 s
161
+ 2023-03-24 16:51:35,826 44k INFO ====> Epoch: 96, cost 25.00 s
162
+ 2023-03-24 16:51:39,583 44k INFO Train Epoch: 97 [0%]
163
+ 2023-03-24 16:51:39,584 44k INFO Losses: [2.5545666217803955, 2.3805904388427734, 11.227080345153809, 19.633548736572266, 0.9872961044311523], step: 4800, lr: 9.880709717466598e-05
164
+ 2023-03-24 16:51:44,661 44k INFO Saving model and optimizer state at iteration 97 to ./logs/44k/G_4800.pth
165
+ 2023-03-24 16:51:46,114 44k INFO Saving model and optimizer state at iteration 97 to ./logs/44k/D_4800.pth
166
+ 2023-03-24 16:52:07,684 44k INFO ====> Epoch: 97, cost 31.86 s
167
+ 2023-03-24 16:52:32,269 44k INFO ====> Epoch: 98, cost 24.58 s
168
+ 2023-03-24 16:52:57,267 44k INFO ====> Epoch: 99, cost 25.00 s
169
+ 2023-03-24 16:53:22,366 44k INFO ====> Epoch: 100, cost 25.10 s
170
+ 2023-03-24 16:53:26,197 44k INFO Train Epoch: 101 [0%]
171
+ 2023-03-24 16:53:26,198 44k INFO Losses: [2.5303902626037598, 2.4984896183013916, 8.828944206237793, 17.48539161682129, 0.9228643178939819], step: 5000, lr: 9.875770288847208e-05
172
+ 2023-03-24 16:53:48,025 44k INFO ====> Epoch: 101, cost 25.66 s
173
+ 2023-03-24 16:54:12,974 44k INFO ====> Epoch: 102, cost 24.95 s
174
+ 2023-03-24 16:54:37,657 44k INFO ====> Epoch: 103, cost 24.68 s
175
+ 2023-03-24 16:55:02,369 44k INFO ====> Epoch: 104, cost 24.71 s
176
+ 2023-03-24 16:55:06,175 44k INFO Train Epoch: 105 [0%]
177
+ 2023-03-24 16:55:06,176 44k INFO Losses: [2.65112042427063, 2.510831832885742, 9.358697891235352, 21.029691696166992, 0.7588745951652527], step: 5200, lr: 9.870833329479095e-05
178
+ 2023-03-24 16:55:27,954 44k INFO ====> Epoch: 105, cost 25.58 s
179
+ 2023-03-24 16:55:52,961 44k INFO ====> Epoch: 106, cost 25.01 s
180
+ 2023-03-24 16:56:18,063 44k INFO ====> Epoch: 107, cost 25.10 s
181
+ 2023-03-24 16:56:43,409 44k INFO ====> Epoch: 108, cost 25.35 s
182
+ 2023-03-24 16:56:47,248 44k INFO Train Epoch: 109 [0%]
183
+ 2023-03-24 16:56:47,250 44k INFO Losses: [2.5201327800750732, 2.3836162090301514, 9.467650413513184, 21.405014038085938, 1.1192044019699097], step: 5400, lr: 9.865898838127865e-05
184
+ 2023-03-24 16:57:08,895 44k INFO ====> Epoch: 109, cost 25.49 s
185
+ 2023-03-24 16:57:33,807 44k INFO ====> Epoch: 110, cost 24.91 s
186
+ 2023-03-24 16:57:58,673 44k INFO ====> Epoch: 111, cost 24.87 s
187
+ 2023-03-24 16:58:23,665 44k INFO ====> Epoch: 112, cost 24.99 s
188
+ 2023-03-24 16:58:27,541 44k INFO Train Epoch: 113 [0%]
189
+ 2023-03-24 16:58:27,542 44k INFO Losses: [2.516652822494507, 1.9198682308197021, 12.64153003692627, 21.67121124267578, 0.9236412644386292], step: 5600, lr: 9.86096681355974e-05
190
+ 2023-03-24 16:58:32,390 44k INFO Saving model and optimizer state at iteration 113 to ./logs/44k/G_5600.pth
191
+ 2023-03-24 16:58:33,837 44k INFO Saving model and optimizer state at iteration 113 to ./logs/44k/D_5600.pth
192
+ 2023-03-24 16:58:55,689 44k INFO ====> Epoch: 113, cost 32.02 s
193
+ 2023-03-24 16:59:20,767 44k INFO ====> Epoch: 114, cost 25.08 s
194
+ 2023-03-24 16:59:45,581 44k INFO ====> Epoch: 115, cost 24.81 s
195
+ 2023-03-24 17:00:10,507 44k INFO ====> Epoch: 116, cost 24.93 s
196
+ 2023-03-24 17:00:14,311 44k INFO Train Epoch: 117 [0%]
197
+ 2023-03-24 17:00:14,312 44k INFO Losses: [2.473012924194336, 2.149684190750122, 7.098073959350586, 17.53461456298828, 0.6158500909805298], step: 5800, lr: 9.85603725454156e-05
198
+ 2023-03-24 17:00:36,112 44k INFO ====> Epoch: 117, cost 25.61 s
199
+ 2023-03-24 17:01:01,203 44k INFO ====> Epoch: 118, cost 25.09 s
200
+ 2023-03-24 17:01:26,147 44k INFO ====> Epoch: 119, cost 24.94 s
201
+ 2023-03-24 17:01:51,183 44k INFO ====> Epoch: 120, cost 25.04 s
202
+ 2023-03-24 17:01:55,020 44k INFO Train Epoch: 121 [0%]
203
+ 2023-03-24 17:01:55,022 44k INFO Losses: [2.182035446166992, 2.618431568145752, 10.66880989074707, 19.815397262573242, 0.8694823980331421], step: 6000, lr: 9.851110159840781e-05
204
+ 2023-03-24 17:02:16,777 44k INFO ====> Epoch: 121, cost 25.59 s
205
+ 2023-03-24 17:02:41,956 44k INFO ====> Epoch: 122, cost 25.18 s
206
+ 2023-03-24 17:03:06,973 44k INFO ====> Epoch: 123, cost 25.02 s
207
+ 2023-03-24 17:03:31,917 44k INFO ====> Epoch: 124, cost 24.94 s
208
+ 2023-03-24 17:03:35,819 44k INFO Train Epoch: 125 [0%]
209
+ 2023-03-24 17:03:35,821 44k INFO Losses: [2.406432628631592, 2.431051731109619, 8.33769416809082, 16.252138137817383, 1.0498191118240356], step: 6200, lr: 9.846185528225477e-05
210
+ 2023-03-24 17:03:57,713 44k INFO ====> Epoch: 125, cost 25.80 s
211
+ 2023-03-24 17:04:22,726 44k INFO ====> Epoch: 126, cost 25.01 s
212
+ 2023-03-24 17:04:47,506 44k INFO ====> Epoch: 127, cost 24.78 s
213
+ 2023-03-24 17:05:12,524 44k INFO ====> Epoch: 128, cost 25.02 s
214
+ 2023-03-24 17:05:16,139 44k INFO Train Epoch: 129 [0%]
215
+ 2023-03-24 17:05:16,140 44k INFO Losses: [2.168229818344116, 2.5004663467407227, 9.397270202636719, 21.672155380249023, 1.026598334312439], step: 6400, lr: 9.841263358464336e-05
216
+ 2023-03-24 17:05:20,927 44k INFO Saving model and optimizer state at iteration 129 to ./logs/44k/G_6400.pth
217
+ 2023-03-24 17:05:22,456 44k INFO Saving model and optimizer state at iteration 129 to ./logs/44k/D_6400.pth
218
+ 2023-03-24 17:05:44,177 44k INFO ====> Epoch: 129, cost 31.65 s
219
+ 2023-03-24 17:06:09,175 44k INFO ====> Epoch: 130, cost 25.00 s
220
+ 2023-03-24 17:06:34,161 44k INFO ====> Epoch: 131, cost 24.99 s
221
+ 2023-03-24 17:06:59,334 44k INFO ====> Epoch: 132, cost 25.17 s
222
+ 2023-03-24 17:07:03,285 44k INFO Train Epoch: 133 [0%]
223
+ 2023-03-24 17:07:03,286 44k INFO Losses: [2.4833321571350098, 1.8855023384094238, 8.147689819335938, 16.213708877563477, 0.935891330242157], step: 6600, lr: 9.836343649326659e-05
224
+ 2023-03-24 17:07:25,196 44k INFO ====> Epoch: 133, cost 25.86 s
225
+ 2023-03-24 17:07:50,174 44k INFO ====> Epoch: 134, cost 24.98 s
226
+ 2023-03-24 17:08:15,531 44k INFO ====> Epoch: 135, cost 25.36 s
227
+ 2023-03-24 17:08:40,614 44k INFO ====> Epoch: 136, cost 25.08 s
228
+ 2023-03-24 17:08:44,405 44k INFO Train Epoch: 137 [0%]
229
+ 2023-03-24 17:08:44,406 44k INFO Losses: [2.6166269779205322, 2.219412088394165, 8.7598295211792, 20.433765411376953, 0.8591962456703186], step: 6800, lr: 9.831426399582366e-05
230
+ 2023-03-24 17:09:06,198 44k INFO ====> Epoch: 137, cost 25.58 s
231
+ 2023-03-24 17:09:31,096 44k INFO ====> Epoch: 138, cost 24.90 s
232
+ 2023-03-24 17:09:56,066 44k INFO ====> Epoch: 139, cost 24.97 s
233
+ 2023-03-24 17:10:21,002 44k INFO ====> Epoch: 140, cost 24.94 s
234
+ 2023-03-24 17:10:24,806 44k INFO Train Epoch: 141 [0%]
235
+ 2023-03-24 17:10:24,807 44k INFO Losses: [2.2800958156585693, 2.6092875003814697, 10.905752182006836, 22.769119262695312, 1.0514625310897827], step: 7000, lr: 9.826511608001993e-05
236
+ 2023-03-24 17:10:46,563 44k INFO ====> Epoch: 141, cost 25.56 s
237
+ 2023-03-24 17:11:11,429 44k INFO ====> Epoch: 142, cost 24.87 s
238
+ 2023-03-24 17:11:36,251 44k INFO ====> Epoch: 143, cost 24.82 s
239
+ 2023-03-24 17:12:00,907 44k INFO ====> Epoch: 144, cost 24.66 s
240
+ 2023-03-24 17:12:04,652 44k INFO Train Epoch: 145 [0%]
241
+ 2023-03-24 17:12:04,654 44k INFO Losses: [2.4433958530426025, 2.1577835083007812, 11.530089378356934, 19.273975372314453, 0.5594630837440491], step: 7200, lr: 9.821599273356685e-05
242
+ 2023-03-24 17:12:09,668 44k INFO Saving model and optimizer state at iteration 145 to ./logs/44k/G_7200.pth
243
+ 2023-03-24 17:12:11,127 44k INFO Saving model and optimizer state at iteration 145 to ./logs/44k/D_7200.pth
244
+ 2023-03-24 17:12:32,817 44k INFO ====> Epoch: 145, cost 31.91 s
245
+ 2023-03-24 17:12:57,545 44k INFO ====> Epoch: 146, cost 24.73 s
246
+ 2023-03-24 17:13:22,383 44k INFO ====> Epoch: 147, cost 24.84 s
247
+ 2023-03-24 17:13:47,376 44k INFO ====> Epoch: 148, cost 24.99 s
248
+ 2023-03-24 17:13:51,018 44k INFO Train Epoch: 149 [0%]
249
+ 2023-03-24 17:13:51,020 44k INFO Losses: [2.541314125061035, 2.248460292816162, 8.099034309387207, 17.675189971923828, 0.6793069243431091], step: 7400, lr: 9.816689394418209e-05
250
+ 2023-03-24 17:14:12,901 44k INFO ====> Epoch: 149, cost 25.52 s
251
+ 2023-03-24 17:14:37,615 44k INFO ====> Epoch: 150, cost 24.71 s
252
+ 2023-03-24 17:15:02,718 44k INFO ====> Epoch: 151, cost 25.10 s
253
+ 2023-03-24 17:15:27,505 44k INFO ====> Epoch: 152, cost 24.79 s
254
+ 2023-03-24 17:15:31,093 44k INFO Train Epoch: 153 [0%]
255
+ 2023-03-24 17:15:31,095 44k INFO Losses: [2.6885759830474854, 1.7870721817016602, 7.810143947601318, 17.377124786376953, 0.9760600328445435], step: 7600, lr: 9.811781969958938e-05
256
+ 2023-03-24 17:15:52,655 44k INFO ====> Epoch: 153, cost 25.15 s
257
+ 2023-03-24 17:16:17,956 44k INFO ====> Epoch: 154, cost 25.30 s
258
+ 2023-03-24 17:16:42,842 44k INFO ====> Epoch: 155, cost 24.89 s
259
+ 2023-03-24 17:17:07,750 44k INFO ====> Epoch: 156, cost 24.91 s
260
+ 2023-03-24 17:17:11,629 44k INFO Train Epoch: 157 [0%]
261
+ 2023-03-24 17:17:11,630 44k INFO Losses: [2.3176026344299316, 2.416011333465576, 11.307724952697754, 21.595233917236328, 0.9688916206359863], step: 7800, lr: 9.806876998751865e-05
262
+ 2023-03-24 17:17:33,336 44k INFO ====> Epoch: 157, cost 25.59 s
263
+ 2023-03-24 17:17:58,244 44k INFO ====> Epoch: 158, cost 24.91 s
264
+ 2023-03-24 17:18:23,111 44k INFO ====> Epoch: 159, cost 24.87 s
265
+ 2023-03-24 17:18:47,965 44k INFO ====> Epoch: 160, cost 24.85 s
266
+ 2023-03-24 17:18:51,778 44k INFO Train Epoch: 161 [0%]
267
+ 2023-03-24 17:18:51,779 44k INFO Losses: [2.335582971572876, 2.200396776199341, 11.696431159973145, 21.497562408447266, 0.9381815791130066], step: 8000, lr: 9.801974479570593e-05
268
+ 2023-03-24 17:18:56,755 44k INFO Saving model and optimizer state at iteration 161 to ./logs/44k/G_8000.pth
269
+ 2023-03-24 17:18:58,161 44k INFO Saving model and optimizer state at iteration 161 to ./logs/44k/D_8000.pth
270
+ 2023-03-24 17:19:19,776 44k INFO ====> Epoch: 161, cost 31.81 s
271
+ 2023-03-24 17:19:44,844 44k INFO ====> Epoch: 162, cost 25.07 s
272
+ 2023-03-24 17:20:09,764 44k INFO ====> Epoch: 163, cost 24.92 s
273
+ 2023-03-24 17:20:34,563 44k INFO ====> Epoch: 164, cost 24.80 s
274
+ 2023-03-24 17:20:38,302 44k INFO Train Epoch: 165 [0%]
275
+ 2023-03-24 17:20:38,303 44k INFO Losses: [2.2087090015411377, 2.4203712940216064, 10.896903991699219, 20.731460571289062, 0.6834594011306763], step: 8200, lr: 9.797074411189339e-05
276
+ 2023-03-24 17:20:59,999 44k INFO ====> Epoch: 165, cost 25.44 s
277
+ 2023-03-24 17:21:25,178 44k INFO ====> Epoch: 166, cost 25.18 s
278
+ 2023-03-24 17:21:50,298 44k INFO ====> Epoch: 167, cost 25.12 s
279
+ 2023-03-24 17:22:15,251 44k INFO ====> Epoch: 168, cost 24.95 s
280
+ 2023-03-24 17:22:18,958 44k INFO Train Epoch: 169 [0%]
281
+ 2023-03-24 17:22:18,959 44k INFO Losses: [2.3192877769470215, 2.2281627655029297, 7.640979766845703, 18.116220474243164, 0.751641571521759], step: 8400, lr: 9.792176792382932e-05
282
+ 2023-03-24 17:22:40,590 44k INFO ====> Epoch: 169, cost 25.34 s
283
+ 2023-03-24 17:23:05,450 44k INFO ====> Epoch: 170, cost 24.86 s
284
+ 2023-03-24 17:23:30,354 44k INFO ====> Epoch: 171, cost 24.90 s
285
+ 2023-03-24 17:23:55,451 44k INFO ====> Epoch: 172, cost 25.10 s
286
+ 2023-03-24 17:23:59,357 44k INFO Train Epoch: 173 [0%]
287
+ 2023-03-24 17:23:59,358 44k INFO Losses: [2.6158812046051025, 2.3054511547088623, 11.013618469238281, 19.365203857421875, 0.7469444870948792], step: 8600, lr: 9.787281621926815e-05
288
+ 2023-03-24 17:24:21,066 44k INFO ====> Epoch: 173, cost 25.61 s
289
