Hugging Face's logo

matijap
/
hera

Token Classification
Flair
PyTorch
German
French
Italian
Model card Files
xet
Community
matijap commited on
Commit
bf5e67a
·
1 Parent(s): 76d2ca8

Upload 3 files

Browse files
Files changed (3) hide show
  1. loss.tsv +41 -0
  2. pytorch_model.bin +3 -0
  3. training.log +1095 -0
loss.tsv ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP BAD_EPOCHS LEARNING_RATE TRAIN_LOSS TRAIN_PRECISION TRAIN_RECALL TRAIN_F1 TRAIN_ACCURACY DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 12:42:38 0 0.1000 1.2624991280691964 0.861785352230072 0.7432 0.7097 0.7261 0.5916 0.9671951532363892 0.687 0.6593 0.6729 0.5341
3
+ 2 12:43:50 0 0.1000 0.6825343231175375 0.5168215036392212 0.7308 0.8282 0.7765 0.661 0.6507071852684021 0.6635 0.7656 0.7109 0.5838
4
+ 3 12:45:03 0 0.1000 0.5183594098256386 0.3893236517906189 0.8172 0.836 0.8265 0.7337 0.4997367858886719 0.7649 0.7985 0.7814 0.6899
5
+ 4 12:46:15 0 0.1000 0.4139997055040813 0.3005659878253937 0.8938 0.8594 0.8763 0.8003 0.3956667482852936 0.8377 0.8132 0.8253 0.74
6
+ 5 12:47:27 1 0.1000 0.36818719446306397 0.24057592451572418 0.8911 0.9028 0.8969 0.8363 0.3635919690132141 0.7993 0.8168 0.808 0.7102
7
+ 6 12:48:38 2 0.1000 0.31204920145584586 0.2524365782737732 0.87 0.8696 0.8698 0.7968 0.3847086429595947 0.7949 0.7949 0.7949 0.7023
8
+ 7 12:49:50 3 0.1000 0.28333256802917633 0.21568651497364044 0.8761 0.893 0.8845 0.8179 0.3467901051044464 0.8123 0.8242 0.8182 0.7258
9
+ 8 12:51:02 4 0.1000 0.2589272867494585 0.19819645583629608 0.8969 0.8918 0.8943 0.8346 0.33811214566230774 0.8178 0.8059 0.8118 0.719
10
+ 9 12:52:14 0 0.0500 0.2208224505362426 0.14378328621387482 0.9301 0.927 0.9285 0.8794 0.2700260877609253 0.8401 0.8278 0.8339 0.7386
11
+ 10 12:53:28 0 0.0500 0.2100470625941066 0.1295960396528244 0.94 0.9319 0.936 0.8938 0.2524380087852478 0.8625 0.8498 0.8561 0.7708
12
+ 11 12:54:42 1 0.0500 0.19424312072392752 0.1299719512462616 0.9298 0.9393 0.9345 0.8907 0.2761968970298767 0.8406 0.8498 0.8452 0.7557
13
+ 12 12:55:55 0 0.0500 0.18922761222213783 0.11139164865016937 0.9456 0.9479 0.9468 0.9092 0.24872702360153198 0.8704 0.8608 0.8656 0.786
14
+ 13 12:57:06 1 0.0500 0.178593337767626 0.11495152860879898 0.94 0.9446 0.9423 0.9011 0.2527526021003723 0.8571 0.8571 0.8571 0.7697
15
+ 14 12:58:17 0 0.0500 0.17087342237277447 0.09600471705198288 0.9559 0.959 0.9574 0.9252 0.2339187115430832 0.8881 0.8718 0.8799 0.8013
16
+ 15 12:59:28 0 0.0500 0.1794916741442902 0.10160095989704132 0.9584 0.9455 0.9519 0.9158 0.22547538578510284 0.8981 0.8718 0.8848 0.8041
17
+ 16 13:00:40 1 0.0500 0.16603657197670119 0.09528940916061401 0.9628 0.9442 0.9534 0.9208 0.2436159998178482 0.8897 0.8571 0.8731 0.8014
18
+ 17 13:01:51 2 0.0500 0.16133079754530325 0.09658616781234741 0.9542 0.9475 0.9508 0.9127 0.25625115633010864 0.8731 0.8571 0.8651 0.7748
19
+ 18 13:03:04 3 0.0500 0.15205490873309382 0.07957068085670471 0.9641 0.9569 0.9605 0.9332 0.24449127912521362 0.8826 0.8535 0.8678 0.7898
20
+ 19 13:04:15 4 0.0500 0.14762112769854846 0.08018826693296432 0.9589 0.9664 0.9626 0.9342 0.2329457700252533 0.8635 0.8571 0.8603 0.7697
21
+ 20 13:05:25 0 0.0250 0.13416617515222157 0.06673520058393478 0.9676 0.9672 0.9674 0.9417 0.22465617954730988 0.8981 0.8718 0.8848 0.8068
22
+ 21 13:06:34 0 0.0250 0.13184886426111447 0.06548392027616501 0.9727 0.9647 0.9687 0.9446 0.22177472710609436 0.8985 0.8755 0.8868 0.8157
23
+ 22 13:07:46 0 0.0250 0.12659577267309013 0.06098590046167374 0.9752 0.9668 0.971 0.9477 0.21829380095005035 0.9049 0.8718 0.8881 0.8095
24
+ 23 13:08:59 0 0.0250 0.1288031041974852 0.05855342745780945 0.9769 0.9717 0.9743 0.9549 0.21743902564048767 0.9053 0.8755 0.8901 0.8129
25
+ 24 13