+ 2023-03-24 17:24:46,070 44k INFO ====> Epoch: 174, cost 25.00 s
290
+ 2023-03-24 17:25:11,054 44k INFO ====> Epoch: 175, cost 24.98 s
291
+ 2023-03-24 17:25:36,049 44k INFO ====> Epoch: 176, cost 25.00 s
292
+ 2023-03-24 17:25:39,843 44k INFO Train Epoch: 177 [0%]
293
+ 2023-03-24 17:25:39,844 44k INFO Losses: [2.4385979175567627, 2.112964153289795, 10.049574851989746, 18.21015739440918, 0.4370778203010559], step: 8800, lr: 9.782388898597041e-05
294
+ 2023-03-24 17:25:44,627 44k INFO Saving model and optimizer state at iteration 177 to ./logs/44k/G_8800.pth
295
+ 2023-03-24 17:25:46,174 44k INFO Saving model and optimizer state at iteration 177 to ./logs/44k/D_8800.pth
296
+ 2023-03-24 17:26:07,968 44k INFO ====> Epoch: 177, cost 31.92 s
297
+ 2023-03-24 17:26:32,811 44k INFO ====> Epoch: 178, cost 24.84 s
298
+ 2023-03-24 17:26:57,954 44k INFO ====> Epoch: 179, cost 25.14 s
299
+ 2023-03-24 17:27:22,976 44k INFO ====> Epoch: 180, cost 25.02 s
300
+ 2023-03-24 17:27:26,840 44k INFO Train Epoch: 181 [0%]
301
+ 2023-03-24 17:27:26,841 44k INFO Losses: [2.577077865600586, 1.9521126747131348, 6.810126304626465, 15.925239562988281, 0.7762280702590942], step: 9000, lr: 9.777498621170277e-05
302
+ 2023-03-24 17:27:48,572 44k INFO ====> Epoch: 181, cost 25.60 s
303
+ 2023-03-24 17:28:13,664 44k INFO ====> Epoch: 182, cost 25.09 s
304
+ 2023-03-24 17:28:38,409 44k INFO ====> Epoch: 183, cost 24.74 s
305
+ 2023-03-24 17:29:03,357 44k INFO ====> Epoch: 184, cost 24.95 s
306
+ 2023-03-24 17:29:07,154 44k INFO Train Epoch: 185 [0%]
307
+ 2023-03-24 17:29:07,155 44k INFO Losses: [2.735002040863037, 2.042769193649292, 6.685863971710205, 18.602622985839844, 0.7408661246299744], step: 9200, lr: 9.772610788423802e-05
308
+ 2023-03-24 17:29:29,064 44k INFO ====> Epoch: 185, cost 25.71 s
309
+ 2023-03-24 17:29:54,104 44k INFO ====> Epoch: 186, cost 25.04 s
310
+ 2023-03-24 17:30:18,964 44k INFO ====> Epoch: 187, cost 24.86 s
311
+ 2023-03-24 17:30:44,025 44k INFO ====> Epoch: 188, cost 25.06 s
312
+ 2023-03-24 17:30:47,877 44k INFO Train Epoch: 189 [0%]
313
+ 2023-03-24 17:30:47,879 44k INFO Losses: [2.4279744625091553, 2.3245105743408203, 9.516021728515625, 18.581796646118164, 0.8581266403198242], step: 9400, lr: 9.767725399135504e-05
314
+ 2023-03-24 17:31:09,751 44k INFO ====> Epoch: 189, cost 25.73 s
315
+ 2023-03-24 17:31:34,675 44k INFO ====> Epoch: 190, cost 24.92 s
316
+ 2023-03-24 17:31:59,606 44k INFO ====> Epoch: 191, cost 24.93 s
317
+ 2023-03-24 17:32:24,562 44k INFO ====> Epoch: 192, cost 24.96 s
318
+ 2023-03-24 17:32:28,382 44k INFO Train Epoch: 193 [0%]
319
+ 2023-03-24 17:32:28,383 44k INFO Losses: [2.2496345043182373, 2.273651123046875, 10.993385314941406, 18.61309051513672, 0.5576750636100769], step: 9600, lr: 9.762842452083883e-05
320
+ 2023-03-24 17:32:33,323 44k INFO Saving model and optimizer state at iteration 193 to ./logs/44k/G_9600.pth
321
+ 2023-03-24 17:32:34,749 44k INFO Saving model and optimizer state at iteration 193 to ./logs/44k/D_9600.pth
322
+ 2023-03-24 17:32:56,434 44k INFO ====> Epoch: 193, cost 31.87 s
323
+ 2023-03-24 17:33:21,551 44k INFO ====> Epoch: 194, cost 25.12 s
324
+ 2023-03-24 17:33:46,348 44k INFO ====> Epoch: 195, cost 24.80 s
325
+ 2023-03-24 17:34:11,094 44k INFO ====> Epoch: 196, cost 24.75 s
326
+ 2023-03-24 17:34:15,062 44k INFO Train Epoch: 197 [0%]
327
+ 2023-03-24 17:34:15,064 44k INFO Losses: [2.400710105895996, 2.40057373046875, 8.645580291748047, 18.94593048095703, 0.579620897769928], step: 9800, lr: 9.757961946048049e-05
328
+ 2023-03-24 17:34:36,942 44k INFO ====> Epoch: 197, cost 25.85 s
329
+ 2023-03-24 17:35:01,995 44k INFO ====> Epoch: 198, cost 25.05 s
330
+ 2023-03-24 17:35:26,534 44k INFO ====> Epoch: 199, cost 24.54 s
331
+ 2023-03-24 17:35:51,658 44k INFO ====> Epoch: 200, cost 25.12 s
332
+ 2023-03-24 17:35:55,306 44k INFO Train Epoch: 201 [0%]
333
+ 2023-03-24 17:35:55,307 44k INFO Losses: [2.255695104598999, 2.3569161891937256, 11.471579551696777, 21.66715431213379, 0.7947320938110352], step: 10000, lr: 9.753083879807726e-05
334
+ 2023-03-24 17:36:16,899 44k INFO ====> Epoch: 201, cost 25.24 s
335
+ 2023-03-24 17:36:41,707 44k INFO ====> Epoch: 202, cost 24.81 s
336
+ 2023-03-24 17:37:06,526 44k INFO ====> Epoch: 203, cost 24.82 s
337
+ 2023-03-24 17:37:31,076 44k INFO ====> Epoch: 204, cost 24.55 s
338
+ 2023-03-24 17:37:34,870 44k INFO Train Epoch: 205 [0%]
339
+ 2023-03-24 17:37:34,872 44k INFO Losses: [2.64894437789917, 2.250037670135498, 7.354074001312256, 16.611326217651367, 0.5304523706436157], step: 10200, lr: 9.748208252143241e-05
340
+ 2023-03-24 17:37:56,565 44k INFO ====> Epoch: 205, cost 25.49 s
341
+ 2023-03-24 17:38:21,446 44k INFO ====> Epoch: 206, cost 24.88 s
342
+ 2023-03-24 17:38:46,504 44k INFO ====> Epoch: 207, cost 25.06 s
343
+ 2023-03-24 17:39:11,285 44k INFO ====> Epoch: 208, cost 24.78 s
344
+ 2023-03-24 17:39:15,102 44k INFO Train Epoch: 209 [0%]
345
+ 2023-03-24 17:39:15,103 44k INFO Losses: [2.1992905139923096, 2.505424976348877, 9.921688079833984, 21.01839828491211, 1.1307833194732666], step: 10400, lr: 9.743335061835535e-05
346
+ 2023-03-24 17:39:20,006 44k INFO Saving model and optimizer state at iteration 209 to ./logs/44k/G_10400.pth
347
+ 2023-03-24 17:39:21,392 44k INFO Saving model and optimizer state at iteration 209 to ./logs/44k/D_10400.pth
348
+ 2023-03-24 17:39:43,026 44k INFO ====> Epoch: 209, cost 31.74 s
349
+ 2023-03-24 17:40:08,180 44k INFO ====> Epoch: 210, cost 25.15 s
350
+ 2023-03-24 17:40:33,240 44k INFO ====> Epoch: 211, cost 25.06 s
351
+ 2023-03-24 17:40:58,174 44k INFO ====> Epoch: 212, cost 24.93 s
352
+ 2023-03-24 17:41:02,009 44k INFO Train Epoch: 213 [0%]
353
+ 2023-03-24 17:41:02,010 44k INFO Losses: [2.2102227210998535, 2.369741678237915, 9.036264419555664, 17.1883602142334, 0.4644804894924164], step: 10600, lr: 9.73846430766616e-05
354
+ 2023-03-24 17:41:23,687 44k INFO ====> Epoch: 213, cost 25.51 s
355
+ 2023-03-24 17:41:48,589 44k INFO ====> Epoch: 214, cost 24.90 s
356
+ 2023-03-24 17:42:13,292 44k INFO ====> Epoch: 215, cost 24.70 s
357
+ 2023-03-24 17:42:38,126 44k INFO ====> Epoch: 216, cost 24.83 s
358
+ 2023-03-24 17:42:42,082 44k INFO Train Epoch: 217 [0%]
359
+ 2023-03-24 17:42:42,084 44k INFO Losses: [2.442573070526123, 2.2122700214385986, 8.983601570129395, 17.680198669433594, 0.7081395387649536], step: 10800, lr: 9.733595988417275e-05
360
+ 2023-03-24 17:43:03,592 44k INFO ====> Epoch: 217, cost 25.47 s
361
+ 2023-03-24 17:43:28,367 44k INFO ====> Epoch: 218, cost 24.78 s
362
+ 2023-03-24 17:43:53,025 44k INFO ====> Epoch: 219, cost 24.66 s
363
+ 2023-03-24 17:44:17,974 44k INFO ====> Epoch: 220, cost 24.95 s
364
+ 2023-03-24 17:44:21,653 44k INFO Train Epoch: 221 [0%]
365
+ 2023-03-24 17:44:21,655 44k INFO Losses: [2.626898765563965, 1.8628339767456055, 10.394402503967285, 15.498502731323242, 0.8859735131263733], step: 11000, lr: 9.728730102871649e-05
366
+ 2023-03-24 17:44:43,321 44k INFO ====> Epoch: 221, cost 25.35 s
367
+ 2023-03-24 17:45:08,220 44k INFO ====> Epoch: 222, cost 24.90 s
368
+ 2023-03-24 17:45:33,125 44k INFO ====> Epoch: 223, cost 24.91 s
369
+ 2023-03-24 17:45:57,806 44k INFO ====> Epoch: 224, cost 24.68 s
370
+ 2023-03-24 17:46:01,570 44k INFO Train Epoch: 225 [0%]
371
+ 2023-03-24 17:46:01,571 44k INFO Losses: [2.50315260887146, 2.425022840499878, 10.251396179199219, 19.030975341796875, 0.6795037984848022], step: 11200, lr: 9.723866649812655e-05
372
+ 2023-03-24 17:46:06,371 44k INFO Saving model and optimizer state at iteration 225 to ./logs/44k/G_11200.pth
373
+ 2023-03-24 17:46:07,942 44k INFO Saving model and optimizer state at iteration 225 to ./logs/44k/D_11200.pth
374
+ 2023-03-24 17:46:29,740 44k INFO ====> Epoch: 225, cost 31.93 s
375
+ 2023-03-24 17:46:54,684 44k INFO ====> Epoch: 226, cost 24.94 s
376
+ 2023-03-24 17:47:19,823 44k INFO ====> Epoch: 227, cost 25.14 s
377
+ 2023-03-24 17:47:44,505 44k INFO ====> Epoch: 228, cost 24.68 s
378
+ 2023-03-24 17:47:48,475 44k INFO Train Epoch: 229 [0%]
379
+ 2023-03-24 17:47:48,476 44k INFO Losses: [2.6375670433044434, 2.0211079120635986, 8.814630508422852, 19.634490966796875, 0.7629879117012024], step: 11400, lr: 9.719005628024282e-05
380
+ 2023-03-24 17:48:09,961 44k INFO ====> Epoch: 229, cost 25.46 s
381
+ 2023-03-24 17:48:35,039 44k INFO ====> Epoch: 230, cost 25.08 s
382
+ 2023-03-24 17:48:59,974 44k INFO ====> Epoch: 231, cost 24.94 s
383
+ 2023-03-24 17:49:24,771 44k INFO ====> Epoch: 232, cost 24.80 s
384
+ 2023-03-24 17:49:28,612 44k INFO Train Epoch: 233 [0%]
385
+ 2023-03-24 17:49:28,613 44k INFO Losses: [2.6957833766937256, 2.0368285179138184, 6.790450096130371, 16.876953125, 0.8912892937660217], step: 11600, lr: 9.714147036291117e-05
386
+ 2023-03-24 17:49:50,306 44k INFO ====> Epoch: 233, cost 25.53 s
387
+ 2023-03-24 17:50:15,336 44k INFO ====> Epoch: 234, cost 25.03 s
388
+ 2023-03-24 17:50:40,138 44k INFO ====> Epoch: 235, cost 24.80 s
389
+ 2023-03-24 17:51:04,920 44k INFO ====> Epoch: 236, cost 24.78 s
390
+ 2023-03-24 17:51:08,682 44k INFO Train Epoch: 237 [0%]
391
+ 2023-03-24 17:51:08,683 44k INFO Losses: [2.3348302841186523, 2.5667877197265625, 7.6211042404174805, 15.862165451049805, 0.9230325818061829], step: 11800, lr: 9.709290873398365e-05
392
+ 2023-03-24 17:51:30,569 44k INFO ====> Epoch: 237, cost 25.65 s
393
+ 2023-03-24 17:51:55,319 44k INFO ====> Epoch: 238, cost 24.75 s
394
+ 2023-03-24 17:52:20,313 44k INFO ====> Epoch: 239, cost 24.99 s
395
+ 2023-03-24 17:52:45,066 44k INFO ====> Epoch: 240, cost 24.75 s
396
+ 2023-03-24 17:52:48,739 44k INFO Train Epoch: 241 [0%]
397
+ 2023-03-24 17:52:48,741 44k INFO Losses: [2.040863513946533, 2.403212547302246, 12.783172607421875, 20.934587478637695, 0.7837891578674316], step: 12000, lr: 9.704437138131832e-05
398
+ 2023-03-24 17:52:53,772 44k INFO Saving model and optimizer state at iteration 241 to ./logs/44k/G_12000.pth
399
+ 2023-03-24 17:52:55,180 44k INFO Saving model and optimizer state at iteration 241 to ./logs/44k/D_12000.pth
400
+ 2023-03-24 17:53:16,867 44k INFO ====> Epoch: 241, cost 31.80 s
401
+ 2023-03-24 17:53:41,966 44k INFO ====> Epoch: 242, cost 25.10 s
402
+ 2023-03-24 17:54:06,816 44k INFO ====> Epoch: 243, cost 24.85 s
403
+ 2023-03-24 17:54:31,779 44k INFO ====> Epoch: 244, cost 24.96 s
404
+ 2023-03-24 17:54:35,451 44k INFO Train Epoch: 245 [0%]
405
+ 2023-03-24 17:54:35,453 44k INFO Losses: [2.5395002365112305, 2.109894037246704, 11.065472602844238, 20.373876571655273, 0.8642053604125977], step: 12200, lr: 9.699585829277933e-05
406
+ 2023-03-24 17:54:57,236 44k INFO ====> Epoch: 245, cost 25.46 s
407
+ 2023-03-24 17:55:22,256 44k INFO ====> Epoch: 246, cost 25.02 s
408
+ 2023-03-24 17:55:47,152 44k INFO ====> Epoch: 247, cost 24.90 s
409
+ 2023-03-24 17:56:12,120 44k INFO ====> Epoch: 248, cost 24.97 s
410
+ 2023-03-24 17:56:15,904 44k INFO Train Epoch: 249 [0%]
411
+ 2023-03-24 17:56:15,906 44k INFO Losses: [2.2659645080566406, 2.8718552589416504, 13.755086898803711, 21.769315719604492, 0.3315553367137909], step: 12400, lr: 9.694736945623688e-05
412
+ 2023-03-24 17:56:37,475 44k INFO ====> Epoch: 249, cost 25.36 s
413
+ 2023-03-24 17:57:02,506 44k INFO ====> Epoch: 250, cost 25.03 s
414
+ 2023-03-24 17:57:27,327 44k INFO ====> Epoch: 251, cost 24.82 s
415
+ 2023-03-24 17:57:52,472 44k INFO ====> Epoch: 252, cost 25.14 s
416
+ 2023-03-24 17:57:56,307 44k INFO Train Epoch: 253 [0%]
417
+ 2023-03-24 17:57:56,308 44k INFO Losses: [2.3076179027557373, 2.63967227935791, 9.53144645690918, 20.38763427734375, 0.9321761727333069], step: 12600, lr: 9.689890485956725e-05
418
+ 2023-03-24 17:58:18,242 44k INFO ====> Epoch: 253, cost 25.77 s
419
+ 2023-03-24 17:58:43,283 44k INFO ====> Epoch: 254, cost 25.04 s
420
+ 2023-03-24 17:59:08,313 44k INFO ====> Epoch: 255, cost 25.03 s
421
+ 2023-03-24 17:59:32,978 44k INFO ====> Epoch: 256, cost 24.66 s
422
+ 2023-03-24 17:59:36,789 44k INFO Train Epoch: 257 [0%]
423
+ 2023-03-24 17:59:36,790 44k INFO Losses: [2.601497173309326, 2.15515398979187, 10.487128257751465, 20.16332244873047, 0.6738835573196411], step: 12800, lr: 9.685046449065278e-05
424
+ 2023-03-24 17:59:41,642 44k INFO Saving model and optimizer state at iteration 257 to ./logs/44k/G_12800.pth
425
+ 2023-03-24 17:59:43,224 44k INFO Saving model and optimizer state at iteration 257 to ./logs/44k/D_12800.pth
426
+ 2023-03-24 18:00:05,012 44k INFO ====> Epoch: 257, cost 32.03 s
427
+ 2023-03-24 18:00:29,964 44k INFO ====> Epoch: 258, cost 24.95 s
428
+ 2023-03-24 18:00:55,127 44k INFO ====> Epoch: 259, cost 25.16 s
429
+ 2023-03-24 18:01:19,839 44k INFO ====> Epoch: 260, cost 24.71 s
430
+ 2023-03-24 18:01:23,558 44k INFO Train Epoch: 261 [0%]
431
+ 2023-03-24 18:01:23,559 44k INFO Losses: [2.4871768951416016, 2.240135908126831, 7.533797740936279, 16.42740821838379, 0.8754367828369141], step: 13000, lr: 9.680204833738185e-05
so-vits-svc-4.0/20230324-JuewaNS-44k/best/D_13600.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54660cd48de970b4509401f732850612d08184a8105bcf5049c697efcd923603
3
+ size 561098185
so-vits-svc-4.0/20230324-JuewaNS-44k/best/G_13600.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33b3974ffd10354d567fd4c89a306ad9b3e4654f3da4436954fbff7ab45f76e3
3
+ size 542789405
so-vits-svc-4.0/20230324-JuewaNS-44k/best/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "JuewaNS": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230324-JuewaNS-44k/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "JuewaNS": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230324-JuewaNS-44k/kmeans_10000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9601e2be6d3e4ab9adf65ba0b6146e2feff47cc4b058091735c6e474d49df682
3
+ size 15450169
so-vits-svc-4.0/20230324-JuewaNS-44k/pruned/P_G_13600.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ce3ffbbd7ddf1657d95761e1238f6d465afbacdb16be23c70ccfc02cba55e0f
3
+ size 180885829
so-vits-svc-4.0/20230324-JuewaNS-44k/pruned/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 800,
5
+ "seed": 1234,
6
+ "epochs": 10000,
7
+ "learning_rate": 0.0001,
8
+ "betas": [
9
+ 0.8,
10
+ 0.99
11
+ ],
12
+ "eps": 1e-09,
13
+ "batch_size": 6,
14
+ "fp16_run": false,
15
+ "lr_decay": 0.999875,
16
+ "segment_size": 10240,
17
+ "init_lr_ratio": 1,
18
+ "warmup_epochs": 0,
19
+ "c_mel": 45,
20
+ "c_kl": 1.0,
21
+ "use_sr": true,
22
+ "max_speclen": 512,
23
+ "port": "8001",
24
+ "keep_ckpts": 99
25
+ },
26
+ "data": {
27
+ "training_files": "filelists/train.txt",
28
+ "validation_files": "filelists/val.txt",
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 44100,
31
+ "filter_length": 2048,
32
+ "hop_length": 512,
33
+ "win_length": 2048,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0.0,
36
+ "mel_fmax": 22050
37
+ },
38
+ "model": {
39
+ "inter_channels": 192,
40
+ "hidden_channels": 192,
41
+ "filter_channels": 768,
42
+ "n_heads": 2,
43
+ "n_layers": 6,
44
+ "kernel_size": 3,
45
+ "p_dropout": 0.1,
46
+ "resblock": "1",
47
+ "resblock_kernel_sizes": [
48
+ 3,
49
+ 7,
50
+ 11
51
+ ],
52
+ "resblock_dilation_sizes": [
53
+ [
54
+ 1,
55
+ 3,
56
+ 5
57
+ ],
58
+ [
59
+ 1,
60
+ 3,
61
+ 5
62
+ ],
63
+ [
64
+ 1,
65
+ 3,
66
+ 5
67
+ ]
68
+ ],
69
+ "upsample_rates": [
70
+ 8,
71
+ 8,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "upsample_initial_channel": 512,
77
+ "upsample_kernel_sizes": [
78
+ 16,
79
+ 16,
80
+ 4,
81
+ 4,
82
+ 4
83
+ ],
84
+ "n_layers_q": 3,
85
+ "use_spectral_norm": false,
86
+ "gin_channels": 256,
87
+ "ssl_dim": 256,
88
+ "n_speakers": 200
89
+ },
90
+ "spk": {
91
+ "JuewaNS": 0
92
+ }
93
+ }
so-vits-svc-4.0/20230324-JuewaNS-44k/train.log ADDED
@@ -0,0 +1,772 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-03-24 18:19:36,643 44k INFO {'train': {'log_interval': 200, 'eval_interval': 800, 'seed': 1234, 'epochs': 10000, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 6, 'fp16_run': False, 'lr_decay': 0.999875, 'segment_size': 10240, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0, 'use_sr': True, 'max_speclen': 512, 'port': '8001', 'keep_ckpts': 99}, 'data': {'training_files': 'filelists/train.txt', 'validation_files': 'filelists/val.txt', 'max_wav_value': 32768.0, 'sampling_rate': 44100, 'filter_length': 2048, 'hop_length': 512, 'win_length': 2048, 'n_mel_channels': 80, 'mel_fmin': 0.0, 'mel_fmax': 22050}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.1, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [8, 8, 2, 2, 2], 'upsample_initial_channel': 512, 'upsample_kernel_sizes': [16, 16, 4, 4, 4], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 256, 'ssl_dim': 256, 'n_speakers': 200}, 'spk': {'JuewaNS': 0}, 'model_dir': './logs/44k'}
2
+ 2023-03-24 18:19:36,643 44k WARNING /root/so-vits-svc-4.0 is not a git repository, therefore hash value comparison will be ignored.
3
+ 2023-03-24 18:19:39,837 44k INFO emb_g.weight is not in the checkpoint
4
+ 2023-03-24 18:19:39,907 44k INFO Loaded checkpoint './logs/44k/G_0.pth' (iteration 0)
5
+ 2023-03-24 18:19:40,077 44k INFO Loaded checkpoint './logs/44k/D_0.pth' (iteration 0)
6
+ 2023-03-24 18:19:47,634 44k INFO Train Epoch: 1 [0%]
7
+ 2023-03-24 18:19:47,635 44k INFO Losses: [2.8651366233825684, 2.346381187438965, 12.116447448730469, 39.413856506347656, 3.4299936294555664], step: 0, lr: 0.0001
8
+ 2023-03-24 18:19:52,952 44k INFO Saving model and optimizer state at iteration 1 to ./logs/44k/G_0.pth
9
+ 2023-03-24 18:19:54,318 44k INFO Saving model and optimizer state at iteration 1 to ./logs/44k/D_0.pth
10
+ 2023-03-24 18:20:09,532 44k INFO ====> Epoch: 1, cost 32.89 s
11
+ 2023-03-24 18:20:24,387 44k INFO ====> Epoch: 2, cost 14.85 s
12
+ 2023-03-24 18:20:39,544 44k INFO ====> Epoch: 3, cost 15.16 s
13
+ 2023-03-24 18:20:54,488 44k INFO ====> Epoch: 4, cost 14.94 s
14
+ 2023-03-24 18:21:09,454 44k INFO ====> Epoch: 5, cost 14.97 s
15
+ 2023-03-24 18:21:24,345 44k INFO ====> Epoch: 6, cost 14.89 s
16
+ 2023-03-24 18:21:39,167 44k INFO ====> Epoch: 7, cost 14.82 s
17
+ 2023-03-24 18:21:44,333 44k INFO Train Epoch: 8 [14%]
18
+ 2023-03-24 18:21:44,335 44k INFO Losses: [2.4412031173706055, 2.5478155612945557, 16.34337043762207, 25.992536544799805, 1.6750681400299072], step: 200, lr: 9.991253280566489e-05
19
+ 2023-03-24 18:21:54,640 44k INFO ====> Epoch: 8, cost 15.47 s
20
+ 2023-03-24 18:22:09,402 44k INFO ====> Epoch: 9, cost 14.76 s
21
+ 2023-03-24 18:22:24,340 44k INFO ====> Epoch: 10, cost 14.94 s
22
+ 2023-03-24 18:22:39,146 44k INFO ====> Epoch: 11, cost 14.81 s
23
+ 2023-03-24 18:22:53,975 44k INFO ====> Epoch: 12, cost 14.83 s
24
+ 2023-03-24 18:23:08,849 44k INFO ====> Epoch: 13, cost 14.87 s
25
+ 2023-03-24 18:23:23,609 44k INFO ====> Epoch: 14, cost 14.76 s
26
+ 2023-03-24 18:23:30,509 44k INFO Train Epoch: 15 [29%]
27
+ 2023-03-24 18:23:30,511 44k INFO Losses: [2.853314161300659, 1.969393253326416, 10.559226036071777, 23.596506118774414, 1.1688786745071411], step: 400, lr: 9.982514211643064e-05
28
+ 2023-03-24 18:23:39,135 44k INFO ====> Epoch: 15, cost 15.53 s
29
+ 2023-03-24 18:23:54,294 44k INFO ====> Epoch: 16, cost 15.16 s
30
+ 2023-03-24 18:24:09,139 44k INFO ====> Epoch: 17, cost 14.85 s
31
+ 2023-03-24 18:24:23,809 44k INFO ====> Epoch: 18, cost 14.67 s
32
+ 2023-03-24 18:24:38,578 44k INFO ====> Epoch: 19, cost 14.77 s
33
+ 2023-03-24 18:24:53,441 44k INFO ====> Epoch: 20, cost 14.86 s
34
+ 2023-03-24 18:25:08,355 44k INFO ====> Epoch: 21, cost 14.91 s
35
+ 2023-03-24 18:25:16,721 44k INFO Train Epoch: 22 [43%]
36
+ 2023-03-24 18:25:16,722 44k INFO Losses: [2.245793342590332, 2.530930995941162, 14.867792129516602, 24.568418502807617, 1.0560626983642578], step: 600, lr: 9.973782786538036e-05
37
+ 2023-03-24 18:25:23,991 44k INFO ====> Epoch: 22, cost 15.64 s
38
+ 2023-03-24 18:25:40,833 44k INFO ====> Epoch: 23, cost 16.84 s
39
+ 2023-03-24 18:25:55,605 44k INFO ====> Epoch: 24, cost 14.77 s
40
+ 2023-03-24 18:26:10,502 44k INFO ====> Epoch: 25, cost 14.90 s
41
+ 2023-03-24 18:26:25,600 44k INFO ====> Epoch: 26, cost 15.10 s
42
+ 2023-03-24 18:26:40,526 44k INFO ====> Epoch: 27, cost 14.93 s
43
+ 2023-03-24 18:26:55,340 44k INFO ====> Epoch: 28, cost 14.81 s
44
+ 2023-03-24 18:27:05,180 44k INFO Train Epoch: 29 [57%]
45
+ 2023-03-24 18:27:05,182 44k INFO Losses: [2.613215923309326, 2.27390718460083, 8.487275123596191, 20.264320373535156, 1.1122970581054688], step: 800, lr: 9.965058998565574e-05
46
+ 2023-03-24 18:27:09,722 44k INFO Saving model and optimizer state at iteration 29 to ./logs/44k/G_800.pth
47
+ 2023-03-24 18:27:11,038 44k INFO Saving model and optimizer state at iteration 29 to ./logs/44k/D_800.pth
48
+ 2023-03-24 18:27:16,521 44k INFO ====> Epoch: 29, cost 21.18 s
49
+ 2023-03-24 18:27:31,451 44k INFO ====> Epoch: 30, cost 14.93 s
50
+ 2023-03-24 18:27:46,263 44k INFO ====> Epoch: 31, cost 14.81 s
51
+ 2023-03-24 18:28:01,205 44k INFO ====> Epoch: 32, cost 14.94 s
52
+ 2023-03-24 18:28:15,968 44k INFO ====> Epoch: 33, cost 14.76 s
53
+ 2023-03-24 18:28:30,790 44k INFO ====> Epoch: 34, cost 14.82 s
54
+ 2023-03-24 18:28:46,345 44k INFO ====> Epoch: 35, cost 15.55 s
55
+ 2023-03-24 18:28:57,976 44k INFO Train Epoch: 36 [71%]
56
+ 2023-03-24 18:28:57,977 44k INFO Losses: [2.115177869796753, 2.6007447242736816, 13.032262802124023, 26.101831436157227, 0.6284221410751343], step: 1000, lr: 9.956342841045691e-05
57
+ 2023-03-24 18:29:01,888 44k INFO ====> Epoch: 36, cost 15.54 s
58
+ 2023-03-24 18:29:16,747 44k INFO ====> Epoch: 37, cost 14.86 s
59
+ 2023-03-24 18:29:31,673 44k INFO ====> Epoch: 38, cost 14.93 s
60
+ 2023-03-24 18:29:46,543 44k INFO ====> Epoch: 39, cost 14.87 s
61
+ 2023-03-24 18:30:01,582 44k INFO ====> Epoch: 40, cost 15.04 s
62
+ 2023-03-24 18:30:16,391 44k INFO ====> Epoch: 41, cost 14.81 s
63
+ 2023-03-24 18:30:31,370 44k INFO ====> Epoch: 42, cost 14.98 s
64
+ 2023-03-24 18:30:45,071 44k INFO Train Epoch: 43 [86%]
65
+ 2023-03-24 18:30:45,072 44k INFO Losses: [2.658895969390869, 2.289456605911255, 4.936427593231201, 20.44791030883789, 0.47966423630714417], step: 1200, lr: 9.947634307304244e-05
66
+ 2023-03-24 18:30:47,487 44k INFO ====> Epoch: 43, cost 16.12 s
67
+ 2023-03-24 18:31:02,415 44k INFO ====> Epoch: 44, cost 14.93 s
68
+ 2023-03-24 18:31:17,262 44k INFO ====> Epoch: 45, cost 14.85 s
69
+ 2023-03-24 18:31:31,991 44k INFO ====> Epoch: 46, cost 14.73 s
70
+ 2023-03-24 18:31:46,701 44k INFO ====> Epoch: 47, cost 14.71 s
71
+ 2023-03-24 18:32:01,612 44k INFO ====> Epoch: 48, cost 14.91 s
72
+ 2023-03-24 18:32:17,899 44k INFO ====> Epoch: 49, cost 16.29 s
73
+ 2023-03-24 18:32:32,832 44k INFO ====> Epoch: 50, cost 14.93 s
74
+ 2023-03-24 18:32:36,346 44k INFO Train Epoch: 51 [0%]
75
+ 2023-03-24 18:32:36,347 44k INFO Losses: [2.707481861114502, 2.2357685565948486, 10.195556640625, 20.226890563964844, 0.812960147857666], step: 1400, lr: 9.937691023999092e-05
76
+ 2023-03-24 18:32:48,230 44k INFO ====> Epoch: 51, cost 15.40 s
77
+ 2023-03-24 18:33:02,841 44k INFO ====> Epoch: 52, cost 14.61 s
78
+ 2023-03-24 18:33:17,559 44k INFO ====> Epoch: 53, cost 14.72 s
79
+ 2023-03-24 18:33:32,818 44k INFO ====> Epoch: 54, cost 15.26 s
80
+ 2023-03-24 18:33:47,747 44k INFO ====> Epoch: 55, cost 14.93 s
81
+ 2023-03-24 18:34:02,348 44k INFO ====> Epoch: 56, cost 14.60 s
82
+ 2023-03-24 18:34:17,404 44k INFO ====> Epoch: 57, cost 15.06 s
83
+ 2023-03-24 18:34:22,685 44k INFO Train Epoch: 58 [14%]
84
+ 2023-03-24 18:34:22,686 44k INFO Losses: [2.843576192855835, 2.228142738342285, 14.787569046020508, 23.943769454956055, 0.993866503238678], step: 1600, lr: 9.928998804478705e-05
85
+ 2023-03-24 18:34:27,079 44k INFO Saving model and optimizer state at iteration 58 to ./logs/44k/G_1600.pth
86
+ 2023-03-24 18:34:28,258 44k INFO Saving model and optimizer state at iteration 58 to ./logs/44k/D_1600.pth
87
+ 2023-03-24 18:34:38,568 44k INFO ====> Epoch: 58, cost 21.16 s
88
+ 2023-03-24 18:34:55,234 44k INFO ====> Epoch: 59, cost 16.67 s
89
+ 2023-03-24 18:35:09,982 44k INFO ====> Epoch: 60, cost 14.75 s
90
+ 2023-03-24 18:35:24,679 44k INFO ====> Epoch: 61, cost 14.70 s
91
+ 2023-03-24 18:35:39,348 44k INFO ====> Epoch: 62, cost 14.67 s
92
+ 2023-03-24 18:35:54,070 44k INFO ====> Epoch: 63, cost 14.72 s
93
+ 2023-03-24 18:36:10,703 44k INFO ====> Epoch: 64, cost 16.63 s
94
+ 2023-03-24 18:36:17,521 44k INFO Train Epoch: 65 [29%]
95
+ 2023-03-24 18:36:17,522 44k INFO Losses: [2.328249931335449, 2.4285080432891846, 12.13663387298584, 19.89910125732422, 1.2783076763153076], step: 1800, lr: 9.92031418779886e-05
96
+ 2023-03-24 18:36:26,007 44k INFO ====> Epoch: 65, cost 15.30 s
97
+ 2023-03-24 18:36:40,756 44k INFO ====> Epoch: 66, cost 14.75 s
98
+ 2023-03-24 18:36:55,574 44k INFO ====> Epoch: 67, cost 14.82 s
99
+ 2023-03-24 18:37:10,496 44k INFO ====> Epoch: 68, cost 14.92 s
100
+ 2023-03-24 18:37:25,888 44k INFO ====> Epoch: 69, cost 15.39 s
101
+ 2023-03-24 18:37:40,640 44k INFO ====> Epoch: 70, cost 14.75 s
102
+ 2023-03-24 18:37:55,697 44k INFO ====> Epoch: 71, cost 15.06 s
103
+ 2023-03-24 18:38:04,052 44k INFO Train Epoch: 72 [43%]
104
+ 2023-03-24 18:38:04,053 44k INFO Losses: [2.5356943607330322, 2.3216590881347656, 12.424395561218262, 23.910093307495117, 0.887087345123291], step: 2000, lr: 9.911637167309565e-05
105
+ 2023-03-24 18:38:11,120 44k INFO ====> Epoch: 72, cost 15.42 s
106
+ 2023-03-24 18:38:26,070 44k INFO ====> Epoch: 73, cost 14.95 s
107
+ 2023-03-24 18:38:41,212 44k INFO ====> Epoch: 74, cost 15.14 s
108
+ 2023-03-24 18:38:57,820 44k INFO ====> Epoch: 75, cost 16.61 s
109
+ 2023-03-24 18:39:12,632 44k INFO ====> Epoch: 76, cost 14.81 s
110
+ 2023-03-24 18:39:29,340 44k INFO ====> Epoch: 77, cost 16.71 s
111
+ 2023-03-24 18:39:44,102 44k INFO ====> Epoch: 78, cost 14.76 s
112
+ 2023-03-24 18:39:55,219 44k INFO Train Epoch: 79 [57%]
113
+ 2023-03-24 18:39:55,220 44k INFO Losses: [2.525271415710449, 2.084883689880371, 9.368517875671387, 17.8083438873291, 0.8916102647781372], step: 2200, lr: 9.902967736366644e-05
114
+ 2023-03-24 18:40:00,683 44k INFO ====> Epoch: 79, cost 16.58 s
115
+ 2023-03-24 18:40:15,489 44k INFO ====> Epoch: 80, cost 14.81 s
116
+ 2023-03-24 18:40:30,311 44k INFO ====> Epoch: 81, cost 14.82 s
117
+ 2023-03-24 18:40:45,505 44k INFO ====> Epoch: 82, cost 15.19 s
118
+ 2023-03-24 18:41:00,536 44k INFO ====> Epoch: 83, cost 15.03 s
119
+ 2023-03-24 18:41:16,174 44k INFO ====> Epoch: 84, cost 15.64 s
120
+ 2023-03-24 18:41:30,993 44k INFO ====> Epoch: 85, cost 14.82 s
121
+ 2023-03-24 18:41:42,384 44k INFO Train Epoch: 86 [71%]
122
+ 2023-03-24 18:41:42,385 44k INFO Losses: [2.3191444873809814, 2.3854219913482666, 11.815009117126465, 19.916088104248047, 0.8946883082389832], step: 2400, lr: 9.894305888331732e-05
123
+ 2023-03-24 18:41:46,894 44k INFO Saving model and optimizer state at iteration 86 to ./logs/44k/G_2400.pth
124
+ 2023-03-24 18:41:48,085 44k INFO Saving model and optimizer state at iteration 86 to ./logs/44k/D_2400.pth
125
+ 2023-03-24 18:41:52,207 44k INFO ====> Epoch: 86, cost 21.21 s
126
+ 2023-03-24 18:42:07,069 44k INFO ====> Epoch: 87, cost 14.86 s
127
+ 2023-03-24 18:42:21,905 44k INFO ====> Epoch: 88, cost 14.84 s
128
+ 2023-03-24 18:42:36,734 44k INFO ====> Epoch: 89, cost 14.83 s
129
+ 2023-03-24 18:42:51,437 44k INFO ====> Epoch: 90, cost 14.70 s
130
+ 2023-03-24 18:43:06,191 44k INFO ====> Epoch: 91, cost 14.75 s
131
+ 2023-03-24 18:43:21,034 44k INFO ====> Epoch: 92, cost 14.84 s
132
+ 2023-03-24 18:43:34,239 44k INFO Train Epoch: 93 [86%]
133
+ 2023-03-24 18:43:34,240 44k INFO Losses: [2.467679023742676, 2.421851634979248, 10.510832786560059, 21.985858917236328, 1.062198281288147], step: 2600, lr: 9.885651616572276e-05
134
+ 2023-03-24 18:43:36,795 44k INFO ====> Epoch: 93, cost 15.76 s
135
+ 2023-03-24 18:43:51,767 44k INFO ====> Epoch: 94, cost 14.97 s
136
+ 2023-03-24 18:44:06,700 44k INFO ====> Epoch: 95, cost 14.93 s
137
+ 2023-03-24 18:44:21,270 44k INFO ====> Epoch: 96, cost 14.57 s
138
+ 2023-03-24 18:44:36,401 44k INFO ====> Epoch: 97, cost 15.13 s
139
+ 2023-03-24 18:44:53,093 44k INFO ====> Epoch: 98, cost 16.69 s
140
+ 2023-03-24 18:45:07,890 44k INFO ====> Epoch: 99, cost 14.80 s
141
+ 2023-03-24 18:45:22,480 44k INFO ====> Epoch: 100, cost 14.59 s
142
+ 2023-03-24 18:45:26,087 44k INFO Train Epoch: 101 [0%]
143
+ 2023-03-24 18:45:26,088 44k INFO Losses: [2.4023709297180176, 2.3098299503326416, 11.050838470458984, 23.734569549560547, 0.7466572523117065], step: 2800, lr: 9.875770288847208e-05
144
+ 2023-03-24 18:45:37,857 44k INFO ====> Epoch: 101, cost 15.38 s
145
+ 2023-03-24 18:45:52,566 44k INFO ====> Epoch: 102, cost 14.71 s
146
+ 2023-03-24 18:46:07,422 44k INFO ====> Epoch: 103, cost 14.86 s
147
+ 2023-03-24 18:46:22,264 44k INFO ====> Epoch: 104, cost 14.84 s
148
+ 2023-03-24 18:46:36,961 44k INFO ====> Epoch: 105, cost 14.70 s
149
+ 2023-03-24 18:46:51,580 44k INFO ====> Epoch: 106, cost 14.62 s
150
+ 2023-03-24 18:47:06,263 44k INFO ====> Epoch: 107, cost 14.68 s
151
+ 2023-03-24 18:47:11,409 44k INFO Train Epoch: 108 [14%]
152
+ 2023-03-24 18:47:11,411 44k INFO Losses: [2.3292808532714844, 2.2167253494262695, 10.71139907836914, 21.37546157836914, 0.8030653595924377], step: 3000, lr: 9.867132229656573e-05
153
+ 2023-03-24 18:47:21,503 44k INFO ====> Epoch: 108, cost 15.24 s
154
+ 2023-03-24 18:47:36,263 44k INFO ====> Epoch: 109, cost 14.76 s
155
+ 2023-03-24 18:47:50,996 44k INFO ====> Epoch: 110, cost 14.73 s
156
+ 2023-03-24 18:48:05,795 44k INFO ====> Epoch: 111, cost 14.80 s
157
+ 2023-03-24 18:48:20,515 44k INFO ====> Epoch: 112, cost 14.72 s
158
+ 2023-03-24 18:48:35,425 44k INFO ====> Epoch: 113, cost 14.91 s
159
+ 2023-03-24 18:48:50,175 44k INFO ====> Epoch: 114, cost 14.75 s
160
+ 2023-03-24 18:48:57,076 44k INFO Train Epoch: 115 [29%]
161
+ 2023-03-24 18:48:57,078 44k INFO Losses: [2.477935314178467, 2.528744697570801, 8.90847110748291, 22.292436599731445, 0.34426790475845337], step: 3200, lr: 9.858501725933955e-05
162
+ 2023-03-24 18:49:02,086 44k INFO Saving model and optimizer state at iteration 115 to ./logs/44k/G_3200.pth
163
+ 2023-03-24 18:49:03,349 44k INFO Saving model and optimizer state at iteration 115 to ./logs/44k/D_3200.pth
164
+ 2023-03-24 18:49:12,380 44k INFO ====> Epoch: 115, cost 22.21 s
165
+ 2023-03-24 18:49:27,120 44k INFO ====> Epoch: 116, cost 14.74 s
166
+ 2023-03-24 18:49:41,757 44k INFO ====> Epoch: 117, cost 14.64 s
167
+ 2023-03-24 18:49:56,948 44k INFO ====> Epoch: 118, cost 15.19 s
168
+ 2023-03-24 18:50:11,603 44k INFO ====> Epoch: 119, cost 14.66 s
169
+ 2023-03-24 18:50:26,404 44k INFO ====> Epoch: 120, cost 14.80 s
170
+ 2023-03-24 18:50:41,127 44k INFO ====> Epoch: 121, cost 14.72 s
171
+ 2023-03-24 18:50:49,536 44k INFO Train Epoch: 122 [43%]
172
+ 2023-03-24 18:50:49,537 44k INFO Losses: [2.4078025817871094, 2.5715935230255127, 8.99583911895752, 17.1897029876709, 1.0879757404327393], step: 3400, lr: 9.8498787710708e-05
173
+ 2023-03-24 18:50:56,712 44k INFO ====> Epoch: 122, cost 15.58 s
174
+ 2023-03-24 18:51:11,449 44k INFO ====> Epoch: 123, cost 14.74 s
175
+ 2023-03-24 18:51:26,066 44k INFO ====> Epoch: 124, cost 14.62 s
176
+ 2023-03-24 18:51:40,977 44k INFO ====> Epoch: 125, cost 14.91 s
177
+ 2023-03-24 18:51:56,070 44k INFO ====> Epoch: 126, cost 15.09 s
178
+ 2023-03-24 18:52:10,818 44k INFO ====> Epoch: 127, cost 14.75 s
179
+ 2023-03-24 18:52:25,534 44k INFO ====> Epoch: 128, cost 14.72 s
180
+ 2023-03-24 18:52:35,476 44k INFO Train Epoch: 129 [57%]
181
+ 2023-03-24 18:52:35,477 44k INFO Losses: [2.5406367778778076, 2.3820292949676514, 9.032157897949219, 18.73415756225586, 0.952826201915741], step: 3600, lr: 9.841263358464336e-05
182
+ 2023-03-24 18:52:40,946 44k INFO ====> Epoch: 129, cost 15.41 s
183
+ 2023-03-24 18:52:55,845 44k INFO ====> Epoch: 130, cost 14.90 s
184
+ 2023-03-24 18:53:11,000 44k INFO ====> Epoch: 131, cost 15.16 s
185
+ 2023-03-24 18:53:25,870 44k INFO ====> Epoch: 132, cost 14.87 s
186
+ 2023-03-24 18:53:40,771 44k INFO ====> Epoch: 133, cost 14.90 s
187
+ 2023-03-24 18:53:55,438 44k INFO ====> Epoch: 134, cost 14.67 s
188
+ 2023-03-24 18:54:10,252 44k INFO ====> Epoch: 135, cost 14.81 s
189
+ 2023-03-24 18:54:21,658 44k INFO Train Epoch: 136 [71%]
190
+ 2023-03-24 18:54:21,659 44k INFO Losses: [2.567310333251953, 2.164691686630249, 10.801031112670898, 24.128572463989258, 0.7646280527114868], step: 3800, lr: 9.832655481517557e-05
191
+ 2023-03-24 18:54:25,608 44k INFO ====> Epoch: 136, cost 15.36 s
192
+ 2023-03-24 18:54:40,505 44k INFO ====> Epoch: 137, cost 14.90 s
193
+ 2023-03-24 18:54:55,428 44k INFO ====> Epoch: 138, cost 14.92 s
194
+ 2023-03-24 18:55:10,583 44k INFO ====> Epoch: 139, cost 15.16 s
195
+ 2023-03-24 18:55:25,273 44k INFO ====> Epoch: 140, cost 14.69 s
196
+ 2023-03-24 18:55:39,865 44k INFO ====> Epoch: 141, cost 14.59 s
197
+ 2023-03-24 18:55:54,556 44k INFO ====> Epoch: 142, cost 14.69 s
198
+ 2023-03-24 18:56:07,672 44k INFO Train Epoch: 143 [86%]
199
+ 2023-03-24 18:56:07,673 44k INFO Losses: [2.4808027744293213, 2.3600611686706543, 9.740200996398926, 22.769962310791016, 0.9508404731750488], step: 4000, lr: 9.824055133639235e-05
200
+ 2023-03-24 18:56:12,373 44k INFO Saving model and optimizer state at iteration 143 to ./logs/44k/G_4000.pth
201
+ 2023-03-24 18:56:13,752 44k INFO Saving model and optimizer state at iteration 143 to ./logs/44k/D_4000.pth
202
+ 2023-03-24 18:56:16,196 44k INFO ====> Epoch: 143, cost 21.64 s
203
+ 2023-03-24 18:56:31,092 44k INFO ====> Epoch: 144, cost 14.90 s
204
+ 2023-03-24 18:56:46,318 44k INFO ====> Epoch: 145, cost 15.23 s
205
+ 2023-03-24 18:57:01,107 44k INFO ====> Epoch: 146, cost 14.79 s
206
+ 2023-03-24 18:57:15,686 44k INFO ====> Epoch: 147, cost 14.58 s
207
+ 2023-03-24 18:57:30,896 44k INFO ====> Epoch: 148, cost 15.21 s
208
+ 2023-03-24 18:57:45,509 44k INFO ====> Epoch: 149, cost 14.61 s
209
+ 2023-03-24 18:58:00,186 44k INFO ====> Epoch: 150, cost 14.68 s
210
+ 2023-03-24 18:58:03,715 44k INFO Train Epoch: 151 [0%]
211
+ 2023-03-24 18:58:03,716 44k INFO Losses: [2.453944683074951, 2.4591901302337646, 8.528477668762207, 19.629995346069336, 1.1228570938110352], step: 4200, lr: 9.814235375455375e-05
212
+ 2023-03-24 18:58:15,478 44k INFO ====> Epoch: 151, cost 15.29 s
213
+ 2023-03-24 18:58:30,175 44k INFO ====> Epoch: 152, cost 14.70 s
214
+ 2023-03-24 18:58:44,994 44k INFO ====> Epoch: 153, cost 14.82 s
215
+ 2023-03-24 18:58:59,953 44k INFO ====> Epoch: 154, cost 14.96 s
216
+ 2023-03-24 18:59:14,793 44k INFO ====> Epoch: 155, cost 14.84 s
217
+ 2023-03-24 18:59:29,507 44k INFO ====> Epoch: 156, cost 14.71 s
218
+ 2023-03-24 18:59:44,303 44k INFO ====> Epoch: 157, cost 14.80 s
219
+ 2023-03-24 18:59:49,734 44k INFO Train Epoch: 158 [14%]
220
+ 2023-03-24 18:59:49,735 44k INFO Losses: [2.514904260635376, 2.1705262660980225, 9.76323413848877, 18.711702346801758, 0.7594791650772095], step: 4400, lr: 9.80565113912702e-05
221
+ 2023-03-24 18:59:59,801 44k INFO ====> Epoch: 158, cost 15.50 s
222
+ 2023-03-24 19:00:14,906 44k INFO ====> Epoch: 159, cost 15.10 s
223
+ 2023-03-24 19:00:29,873 44k INFO ====> Epoch: 160, cost 14.97 s
224
+ 2023-03-24 19:00:44,776 44k INFO ====> Epoch: 161, cost 14.90 s
225
+ 2023-03-24 19:00:59,727 44k INFO ====> Epoch: 162, cost 14.95 s
226
+ 2023-03-24 19:01:14,734 44k INFO ====> Epoch: 163, cost 15.01 s
227
+ 2023-03-24 19:01:29,720 44k INFO ====> Epoch: 164, cost 14.99 s
228
+ 2023-03-24 19:01:36,500 44k INFO Train Epoch: 165 [29%]
229
+ 2023-03-24 19:01:36,501 44k INFO Losses: [2.6813385486602783, 2.1468124389648438, 7.211049556732178, 16.662872314453125, 0.9376119375228882], step: 4600, lr: 9.797074411189339e-05
230
+ 2023-03-24 19:01:45,150 44k INFO ====> Epoch: 165, cost 15.43 s
231
+ 2023-03-24 19:02:00,181 44k INFO ====> Epoch: 166, cost 15.03 s
232
+ 2023-03-24 19:02:15,094 44k INFO ====> Epoch: 167, cost 14.91 s
233
+ 2023-03-24 19:02:30,176 44k INFO ====> Epoch: 168, cost 15.08 s
234
+ 2023-03-24 19:02:45,178 44k INFO ====> Epoch: 169, cost 15.00 s
235
+ 2023-03-24 19:03:00,258 44k INFO ====> Epoch: 170, cost 15.08 s
236
+ 2023-03-24 19:03:15,267 44k INFO ====> Epoch: 171, cost 15.01 s
237
+ 2023-03-24 19:03:23,715 44k INFO Train Epoch: 172 [43%]
238
+ 2023-03-24 19:03:23,716 44k INFO Losses: [2.2882304191589355, 2.3772926330566406, 9.894192695617676, 22.153350830078125, 0.46218857169151306], step: 4800, lr: 9.78850518507495e-05
239
+ 2023-03-24 19:03:28,574 44k INFO Saving model and optimizer state at iteration 172 to ./logs/44k/G_4800.pth
240
+ 2023-03-24 19:03:29,794 44k INFO Saving model and optimizer state at iteration 172 to ./logs/44k/D_4800.pth
241
+ 2023-03-24 19:03:36,948 44k INFO ====> Epoch: 172, cost 21.68 s
242
+ 2023-03-24 19:03:52,073 44k INFO ====> Epoch: 173, cost 15.12 s
243
+ 2023-03-24 19:04:06,911 44k INFO ====> Epoch: 174, cost 14.84 s
244
+ 2023-03-24 19:04:21,629 44k INFO ====> Epoch: 175, cost 14.72 s
245
+ 2023-03-24 19:04:36,595 44k INFO ====> Epoch: 176, cost 14.97 s
246
+ 2023-03-24 19:04:51,655 44k INFO ====> Epoch: 177, cost 15.06 s
247
+ 2023-03-24 19:05:06,846 44k INFO ====> Epoch: 178, cost 15.19 s
248
+ 2023-03-24 19:05:16,796 44k INFO Train Epoch: 179 [57%]
249
+ 2023-03-24 19:05:16,798 44k INFO Losses: [1.827094554901123, 2.9733214378356934, 12.0966215133667, 21.781890869140625, 0.7995937466621399], step: 5000, lr: 9.779943454222217e-05
250
+ 2023-03-24 19:05:22,243 44k INFO ====> Epoch: 179, cost 15.40 s
251
+ 2023-03-24 19:05:37,206 44k INFO ====> Epoch: 180, cost 14.96 s
252
+ 2023-03-24 19:05:52,118 44k INFO ====> Epoch: 181, cost 14.91 s
253
+ 2023-03-24 19:06:06,759 44k INFO ====> Epoch: 182, cost 14.64 s
254
+ 2023-03-24 19:06:21,816 44k INFO ====> Epoch: 183, cost 15.06 s
255
+ 2023-03-24 19:06:36,774 44k INFO ====> Epoch: 184, cost 14.96 s
256
+ 2023-03-24 19:06:51,756 44k INFO ====> Epoch: 185, cost 14.98 s
257
+ 2023-03-24 19:07:03,944 44k INFO Train Epoch: 186 [71%]
258
+ 2023-03-24 19:07:03,946 44k INFO Losses: [2.455173969268799, 2.5180628299713135, 8.574335098266602, 22.821983337402344, 0.8425252437591553], step: 5200, lr: 9.771389212075249e-05
259
+ 2023-03-24 19:07:07,998 44k INFO ====> Epoch: 186, cost 16.24 s
260
+ 2023-03-24 19:07:23,004 44k INFO ====> Epoch: 187, cost 15.01 s
261
+ 2023-03-24 19:07:37,921 44k INFO ====> Epoch: 188, cost 14.92 s
262
+ 2023-03-24 19:07:52,959 44k INFO ====> Epoch: 189, cost 15.04 s
263
+ 2023-03-24 19:08:08,037 44k INFO ====> Epoch: 190, cost 15.08 s
264
+ 2023-03-24 19:08:22,948 44k INFO ====> Epoch: 191, cost 14.91 s
265
+ 2023-03-24 19:08:38,306 44k INFO ====> Epoch: 192, cost 15.36 s
266
+ 2023-03-24 19:08:51,376 44k INFO Train Epoch: 193 [86%]
267
+ 2023-03-24 19:08:51,377 44k INFO Losses: [2.3449130058288574, 2.547933578491211, 11.487386703491211, 22.50403594970703, 0.6760184168815613], step: 5400, lr: 9.762842452083883e-05
268
+ 2023-03-24 19:08:53,739 44k INFO ====> Epoch: 193, cost 15.43 s
269
+ 2023-03-24 19:09:08,709 44k INFO ====> Epoch: 194, cost 14.97 s
270
+ 2023-03-24 19:09:23,456 44k INFO ====> Epoch: 195, cost 14.75 s
271
+ 2023-03-24 19:09:38,275 44k INFO ====> Epoch: 196, cost 14.82 s
272
+ 2023-03-24 19:09:53,149 44k INFO ====> Epoch: 197, cost 14.87 s
273
+ 2023-03-24 19:10:07,952 44k INFO ====> Epoch: 198, cost 14.80 s
274
+ 2023-03-24 19:10:22,701 44k INFO ====> Epoch: 199, cost 14.75 s
275
+ 2023-03-24 19:10:37,606 44k INFO ====> Epoch: 200, cost 14.91 s
276
+ 2023-03-24 19:10:41,351 44k INFO Train Epoch: 201 [0%]
277
+ 2023-03-24 19:10:41,352 44k INFO Losses: [2.192190647125244, 2.452057123184204, 14.672394752502441, 22.408723831176758, 0.7568256855010986], step: 5600, lr: 9.753083879807726e-05
278
+ 2023-03-24 19:10:46,121 44k INFO Saving model and optimizer state at iteration 201 to ./logs/44k/G_5600.pth
279
+ 2023-03-24 19:10:47,341 44k INFO Saving model and optimizer state at iteration 201 to ./logs/44k/D_5600.pth
280
+ 2023-03-24 19:10:59,171 44k INFO ====> Epoch: 201, cost 21.56 s
281
+ 2023-03-24 19:11:14,778 44k INFO ====> Epoch: 202, cost 15.61 s
282
+ 2023-03-24 19:11:30,111 44k INFO ====> Epoch: 203, cost 15.33 s
283
+ 2023-03-24 19:11:45,174 44k INFO ====> Epoch: 204, cost 15.06 s
284
+ 2023-03-24 19:12:00,190 44k INFO ====> Epoch: 205, cost 15.02 s
285
+ 2023-03-24 19:12:16,844 44k INFO ====> Epoch: 206, cost 16.65 s
286
+ 2023-03-24 19:12:31,854 44k INFO ====> Epoch: 207, cost 15.01 s
287
+ 2023-03-24 19:12:37,138 44k INFO Train Epoch: 208 [14%]
288
+ 2023-03-24 19:12:37,139 44k INFO Losses: [2.259969472885132, 2.62406063079834, 11.825155258178711, 20.026582717895508, 0.8043803572654724], step: 5800, lr: 9.744553130976908e-05
289
+ 2023-03-24 19:12:47,441 44k INFO ====> Epoch: 208, cost 15.59 s
290
+ 2023-03-24 19:13:02,643 44k INFO ====> Epoch: 209, cost 15.20 s
291
+ 2023-03-24 19:13:17,512 44k INFO ====> Epoch: 210, cost 14.87 s
292
+ 2023-03-24 19:13:32,485 44k INFO ====> Epoch: 211, cost 14.97 s
293
+ 2023-03-24 19:13:47,340 44k INFO ====> Epoch: 212, cost 14.86 s
294
+ 2023-03-24 19:14:02,260 44k INFO ====> Epoch: 213, cost 14.92 s
295
+ 2023-03-24 19:14:17,409 44k INFO ====> Epoch: 214, cost 15.15 s
296
+ 2023-03-24 19:14:24,344 44k INFO Train Epoch: 215 [29%]
297
+ 2023-03-24 19:14:24,345 44k INFO Losses: [2.371037721633911, 2.795213222503662, 12.122397422790527, 23.133861541748047, 0.5996092557907104], step: 6000, lr: 9.736029843752747e-05
298
+ 2023-03-24 19:14:32,970 44k INFO ====> Epoch: 215, cost 15.56 s
299
+ 2023-03-24 19:14:47,969 44k INFO ====> Epoch: 216, cost 15.00 s
300
+ 2023-03-24 19:15:02,915 44k INFO ====> Epoch: 217, cost 14.95 s
301
+ 2023-03-24 19:15:17,769 44k INFO ====> Epoch: 218, cost 14.85 s
302
+ 2023-03-24 19:15:32,645 44k INFO ====> Epoch: 219, cost 14.88 s
303
+ 2023-03-24 19:15:47,617 44k INFO ====> Epoch: 220, cost 14.97 s
304
+ 2023-03-24 19:16:02,732 44k INFO ====> Epoch: 221, cost 15.11 s
305
+ 2023-03-24 19:16:11,258 44k INFO Train Epoch: 222 [43%]
306
+ 2023-03-24 19:16:11,259 44k INFO Losses: [2.0126640796661377, 2.6407198905944824, 16.501447677612305, 22.289186477661133, 0.433843731880188], step: 6200, lr: 9.727514011608789e-05
307
+ 2023-03-24 19:16:18,504 44k INFO ====> Epoch: 222, cost 15.77 s
308
+ 2023-03-24 19:16:33,328 44k INFO ====> Epoch: 223, cost 14.82 s
309
+ 2023-03-24 19:16:48,335 44k INFO ====> Epoch: 224, cost 15.01 s
310
+ 2023-03-24 19:17:05,146 44k INFO ====> Epoch: 225, cost 16.81 s
311
+ 2023-03-24 19:17:20,324 44k INFO ====> Epoch: 226, cost 15.18 s
312
+ 2023-03-24 19:17:35,399 44k INFO ====> Epoch: 227, cost 15.07 s
313
+ 2023-03-24 19:17:50,487 44k INFO ====> Epoch: 228, cost 15.09 s
314
+ 2023-03-24 19:18:00,538 44k INFO Train Epoch: 229 [57%]
315
+ 2023-03-24 19:18:00,540 44k INFO Losses: [2.4733309745788574, 2.2934553623199463, 10.714012145996094, 20.784181594848633, 0.9193792343139648], step: 6400, lr: 9.719005628024282e-05
316
+ 2023-03-24 19:18:05,406 44k INFO Saving model and optimizer state at iteration 229 to ./logs/44k/G_6400.pth
317
+ 2023-03-24 19:18:06,751 44k INFO Saving model and optimizer state at iteration 229 to ./logs/44k/D_6400.pth
318
+ 2023-03-24 19:18:12,417 44k INFO ====> Epoch: 229, cost 21.93 s
319
+ 2023-03-24 19:18:27,279 44k INFO ====> Epoch: 230, cost 14.86 s
320
+ 2023-03-24 19:18:42,506 44k INFO ====> Epoch: 231, cost 15.23 s
321
+ 2023-03-24 19:18:57,429 44k INFO ====> Epoch: 232, cost 14.92 s
322
+ 2023-03-24 19:19:12,291 44k INFO ====> Epoch: 233, cost 14.86 s
323
+ 2023-03-24 19:19:27,176 44k INFO ====> Epoch: 234, cost 14.89 s
324
+ 2023-03-24 19:19:42,112 44k INFO ====> Epoch: 235, cost 14.94 s
325
+ 2023-03-24 19:19:53,840 44k INFO Train Epoch: 236 [71%]
326
+ 2023-03-24 19:19:53,842 44k INFO Losses: [2.2217257022857666, 2.443718433380127, 9.876315116882324, 23.468894958496094, 0.6284939050674438], step: 6600, lr: 9.710504686484176e-05
327
+ 2023-03-24 19:19:57,698 44k INFO ====> Epoch: 236, cost 15.59 s
328
+ 2023-03-24 19:20:12,632 44k INFO ====> Epoch: 237, cost 14.93 s
329
+ 2023-03-24 19:20:27,643 44k INFO ====> Epoch: 238, cost 15.01 s
330
+ 2023-03-24 19:20:42,666 44k INFO ====> Epoch: 239, cost 15.02 s
331
+ 2023-03-24 19:20:57,630 44k INFO ====> Epoch: 240, cost 14.96 s
332
+ 2023-03-24 19:21:12,570 44k INFO ====> Epoch: 241, cost 14.94 s
333
+ 2023-03-24 19:21:27,365 44k INFO ====> Epoch: 242, cost 14.79 s
334
+ 2023-03-24 19:21:40,487 44k INFO Train Epoch: 243 [86%]
335
+ 2023-03-24 19:21:40,488 44k INFO Losses: [2.344350576400757, 2.5619940757751465, 9.264684677124023, 17.457204818725586, 0.9792990684509277], step: 6800, lr: 9.702011180479129e-05
336
+ 2023-03-24 19:21:42,810 44k INFO ====> Epoch: 243, cost 15.45 s
337
+ 2023-03-24 19:21:57,984 44k INFO ====> Epoch: 244, cost 15.17 s
338
+ 2023-03-24 19:22:13,342 44k INFO ====> Epoch: 245, cost 15.36 s
339
+ 2023-03-24 19:22:30,194 44k INFO ====> Epoch: 246, cost 16.85 s
340
+ 2023-03-24 19:22:45,246 44k INFO ====> Epoch: 247, cost 15.05 s
341
+ 2023-03-24 19:23:00,158 44k INFO ====> Epoch: 248, cost 14.91 s
342
+ 2023-03-24 19:23:15,630 44k INFO ====> Epoch: 249, cost 15.47 s
343
+ 2023-03-24 19:23:30,865 44k INFO ====> Epoch: 250, cost 15.24 s
344
+ 2023-03-24 19:23:34,804 44k INFO Train Epoch: 251 [0%]
345
+ 2023-03-24 19:23:34,805 44k INFO Losses: [2.6337828636169434, 2.0596632957458496, 8.9087553024292, 17.73447608947754, 0.9439815878868103], step: 7000, lr: 9.692313412867544e-05
346
+ 2023-03-24 19:23:46,586 44k INFO ====> Epoch: 251, cost 15.72 s
347
+ 2023-03-24 19:24:01,932 44k INFO ====> Epoch: 252, cost 15.35 s
348
+ 2023-03-24 19:24:16,697 44k INFO ====> Epoch: 253, cost 14.77 s
349
+ 2023-03-24 19:24:31,744 44k INFO ====> Epoch: 254, cost 15.05 s
350
+ 2023-03-24 19:24:46,613 44k INFO ====> Epoch: 255, cost 14.87 s
351
+ 2023-03-24 19:25:01,535 44k INFO ====> Epoch: 256, cost 14.92 s
352
+ 2023-03-24 19:25:16,411 44k INFO ====> Epoch: 257, cost 14.88 s
353
+ 2023-03-24 19:25:21,922 44k INFO Train Epoch: 258 [14%]
354
+ 2023-03-24 19:25:21,924 44k INFO Losses: [2.2161927223205566, 2.5849497318267822, 11.678217887878418, 19.252304077148438, 0.6322188973426819], step: 7200, lr: 9.683835818259144e-05
355
+ 2023-03-24 19:25:26,865 44k INFO Saving model and optimizer state at iteration 258 to ./logs/44k/G_7200.pth
356
+ 2023-03-24 19:25:28,030 44k INFO Saving model and optimizer state at iteration 258 to ./logs/44k/D_7200.pth
357
+ 2023-03-24 19:25:38,272 44k INFO ====> Epoch: 258, cost 21.86 s
358
+ 2023-03-24 19:25:53,319 44k INFO ====> Epoch: 259, cost 15.05 s
359
+ 2023-03-24 19:26:08,204 44k INFO ====> Epoch: 260, cost 14.88 s
360
+ 2023-03-24 19:26:23,726 44k INFO ====> Epoch: 261, cost 15.52 s
361
+ 2023-03-24 19:26:38,714 44k INFO ====> Epoch: 262, cost 14.99 s
362
+ 2023-03-24 19:26:53,504 44k INFO ====> Epoch: 263, cost 14.79 s
363
+ 2023-03-24 19:27:08,499 44k INFO ====> Epoch: 264, cost 14.99 s
364
+ 2023-03-24 19:27:15,332 44k INFO Train Epoch: 265 [29%]
365
+ 2023-03-24 19:27:15,333 44k INFO Losses: [2.4472720623016357, 2.4915881156921387, 8.705090522766113, 18.240211486816406, 1.0029183626174927], step: 7400, lr: 9.675365638764893e-05
366
+ 2023-03-24 19:27:24,059 44k INFO ====> Epoch: 265, cost 15.56 s
367
+ 2023-03-24 19:27:39,028 44k INFO ====> Epoch: 266, cost 14.97 s
368
+ 2023-03-24 19:27:53,939 44k INFO ====> Epoch: 267, cost 14.91 s
369
+ 2023-03-24 19:28:08,716 44k INFO ====> Epoch: 268, cost 14.78 s
370
+ 2023-03-24 19:28:23,528 44k INFO ====> Epoch: 269, cost 14.81 s
371
+ 2023-03-24 19:28:38,338 44k INFO ====> Epoch: 270, cost 14.81 s
372
+ 2023-03-24 19:28:53,337 44k INFO ====> Epoch: 271, cost 15.00 s
373
+ 2023-03-24 19:29:01,906 44k INFO Train Epoch: 272 [43%]
374
+ 2023-03-24 19:29:01,907 44k INFO Losses: [2.0804944038391113, 2.359567403793335, 10.937973022460938, 18.14143943786621, 0.7806757092475891], step: 7600, lr: 9.666902867899003e-05
375
+ 2023-03-24 19:29:08,965 44k INFO ====> Epoch: 272, cost 15.63 s
376
+ 2023-03-24 19:29:24,083 44k INFO ====> Epoch: 273, cost 15.12 s
377
+ 2023-03-24 19:29:39,161 44k INFO ====> Epoch: 274, cost 15.08 s
378
+ 2023-03-24 19:29:54,309 44k INFO ====> Epoch: 275, cost 15.15 s
379
+ 2023-03-24 19:30:10,333 44k INFO ====> Epoch: 276, cost 16.02 s
380
+ 2023-03-24 19:30:25,682 44k INFO ====> Epoch: 277, cost 15.35 s
381
+ 2023-03-24 19:30:40,647 44k INFO ====> Epoch: 278, cost 14.97 s
382
+ 2023-03-24 19:30:50,698 44k INFO Train Epoch: 279 [57%]
383
+ 2023-03-24 19:30:50,699 44k INFO Losses: [2.1999661922454834, 2.28269624710083, 10.674137115478516, 19.649080276489258, 0.7525098323822021], step: 7800, lr: 9.658447499181352e-05
384
+ 2023-03-24 19:30:56,292 44k INFO ====> Epoch: 279, cost 15.65 s
385
+ 2023-03-24 19:31:11,237 44k INFO ====> Epoch: 280, cost 14.94 s
386
+ 2023-03-24 19:31:26,358 44k INFO ====> Epoch: 281, cost 15.12 s
387
+ 2023-03-24 19:31:43,106 44k INFO ====> Epoch: 282, cost 16.75 s
388
+ 2023-03-24 19:31:58,007 44k INFO ====> Epoch: 283, cost 14.90 s
389
+ 2023-03-24 19:32:13,921 44k INFO ====> Epoch: 284, cost 15.91 s
390
+ 2023-03-24 19:32:28,943 44k INFO ====> Epoch: 285, cost 15.02 s
391
+ 2023-03-24 19:32:40,760 44k INFO Train Epoch: 286 [71%]
392
+ 2023-03-24 19:32:40,762 44k INFO Losses: [2.2327651977539062, 2.5476839542388916, 12.686872482299805, 24.49921417236328, 0.5556154251098633], step: 8000, lr: 9.649999526137489e-05
393
+ 2023-03-24 19:32:45,305 44k INFO Saving model and optimizer state at iteration 286 to ./logs/44k/G_8000.pth
394
+ 2023-03-24 19:32:46,623 44k INFO Saving model and optimizer state at iteration 286 to ./logs/44k/D_8000.pth
395
+ 2023-03-24 19:32:50,602 44k INFO ====> Epoch: 286, cost 21.66 s
396
+ 2023-03-24 19:33:05,551 44k INFO ====> Epoch: 287, cost 14.95 s
397
+ 2023-03-24 19:33:20,539 44k INFO ====> Epoch: 288, cost 14.99 s
398
+ 2023-03-24 19:33:35,651 44k INFO ====> Epoch: 289, cost 15.11 s
399
+ 2023-03-24 19:33:50,602 44k INFO ====> Epoch: 290, cost 14.95 s
400
+ 2023-03-24 19:34:06,275 44k INFO ====> Epoch: 291, cost 15.67 s
401
+ 2023-03-24 19:34:22,007 44k INFO ====> Epoch: 292, cost 15.73 s
402
+ 2023-03-24 19:34:35,151 44k INFO Train Epoch: 293 [86%]
403
+ 2023-03-24 19:34:35,152 44k INFO Losses: [2.3618199825286865, 2.3041739463806152, 10.38995361328125, 22.8865966796875, 0.8566780090332031], step: 8200, lr: 9.641558942298625e-05
404
+ 2023-03-24 19:34:37,465 44k INFO ====> Epoch: 293, cost 15.46 s
405
+ 2023-03-24 19:34:52,452 44k INFO ====> Epoch: 294, cost 14.99 s
406
+ 2023-03-24 19:35:07,703 44k INFO ====> Epoch: 295, cost 15.25 s
407
+ 2023-03-24 19:35:22,584 44k INFO ====> Epoch: 296, cost 14.88 s
408
+ 2023-03-24 19:35:37,541 44k INFO ====> Epoch: 297, cost 14.96 s
409
+ 2023-03-24 19:35:52,483 44k INFO ====> Epoch: 298, cost 14.94 s
410
+ 2023-03-24 19:36:07,363 44k INFO ====> Epoch: 299, cost 14.88 s
411
+ 2023-03-24 19:36:22,355 44k INFO ====> Epoch: 300, cost 14.99 s
412
+ 2023-03-24 19:36:26,061 44k INFO Train Epoch: 301 [0%]
413
+ 2023-03-24 19:36:26,063 44k INFO Losses: [2.2221317291259766, 2.401972770690918, 8.367659568786621, 17.427534103393555, 0.6027737259864807], step: 8400, lr: 9.631921600483981e-05
414
+ 2023-03-24 19:36:37,917 44k INFO ====> Epoch: 301, cost 15.56 s
415
+ 2023-03-24 19:36:52,952 44k INFO ====> Epoch: 302, cost 15.03 s
416
+ 2023-03-24 19:37:08,118 44k INFO ====> Epoch: 303, cost 15.17 s
417
+ 2023-03-24 19:37:23,732 44k INFO ====> Epoch: 304, cost 15.61 s
418
+ 2023-03-24 19:37:38,791 44k INFO ====> Epoch: 305, cost 15.06 s
419
+ 2023-03-24 19:37:53,968 44k INFO ====> Epoch: 306, cost 15.18 s
420
+ 2023-03-24 19:38:09,012 44k INFO ====> Epoch: 307, cost 15.04 s
421
+ 2023-03-24 19:38:14,255 44k INFO Train Epoch: 308 [14%]
422
+ 2023-03-24 19:38:14,256 44k INFO Losses: [2.4681646823883057, 2.6585354804992676, 11.240384101867676, 21.601070404052734, 0.6029421091079712], step: 8600, lr: 9.62349682889948e-05
423
+ 2023-03-24 19:38:24,415 44k INFO ====> Epoch: 308, cost 15.40 s
424
+ 2023-03-24 19:38:39,411 44k INFO ====> Epoch: 309, cost 15.00 s
425
+ 2023-03-24 19:38:54,178 44k INFO ====> Epoch: 310, cost 14.77 s
426
+ 2023-03-24 19:39:09,389 44k INFO ====> Epoch: 311, cost 15.21 s
427
+ 2023-03-24 19:39:25,427 44k INFO ====> Epoch: 312, cost 16.04 s
428
+ 2023-03-24 19:39:41,537 44k INFO ====> Epoch: 313, cost 16.11 s
429
+ 2023-03-24 19:39:56,927 44k INFO ====> Epoch: 314, cost 15.39 s
430
+ 2023-03-24 19:40:03,872 44k INFO Train Epoch: 315 [29%]
431
+ 2023-03-24 19:40:03,873 44k INFO Losses: [2.458681583404541, 2.419649839401245, 11.656747817993164, 19.785127639770508, 0.44571802020072937], step: 8800, lr: 9.615079426226314e-05
432
+ 2023-03-24 19:40:08,437 44k INFO Saving model and optimizer state at iteration 315 to ./logs/44k/G_8800.pth
433
+ 2023-03-24 19:40:09,602 44k INFO Saving model and optimizer state at iteration 315 to ./logs/44k/D_8800.pth
434
+ 2023-03-24 19:40:18,502 44k INFO ====> Epoch: 315, cost 21.57 s
435
+ 2023-03-24 19:40:33,824 44k INFO ====> Epoch: 316, cost 15.32 s
436
+ 2023-03-24 19:40:49,069 44k INFO ====> Epoch: 317, cost 15.24 s
437
+ 2023-03-24 19:41:04,213 44k INFO ====> Epoch: 318, cost 15.14 s
438
+ 2023-03-24 19:41:19,512 44k INFO ====> Epoch: 319, cost 15.30 s
439
+ 2023-03-24 19:41:33,945 44k INFO ====> Epoch: 320, cost 14.43 s
440
+ 2023-03-24 19:41:48,414 44k INFO ====> Epoch: 321, cost 14.47 s
441
+ 2023-03-24 19:41:56,657 44k INFO Train Epoch: 322 [43%]
442
+ 2023-03-24 19:41:56,658 44k INFO Losses: [2.6525814533233643, 2.5892791748046875, 9.840317726135254, 18.47782325744629, 0.5226555466651917], step: 9000, lr: 9.606669386019102e-05
443
+ 2023-03-24 19:42:03,437 44k INFO ====> Epoch: 322, cost 15.02 s
444
+ 2023-03-24 19:42:18,723 44k INFO ====> Epoch: 323, cost 15.29 s
445
+ 2023-03-24 19:42:33,408 44k INFO ====> Epoch: 324, cost 14.68 s
446
+ 2023-03-24 19:42:47,755 44k INFO ====> Epoch: 325, cost 14.35 s
447
+ 2023-03-24 19:43:04,032 44k INFO ====> Epoch: 326, cost 16.28 s
448
+ 2023-03-24 19:43:18,385 44k INFO ====> Epoch: 327, cost 14.35 s
449
+ 2023-03-24 19:43:34,615 44k INFO ====> Epoch: 328, cost 16.23 s
450
+ 2023-03-24 19:43:44,189 44k INFO Train Epoch: 329 [57%]
451
+ 2023-03-24 19:43:44,191 44k INFO Losses: [2.311946153640747, 2.3908088207244873, 10.318487167358398, 20.419300079345703, 0.5420147776603699], step: 9200, lr: 9.5982667018381e-05
452
+ 2023-03-24 19:43:49,497 44k INFO ====> Epoch: 329, cost 14.88 s
453
+ 2023-03-24 19:44:03,799 44k INFO ====> Epoch: 330, cost 14.30 s
454
+ 2023-03-24 19:44:18,860 44k INFO ====> Epoch: 331, cost 15.06 s
455
+ 2023-03-24 19:44:33,888 44k INFO ====> Epoch: 332, cost 15.03 s
456
+ 2023-03-24 19:44:49,738 44k INFO ====> Epoch: 333, cost 15.85 s
457
+ 2023-03-24 19:45:04,196 44k INFO ====> Epoch: 334, cost 14.46 s
458
+ 2023-03-24 19:45:18,582 44k INFO ====> Epoch: 335, cost 14.39 s
459
+ 2023-03-24 19:45:29,760 44k INFO Train Epoch: 336 [71%]
460
+ 2023-03-24 19:45:29,761 44k INFO Losses: [2.1289186477661133, 2.669475555419922, 12.912238121032715, 24.485408782958984, 0.7686963677406311], step: 9400, lr: 9.589871367249203e-05
461
+ 2023-03-24 19:45:33,538 44k INFO ====> Epoch: 336, cost 14.96 s
462
+ 2023-03-24 19:45:47,998 44k INFO ====> Epoch: 337, cost 14.46 s
463
+ 2023-03-24 19:46:02,426 44k INFO ====> Epoch: 338, cost 14.43 s
464
+ 2023-03-24 19:46:16,729 44k INFO ====> Epoch: 339, cost 14.30 s
465
+ 2023-03-24 19:46:31,129 44k INFO ====> Epoch: 340, cost 14.40 s
466
+ 2023-03-24 19:46:45,835 44k INFO ====> Epoch: 341, cost 14.71 s
467
+ 2023-03-24 19:47:00,347 44k INFO ====> Epoch: 342, cost 14.51 s
468
+ 2023-03-24 19:47:13,112 44k INFO Train Epoch: 343 [86%]
469
+ 2023-03-24 19:47:13,114 44k INFO Losses: [2.662191867828369, 2.280200242996216, 5.206782817840576, 16.19651222229004, 1.028341293334961], step: 9600, lr: 9.581483375823925e-05
470
+ 2023-03-24 19:47:17,473 44k INFO Saving model and optimizer state at iteration 343 to ./logs/44k/G_9600.pth
471
+ 2023-03-24 19:47:18,606 44k INFO Saving model and optimizer state at iteration 343 to ./logs/44k/D_9600.pth
472
+ 2023-03-24 19:47:20,890 44k INFO ====> Epoch: 343, cost 20.54 s
473
+ 2023-03-24 19:47:35,430 44k INFO ====> Epoch: 344, cost 14.54 s
474
+ 2023-03-24 19:47:51,623 44k INFO ====> Epoch: 345, cost 16.19 s
475
+ 2023-03-24 19:48:06,057 44k INFO ====> Epoch: 346, cost 14.43 s
476
+ 2023-03-24 19:48:20,464 44k INFO ====> Epoch: 347, cost 14.41 s
477
+ 2023-03-24 19:48:34,775 44k INFO ====> Epoch: 348, cost 14.31 s
478
+ 2023-03-24 19:48:49,433 44k INFO ====> Epoch: 349, cost 14.66 s
479
+ 2023-03-24 19:49:03,842 44k INFO ====> Epoch: 350, cost 14.41 s
480
+ 2023-03-24 19:49:07,240 44k INFO Train Epoch: 351 [0%]
481
+ 2023-03-24 19:49:07,241 44k INFO Losses: [2.543008804321289, 2.433940887451172, 9.013635635375977, 19.387483596801758, 0.9613720178604126], step: 9800, lr: 9.571906083299264e-05
482
+ 2023-03-24 19:49:18,951 44k INFO ====> Epoch: 351, cost 15.11 s
483
+ 2023-03-24 19:49:33,480 44k INFO ====> Epoch: 352, cost 14.53 s
484
+ 2023-03-24 19:49:49,878 44k INFO ====> Epoch: 353, cost 16.40 s
485
+ 2023-03-24 19:50:04,253 44k INFO ====> Epoch: 354, cost 14.37 s
486
+ 2023-03-24 19:50:18,739 44k INFO ====> Epoch: 355, cost 14.49 s
487
+ 2023-03-24 19:50:33,554 44k INFO ====> Epoch: 356, cost 14.81 s
488
+ 2023-03-24 19:50:48,130 44k INFO ====> Epoch: 357, cost 14.58 s
489
+ 2023-03-24 19:50:53,068 44k INFO Train Epoch: 358 [14%]
490
+ 2023-03-24 19:50:53,069 44k INFO Losses: [2.408367156982422, 2.6140289306640625, 10.188366889953613, 17.812231063842773, 0.4619278311729431], step: 10000, lr: 9.56353380560381e-05
491
+ 2023-03-24 19:51:02,994 44k INFO ====> Epoch: 358, cost 14.86 s
492
+ 2023-03-24 19:51:18,435 44k INFO ====> Epoch: 359, cost 15.44 s
493
+ 2023-03-24 19:51:32,740 44k INFO ====> Epoch: 360, cost 14.30 s
494
+ 2023-03-24 19:51:47,268 44k INFO ====> Epoch: 361, cost 14.53 s
495
+ 2023-03-24 19:52:01,662 44k INFO ====> Epoch: 362, cost 14.39 s
496
+ 2023-03-24 19:52:16,120 44k INFO ====> Epoch: 363, cost 14.46 s
497
+ 2023-03-24 19:52:30,679 44k INFO ====> Epoch: 364, cost 14.56 s
498
+ 2023-03-24 19:52:37,280 44k INFO Train Epoch: 365 [29%]
499
+ 2023-03-24 19:52:37,282 44k INFO Losses: [2.3101587295532227, 2.3744843006134033, 8.811784744262695, 20.133663177490234, 0.5683051347732544], step: 10200, lr: 9.555168850904757e-05
500
+ 2023-03-24 19:52:45,820 44k INFO ====> Epoch: 365, cost 15.14 s
501
+ 2023-03-24 19:53:00,110 44k INFO ====> Epoch: 366, cost 14.29 s
502
+ 2023-03-24 19:53:15,540 44k INFO ====> Epoch: 367, cost 15.43 s
503
+ 2023-03-24 19:53:30,189 44k INFO ====> Epoch: 368, cost 14.65 s
504
+ 2023-03-24 19:53:44,568 44k INFO ====> Epoch: 369, cost 14.38 s
505
+ 2023-03-24 19:53:58,908 44k INFO ====> Epoch: 370, cost 14.34 s
506
+ 2023-03-24 19:54:13,226 44k INFO ====> Epoch: 371, cost 14.32 s
507
+ 2023-03-24 19:54:21,466 44k INFO Train Epoch: 372 [43%]
508
+ 2023-03-24 19:54:21,467 44k INFO Losses: [2.4247655868530273, 2.617013454437256, 7.833914756774902, 15.284951210021973, 0.39104482531547546], step: 10400, lr: 9.546811212796888e-05
509
+ 2023-03-24 19:54:25,622 44k INFO Saving model and optimizer state at iteration 372 to ./logs/44k/G_10400.pth
510
+ 2023-03-24 19:54:26,937 44k INFO Saving model and optimizer state at iteration 372 to ./logs/44k/D_10400.pth
511
+ 2023-03-24 19:54:33,855 44k INFO ====> Epoch: 372, cost 20.63 s
512
+ 2023-03-24 19:54:48,244 44k INFO ====> Epoch: 373, cost 14.39 s
513
+ 2023-03-24 19:55:04,755 44k INFO ====> Epoch: 374, cost 16.51 s
514
+ 2023-03-24 19:55:20,840 44k INFO ====> Epoch: 375, cost 16.09 s
515
+ 2023-03-24 19:55:35,407 44k INFO ====> Epoch: 376, cost 14.57 s
516
+ 2023-03-24 19:55:49,810 44k INFO ====> Epoch: 377, cost 14.40 s
517
+ 2023-03-24 19:56:04,253 44k INFO ====> Epoch: 378, cost 14.44 s
518
+ 2023-03-24 19:56:14,055 44k INFO Train Epoch: 379 [57%]
519
+ 2023-03-24 19:56:14,056 44k INFO Losses: [2.562399387359619, 2.2941503524780273, 9.387338638305664, 20.59601593017578, 0.7333645224571228], step: 10600, lr: 9.538460884880585e-05
520
+ 2023-03-24 19:56:19,432 44k INFO ====> Epoch: 379, cost 15.18 s
521
+ 2023-03-24 19:56:33,776 44k INFO ====> Epoch: 380, cost 14.34 s
522
+ 2023-03-24 19:56:48,194 44k INFO ====> Epoch: 381, cost 14.42 s
523
+ 2023-03-24 19:57:02,933 44k INFO ====> Epoch: 382, cost 14.74 s
524
+ 2023-03-24 19:57:17,422 44k INFO ====> Epoch: 383, cost 14.49 s
525
+ 2023-03-24 19:57:32,443 44k INFO ====> Epoch: 384, cost 15.02 s
526
+ 2023-03-24 19:57:48,657 44k INFO ====> Epoch: 385, cost 16.21 s
527
+ 2023-03-24 19:58:01,332 44k INFO Train Epoch: 386 [71%]
528
+ 2023-03-24 19:58:01,333 44k INFO Losses: [2.2014482021331787, 2.4676952362060547, 14.335899353027344, 22.053157806396484, 0.9596916437149048], step: 10800, lr: 9.530117860761828e-05
529
+ 2023-03-24 19:58:05,553 44k INFO ====> Epoch: 386, cost 16.90 s
530
+ 2023-03-24 19:58:20,173 44k INFO ====> Epoch: 387, cost 14.62 s
531
+ 2023-03-24 19:58:34,640 44k INFO ====> Epoch: 388, cost 14.47 s
532
+ 2023-03-24 19:58:48,960 44k INFO ====> Epoch: 389, cost 14.32 s
533
+ 2023-03-24 19:59:03,473 44k INFO ====> Epoch: 390, cost 14.51 s
534
+ 2023-03-24 19:59:17,969 44k INFO ====> Epoch: 391, cost 14.50 s
535
+ 2023-03-24 19:59:32,349 44k INFO ====> Epoch: 392, cost 14.38 s
536
+ 2023-03-24 19:59:45,153 44k INFO Train Epoch: 393 [86%]
537
+ 2023-03-24 19:59:45,154 44k INFO Losses: [2.2285561561584473, 2.763125419616699, 10.18687629699707, 21.496538162231445, 0.8079171180725098], step: 11000, lr: 9.52178213405219e-05
538
+ 2023-03-24 19:59:47,353 44k INFO ====> Epoch: 393, cost 15.00 s
539
+ 2023-03-24 20:00:01,806 44k INFO ====> Epoch: 394, cost 14.45 s
540
+ 2023-03-24 20:00:16,188 44k INFO ====> Epoch: 395, cost 14.38 s
541
+ 2023-03-24 20:00:31,245 44k INFO ====> Epoch: 396, cost 15.06 s
542
+ 2023-03-24 20:00:46,297 44k INFO ====> Epoch: 397, cost 15.05 s
543
+ 2023-03-24 20:01:01,571 44k INFO ====> Epoch: 398, cost 15.27 s
544
+ 2023-03-24 20:01:15,927 44k INFO ====> Epoch: 399, cost 14.36 s
545
+ 2023-03-24 20:01:30,224 44k INFO ====> Epoch: 400, cost 14.30 s
546
+ 2023-03-24 20:01:33,793 44k INFO Train Epoch: 401 [0%]
547
+ 2023-03-24 20:01:33,794 44k INFO Losses: [2.341519594192505, 2.1632702350616455, 10.988146781921387, 21.963043212890625, 0.8887688517570496], step: 11200, lr: 9.512264516656537e-05
548
+ 2023-03-24 20:01:38,128 44k INFO Saving model and optimizer state at iteration 401 to ./logs/44k/G_11200.pth
549
+ 2023-03-24 20:01:39,259 44k INFO Saving model and optimizer state at iteration 401 to ./logs/44k/D_11200.pth
550
+ 2023-03-24 20:01:51,386 44k INFO ====> Epoch: 401, cost 21.16 s
551
+ 2023-03-24 20:02:05,793 44k INFO ====> Epoch: 402, cost 14.41 s
552
+ 2023-03-24 20:02:20,365 44k INFO ====> Epoch: 403, cost 14.57 s
553
+ 2023-03-24 20:02:34,880 44k INFO ====> Epoch: 404, cost 14.51 s
554
+ 2023-03-24 20:02:49,330 44k INFO ====> Epoch: 405, cost 14.45 s
555
+ 2023-03-24 20:03:03,690 44k INFO ====> Epoch: 406, cost 14.36 s
556
+ 2023-03-24 20:03:18,139 44k INFO ====> Epoch: 407, cost 14.45 s
557
+ 2023-03-24 20:03:23,289 44k INFO Train Epoch: 408 [14%]
558
+ 2023-03-24 20:03:23,290 44k INFO Losses: [2.1299514770507812, 2.5966148376464844, 12.680700302124023, 19.583683013916016, 0.708603024482727], step: 11400, lr: 9.503944405766085e-05
559
+ 2023-03-24 20:03:33,270 44k INFO ====> Epoch: 408, cost 15.13 s
560
+ 2023-03-24 20:03:47,805 44k INFO ====> Epoch: 409, cost 14.54 s
561
+ 2023-03-24 20:04:02,389 44k INFO ====> Epoch: 410, cost 14.58 s
562
+ 2023-03-24 20:04:16,789 44k INFO ====> Epoch: 411, cost 14.40 s
563
+ 2023-03-24 20:04:31,317 44k INFO ====> Epoch: 412, cost 14.53 s
564
+ 2023-03-24 20:04:45,740 44k INFO ====> Epoch: 413, cost 14.42 s
565
+ 2023-03-24 20:05:00,307 44k INFO ====> Epoch: 414, cost 14.57 s
566
+ 2023-03-24 20:05:07,625 44k INFO Train Epoch: 415 [29%]
567
+ 2023-03-24 20:05:07,627 44k INFO Losses: [2.1400105953216553, 2.480010747909546, 14.507291793823242, 21.961090087890625, 1.0544546842575073], step: 11600, lr: 9.495631572243191e-05
568
+ 2023-03-24 20:05:17,591 44k INFO ====> Epoch: 415, cost 17.28 s
569
+ 2023-03-24 20:05:32,326 44k INFO ====> Epoch: 416, cost 14.73 s
570
+ 2023-03-24 20:05:46,878 44k INFO ====> Epoch: 417, cost 14.55 s
571
+ 2023-03-24 20:06:01,677 44k INFO ====> Epoch: 418, cost 14.80 s
572
+ 2023-03-24 20:06:16,322 44k INFO ====> Epoch: 419, cost 14.64 s
573
+ 2023-03-24 20:06:31,265 44k INFO ====> Epoch: 420, cost 14.94 s
574
+ 2023-03-24 20:06:46,256 44k INFO ====> Epoch: 421, cost 14.99 s
575
+ 2023-03-24 20:06:54,496 44k INFO Train Epoch: 422 [43%]
576
+ 2023-03-24 20:06:54,498 44k INFO Losses: [2.521395444869995, 2.023440361022949, 10.221491813659668, 16.555747985839844, 0.37890228629112244], step: 11800, lr: 9.487326009722552e-05
577
+ 2023-03-24 20:07:01,372 44k INFO ====> Epoch: 422, cost 15.12 s
578
+ 2023-03-24 20:07:17,538 44k INFO ====> Epoch: 423, cost 16.17 s
579
+ 2023-03-24 20:07:32,137 44k INFO ====> Epoch: 424, cost 14.60 s
580
+ 2023-03-24 20:07:46,367 44k INFO ====> Epoch: 425, cost 14.23 s
581
+ 2023-03-24 20:08:00,897 44k INFO ====> Epoch: 426, cost 14.53 s
582
+ 2023-03-24 20:08:16,044 44k INFO ====> Epoch: 427, cost 15.15 s
583
+ 2023-03-24 20:08:30,473 44k INFO ====> Epoch: 428, cost 14.43 s
584
+ 2023-03-24 20:08:41,356 44k INFO Train Epoch: 429 [57%]
585
+ 2023-03-24 20:08:41,357 44k INFO Losses: [2.484126329421997, 2.4067256450653076, 9.9367036819458, 19.814332962036133, 0.9535115361213684], step: 12000, lr: 9.479027711844423e-05
586
+ 2023-03-24 20:08:45,857 44k INFO Saving model and optimizer state at iteration 429 to ./logs/44k/G_12000.pth
587
+ 2023-03-24 20:08:46,998 44k INFO Saving model and optimizer state at iteration 429 to ./logs/44k/D_12000.pth
588
+ 2023-03-24 20:08:52,322 44k INFO ====> Epoch: 429, cost 21.85 s
589
+ 2023-03-24 20:09:06,682 44k INFO ====> Epoch: 430, cost 14.36 s
590
+ 2023-03-24 20:09:21,042 44k INFO ====> Epoch: 431, cost 14.36 s
591
+ 2023-03-24 20:09:35,541 44k INFO ====> Epoch: 432, cost 14.50 s
592
+ 2023-03-24 20:09:50,036 44k INFO ====> Epoch: 433, cost 14.50 s
593
+ 2023-03-24 20:10:04,556 44k INFO ====> Epoch: 434, cost 14.52 s
594
+ 2023-03-24 20:10:18,950 44k INFO ====> Epoch: 435, cost 14.39 s
595
+ 2023-03-24 20:10:30,293 44k INFO Train Epoch: 436 [71%]
596
+ 2023-03-24 20:10:30,294 44k INFO Losses: [2.1555025577545166, 2.3229312896728516, 7.884840488433838, 16.179161071777344, 0.5915942192077637], step: 12200, lr: 9.470736672254626e-05
597
+ 2023-03-24 20:10:34,099 44k INFO ====> Epoch: 436, cost 15.15 s
598
+ 2023-03-24 20:10:48,609 44k INFO ====> Epoch: 437, cost 14.51 s
599
+ 2023-03-24 20:11:02,959 44k INFO ====> Epoch: 438, cost 14.35 s
600
+ 2023-03-24 20:11:17,503 44k INFO ====> Epoch: 439, cost 14.54 s
601
+ 2023-03-24 20:11:32,290 44k INFO ====> Epoch: 440, cost 14.79 s
602
+ 2023-03-24 20:11:47,198 44k INFO ====> Epoch: 441, cost 14.91 s
603
+ 2023-03-24 20:12:01,679 44k INFO ====> Epoch: 442, cost 14.48 s
604
+ 2023-03-24 20:12:14,509 44k INFO Train Epoch: 443 [86%]
605
+ 2023-03-24 20:12:14,510 44k INFO Losses: [2.2263095378875732, 2.628852367401123, 10.052955627441406, 17.378814697265625, 0.6211011409759521], step: 12400, lr: 9.46245288460454e-05
606
+ 2023-03-24 20:12:16,691 44k INFO ====> Epoch: 443, cost 15.01 s
607
+ 2023-03-24 20:12:31,424 44k INFO ====> Epoch: 444, cost 14.73 s
608
+ 2023-03-24 20:12:45,819 44k INFO ====> Epoch: 445, cost 14.39 s
609
+ 2023-03-24 20:13:00,366 44k INFO ====> Epoch: 446, cost 14.55 s
610
+ 2023-03-24 20:13:15,743 44k INFO ====> Epoch: 447, cost 15.38 s
611
+ 2023-03-24 20:13:30,449 44k INFO ====> Epoch: 448, cost 14.71 s
612
+ 2023-03-24 20:13:45,006 44k INFO ====> Epoch: 449, cost 14.56 s
613
+ 2023-03-24 20:13:59,691 44k INFO ====> Epoch: 450, cost 14.69 s
614
+ 2023-03-24 20:14:03,302 44k INFO Train Epoch: 451 [0%]
615
+ 2023-03-24 20:14:03,304 44k INFO Losses: [2.2450530529022217, 2.55692720413208, 9.89914321899414, 16.104248046875, 0.7876774668693542], step: 12600, lr: 9.452994570508276e-05
616
+ 2023-03-24 20:14:15,011 44k INFO ====> Epoch: 451, cost 15.32 s
617
+ 2023-03-24 20:14:29,690 44k INFO ====> Epoch: 452, cost 14.68 s
618
+ 2023-03-24 20:14:44,367 44k INFO ====> Epoch: 453, cost 14.68 s
619
+ 2023-03-24 20:14:58,869 44k INFO ====> Epoch: 454, cost 14.50 s
620
+ 2023-03-24 20:15:13,461 44k INFO ====> Epoch: 455, cost 14.59 s
621
+ 2023-03-24 20:15:27,995 44k INFO ====> Epoch: 456, cost 14.53 s
622
+ 2023-03-24 20:15:42,342 44k INFO ====> Epoch: 457, cost 14.35 s
623
+ 2023-03-24 20:15:47,531 44k INFO Train Epoch: 458 [14%]
624
+ 2023-03-24 20:15:47,532 44k INFO Losses: [2.1057448387145996, 2.3418726921081543, 14.067486763000488, 20.251667022705078, 0.6368907690048218], step: 12800, lr: 9.4447263013768e-05
625
+ 2023-03-24 20:15:51,836 44k INFO Saving model and optimizer state at iteration 458 to ./logs/44k/G_12800.pth
626
+ 2023-03-24 20:15:53,108 44k INFO Saving model and optimizer state at iteration 458 to ./logs/44k/D_12800.pth
627
+ 2023-03-24 20:16:03,015 44k INFO ====> Epoch: 458, cost 20.67 s
628
+ 2023-03-24 20:16:17,577 44k INFO ====> Epoch: 459, cost 14.56 s
629
+ 2023-03-24 20:16:31,875 44k INFO ====> Epoch: 460, cost 14.30 s
630
+ 2023-03-24 20:16:46,440 44k INFO ====> Epoch: 461, cost 14.57 s
631
+ 2023-03-24 20:17:00,950 44k INFO ====> Epoch: 462, cost 14.51 s
632
+ 2023-03-24 20:17:15,290 44k INFO ====> Epoch: 463, cost 14.34 s
633
+ 2023-03-24 20:17:29,633 44k INFO ====> Epoch: 464, cost 14.34 s
634
+ 2023-03-24 20:17:36,274 44k INFO Train Epoch: 465 [29%]
635
+ 2023-03-24 20:17:36,275 44k INFO Losses: [2.522308349609375, 2.7243692874908447, 11.208446502685547, 20.988624572753906, 0.5713735818862915], step: 13000, lr: 9.436465264268356e-05
636
+ 2023-03-24 20:17:44,555 44k INFO ====> Epoch: 465, cost 14.92 s
637
+ 2023-03-24 20:17:59,035 44k INFO ====> Epoch: 466, cost 14.48 s
638
+ 2023-03-24 20:18:13,574 44k INFO ====> Epoch: 467, cost 14.54 s
639
+ 2023-03-24 20:18:28,138 44k INFO ====> Epoch: 468, cost 14.56 s
640
+ 2023-03-24 20:18:42,637 44k INFO ====> Epoch: 469, cost 14.50 s
641
+ 2023-03-24 20:18:57,185 44k INFO ====> Epoch: 470, cost 14.55 s
642
+ 2023-03-24 20:19:11,671 44k INFO ====> Epoch: 471, cost 14.49 s
643
+ 2023-03-24 20:19:19,879 44k INFO Train Epoch: 472 [43%]
644
+ 2023-03-24 20:19:19,880 44k INFO Losses: [2.4117305278778076, 2.0530574321746826, 8.800864219665527, 12.694659233093262, 0.7633107900619507], step: 13200, lr: 9.428211452857292e-05
645
+ 2023-03-24 20:19:26,821 44k INFO ====> Epoch: 472, cost 15.15 s
646
+ 2023-03-24 20:19:41,477 44k INFO ====> Epoch: 473, cost 14.66 s
647
+ 2023-03-24 20:19:57,261 44k INFO ====> Epoch: 474, cost 15.78 s
648
+ 2023-03-24 20:20:11,794 44k INFO ====> Epoch: 475, cost 14.53 s
649
+ 2023-03-24 20:20:26,224 44k INFO ====> Epoch: 476, cost 14.43 s
650
+ 2023-03-24 20:20:41,168 44k INFO ====> Epoch: 477, cost 14.94 s
651
+ 2023-03-24 20:20:55,667 44k INFO ====> Epoch: 478, cost 14.50 s
652
+ 2023-03-24 20:21:05,331 44k INFO Train Epoch: 479 [57%]
653
+ 2023-03-24 20:21:05,332 44k INFO Losses: [2.286837100982666, 2.894473075866699, 11.783750534057617, 20.42172622680664, 0.7042444944381714], step: 13400, lr: 9.419964860823498e-05
654
+ 2023-03-24 20:21:10,655 44k INFO ====> Epoch: 479, cost 14.99 s
655
+ 2023-03-24 20:21:25,167 44k INFO ====> Epoch: 480, cost 14.51 s
656
+ 2023-03-24 20:21:39,608 44k INFO ====> Epoch: 481, cost 14.44 s
657
+ 2023-03-24 20:21:53,961 44k INFO ====> Epoch: 482, cost 14.35 s
658
+ 2023-03-24 20:22:09,678 44k INFO ====> Epoch: 483, cost 15.72 s
659
+ 2023-03-24 20:22:24,111 44k INFO ====> Epoch: 484, cost 14.43 s
660
+ 2023-03-24 20:22:38,527 44k INFO ====> Epoch: 485, cost 14.42 s
661
+ 2023-03-24 20:22:49,768 44k INFO Train Epoch: 486 [71%]
662
+ 2023-03-24 20:22:49,769 44k INFO Losses: [2.4765379428863525, 2.358840227127075, 9.544150352478027, 17.993436813354492, 0.4492684304714203], step: 13600, lr: 9.411725481852385e-05
663
+ 2023-03-24 20:22:54,040 44k INFO Saving model and optimizer state at iteration 486 to ./logs/44k/G_13600.pth
664
+ 2023-03-24 20:22:55,162 44k INFO Saving model and optimizer state at iteration 486 to ./logs/44k/D_13600.pth
665
+ 2023-03-24 20:22:58,945 44k INFO ====> Epoch: 486, cost 20.42 s
666
+ 2023-03-24 20:23:13,211 44k INFO ====> Epoch: 487, cost 14.27 s
667
+ 2023-03-24 20:23:29,478 44k INFO ====> Epoch: 488, cost 16.27 s
668
+ 2023-03-24 20:23:43,887 44k INFO ====> Epoch: 489, cost 14.41 s
669
+ 2023-03-24 20:23:58,565 44k INFO ====> Epoch: 490, cost 14.68 s
670
+ 2023-03-24 20:24:13,027 44k INFO ====> Epoch: 491, cost 14.46 s
671
+ 2023-03-24 20:24:27,298 44k INFO ====> Epoch: 492, cost 14.27 s
672
+ 2023-03-24 20:24:40,013 44k INFO Train Epoch: 493 [86%]
673
+ 2023-03-24 20:24:40,014 44k INFO Losses: [2.4640090465545654, 2.574191093444824, 11.566976547241211, 20.15541648864746, 0.44430384039878845], step: 13800, lr: 9.403493309634886e-05
674
+ 2023-03-24 20:24:42,220 44k INFO ====> Epoch: 493, cost 14.92 s
675
+ 2023-03-24 20:24:56,688 44k INFO ====> Epoch: 494, cost 14.47 s
676
+ 2023-03-24 20:25:11,153 44k INFO ====> Epoch: 495, cost 14.47 s
677
+ 2023-03-24 20:25:25,566 44k INFO ====> Epoch: 496, cost 14.41 s
678
+ 2023-03-24 20:25:39,817 44k INFO ====> Epoch: 497, cost 14.25 s
679
+ 2023-03-24 20:25:54,155 44k INFO ====> Epoch: 498, cost 14.34 s
680
+ 2023-03-24 20:26:08,438 44k INFO ====> Epoch: 499, cost 14.28 s
681
+ 2023-03-24 20:26:22,930 44k INFO ====> Epoch: 500, cost 14.49 s
682
+ 2023-03-24 20:26:26,528 44k INFO Train Epoch: 501 [0%]
683
+ 2023-03-24 20:26:26,529 44k INFO Losses: [2.4643354415893555, 2.21096134185791, 12.568891525268555, 17.2219181060791, 0.6204159259796143], step: 14000, lr: 9.394093929325224e-05
684
+ 2023-03-24 20:26:38,142 44k INFO ====> Epoch: 501, cost 15.21 s
685
+ 2023-03-24 20:26:52,567 44k INFO ====> Epoch: 502, cost 14.43 s
686
+ 2023-03-24 20:27:07,141 44k INFO ====> Epoch: 503, cost 14.57 s
687
+ 2023-03-24 20:27:21,499 44k INFO ====> Epoch: 504, cost 14.36 s
688
+ 2023-03-24 20:27:35,990 44k INFO ====> Epoch: 505, cost 14.49 s
689
+ 2023-03-24 20:27:50,442 44k INFO ====> Epoch: 506, cost 14.45 s
690
+ 2023-03-24 20:28:04,909 44k INFO ====> Epoch: 507, cost 14.47 s
691
+ 2023-03-24 20:28:09,895 44k INFO Train Epoch: 508 [14%]
692
+ 2023-03-24 20:28:09,896 44k INFO Losses: [2.190594434738159, 2.6167070865631104, 19.042556762695312, 21.837785720825195, 0.6685327291488647], step: 14200, lr: 9.385877178932038e-05
693
+ 2023-03-24 20:28:19,888 44k INFO ====> Epoch: 508, cost 14.98 s
694
+ 2023-03-24 20:28:34,282 44k INFO ====> Epoch: 509, cost 14.39 s
695
+ 2023-03-24 20:28:48,842 44k INFO ====> Epoch: 510, cost 14.56 s
696
+ 2023-03-24 20:29:03,258 44k INFO ====> Epoch: 511, cost 14.42 s
697
+ 2023-03-24 20:29:17,509 44k INFO ====> Epoch: 512, cost 14.25 s
698
+ 2023-03-24 20:29:32,003 44k INFO ====> Epoch: 513, cost 14.49 s
699
+ 2023-03-24 20:29:46,533 44k INFO ====> Epoch: 514, cost 14.53 s
700
+ 2023-03-24 20:29:53,094 44k INFO Train Epoch: 515 [29%]
701
+ 2023-03-24 20:29:53,095 44k INFO Losses: [2.2295584678649902, 2.565598964691162, 12.276839256286621, 20.938312530517578, 0.5135537385940552], step: 14400, lr: 9.377667615499888e-05
702
+ 2023-03-24 20:29:57,584 44k INFO Saving model and optimizer state at iteration 515 to ./logs/44k/G_14400.pth
703
+ 2023-03-24 20:29:58,746 44k INFO Saving model and optimizer state at iteration 515 to ./logs/44k/D_14400.pth
704
+ 2023-03-24 20:30:07,099 44k INFO ====> Epoch: 515, cost 20.57 s
705
+ 2023-03-24 20:30:21,493 44k INFO ====> Epoch: 516, cost 14.39 s
706
+ 2023-03-24 20:30:35,841 44k INFO ====> Epoch: 517, cost 14.35 s
707
+ 2023-03-24 20:30:50,339 44k INFO ====> Epoch: 518, cost 14.50 s
708
+ 2023-03-24 20:31:04,644 44k INFO ====> Epoch: 519, cost 14.31 s
709
+ 2023-03-24 20:31:19,043 44k INFO ====> Epoch: 520, cost 14.40 s
710
+ 2023-03-24 20:31:33,479 44k INFO ====> Epoch: 521, cost 14.44 s
711
+ 2023-03-24 20:31:41,552 44k INFO Train Epoch: 522 [43%]
712
+ 2023-03-24 20:31:41,553 44k INFO Losses: [2.4139742851257324, 2.372035264968872, 8.859540939331055, 14.895163536071777, 0.356738805770874], step: 14600, lr: 9.36946523274254e-05
713
+ 2023-03-24 20:31:48,526 44k INFO ====> Epoch: 522, cost 15.05 s
714
+ 2023-03-24 20:32:03,063 44k INFO ====> Epoch: 523, cost 14.54 s
715
+ 2023-03-24 20:32:17,498 44k INFO ====> Epoch: 524, cost 14.43 s
716
+ 2023-03-24 20:32:31,840 44k INFO ====> Epoch: 525, cost 14.34 s
717
+ 2023-03-24 20:32:46,313 44k INFO ====> Epoch: 526, cost 14.47 s
718
+ 2023-03-24 20:33:02,532 44k INFO ====> Epoch: 527, cost 16.22 s
719
+ 2023-03-24 20:33:17,116 44k INFO ====> Epoch: 528, cost 14.58 s
720
+ 2023-03-24 20:33:26,788 44k INFO Train Epoch: 529 [57%]
721
+ 2023-03-24 20:33:26,789 44k INFO Losses: [2.420297861099243, 2.644667148590088, 11.601181983947754, 20.18661880493164, 0.6209089159965515], step: 14800, lr: 9.361270024379255e-05
722
+ 2023-03-24 20:33:32,068 44k INFO ====> Epoch: 529, cost 14.95 s
723
+ 2023-03-24 20:33:46,379 44k INFO ====> Epoch: 530, cost 14.31 s
724
+ 2023-03-24 20:34:00,960 44k INFO ====> Epoch: 531, cost 14.58 s
725
+ 2023-03-24 20:34:15,291 44k INFO ====> Epoch: 532, cost 14.33 s
726
+ 2023-03-24 20:34:29,582 44k INFO ====> Epoch: 533, cost 14.29 s
727
+ 2023-03-24 20:34:44,315 44k INFO ====> Epoch: 534, cost 14.73 s
728
+ 2023-03-24 20:34:58,766 44k INFO ====> Epoch: 535, cost 14.45 s
729
+ 2023-03-24 20:35:10,040 44k INFO Train Epoch: 536 [71%]
730
+ 2023-03-24 20:35:10,042 44k INFO Losses: [2.2677812576293945, 2.56514310836792, 10.31601333618164, 19.94939613342285, 0.9051366448402405], step: 15000, lr: 9.353081984134796e-05
731
+ 2023-03-24 20:35:13,887 44k INFO ====> Epoch: 536, cost 15.12 s
732
+ 2023-03-24 20:35:28,286 44k INFO ====> Epoch: 537, cost 14.40 s
733
+ 2023-03-24 20:35:42,698 44k INFO ====> Epoch: 538, cost 14.41 s
734
+ 2023-03-24 20:35:57,094 44k INFO ====> Epoch: 539, cost 14.40 s
735
+ 2023-03-24 20:36:11,458 44k INFO ====> Epoch: 540, cost 14.36 s
736
+ 2023-03-24 20:36:25,885 44k INFO ====> Epoch: 541, cost 14.43 s
737
+ 2023-03-24 20:36:40,508 44k INFO ====> Epoch: 542, cost 14.62 s
738
+ 2023-03-24 20:36:53,595 44k INFO Train Epoch: 543 [86%]
739
+ 2023-03-24 20:36:53,596 44k INFO Losses: [2.5094823837280273, 2.4331843852996826, 6.562588691711426, 18.474308013916016, 0.5865221619606018], step: 15200, lr: 9.344901105739411e-05
740
+ 2023-03-24 20:36:57,995 44k INFO Saving model and optimizer state at iteration 543 to ./logs/44k/G_15200.pth
741
+ 2023-03-24 20:36:59,235 44k INFO Saving model and optimizer state at iteration 543 to ./logs/44k/D_15200.pth
742
+ 2023-03-24 20:37:01,473 44k INFO ====> Epoch: 543, cost 20.96 s
743
+ 2023-03-24 20:37:15,632 44k INFO ====> Epoch: 544, cost 14.16 s
744
+ 2023-03-24 20:37:29,945 44k INFO ====> Epoch: 545, cost 14.31 s
745
+ 2023-03-24 20:37:44,127 44k INFO ====> Epoch: 546, cost 14.18 s
746
+ 2023-03-24 20:37:58,227 44k INFO ====> Epoch: 547, cost 14.10 s
747
+ 2023-03-24 20:38:12,354 44k INFO ====> Epoch: 548, cost 14.13 s
748
+ 2023-03-24 20:38:26,594 44k INFO ====> Epoch: 549, cost 14.24 s
749
+ 2023-03-24 20:38:41,002 44k INFO ====> Epoch: 550, cost 14.41 s
750
+ 2023-03-24 20:38:44,471 44k INFO Train Epoch: 551 [0%]
751
+ 2023-03-24 20:38:44,473 44k INFO Losses: [2.3259267807006836, 2.6285746097564697, 11.082757949829102, 20.842479705810547, 0.6062003374099731], step: 15400, lr: 9.335560292005964e-05
752
+ 2023-03-24 20:38:55,883 44k INFO ====> Epoch: 551, cost 14.88 s
753
+ 2023-03-24 20:39:10,186 44k INFO ====> Epoch: 552, cost 14.30 s
754
+ 2023-03-24 20:39:24,305 44k INFO ====> Epoch: 553, cost 14.12 s
755
+ 2023-03-24 20:39:38,479 44k INFO ====> Epoch: 554, cost 14.17 s
756
+ 2023-03-24 20:39:52,592 44k INFO ====> Epoch: 555, cost 14.11 s
757
+ 2023-03-24 20:40:08,191 44k INFO ====> Epoch: 556, cost 15.60 s
758
+ 2023-03-24 20:40:22,344 44k INFO ====> Epoch: 557, cost 14.15 s
759
+ 2023-03-24 20:40:27,327 44k INFO Train Epoch: 558 [14%]
760
+ 2023-03-24 20:40:27,328 44k INFO Losses: [2.343079090118408, 2.480746030807495, 14.084281921386719, 19.00463104248047, 0.4086940586566925], step: 15600, lr: 9.327394739343082e-05
761
+ 2023-03-24 20:40:37,225 44k INFO ====> Epoch: 558, cost 14.88 s
762
+ 2023-03-24 20:40:51,779 44k INFO ====> Epoch: 559, cost 14.55 s
763
+ 2023-03-24 20:41:05,973 44k INFO ====> Epoch: 560, cost 14.19 s
764
+ 2023-03-24 20:41:20,119 44k INFO ====> Epoch: 561, cost 14.15 s
765
+ 2023-03-24 20:41:34,410 44k INFO ====> Epoch: 562, cost 14.29 s
766
+ 2023-03-24 20:41:48,570 44k INFO ====> Epoch: 563, cost 14.16 s
767
+ 2023-03-24 20:42:02,932 44k INFO ====> Epoch: 564, cost 14.36 s
768
+ 2023-03-24 20:42:09,543 44k INFO Train Epoch: 565 [29%]
769
+ 2023-03-24 20:42:09,545 44k INFO Losses: [2.1794023513793945, 2.599058151245117, 14.124211311340332, 20.959125518798828, 0.5926719307899475], step: 15800, lr: 9.319236328860017e-05
770
+ 2023-03-24 20:42:17,885 44k INFO ====> Epoch: 565, cost 14.95 s
771
+ 2023-03-24 20:42:32,242 44k INFO ====> Epoch: 566, cost 14.36 s
772
+ 2023-03-24 20:42:46,728 44k INFO ====> Epoch: 567, cost 14.49 s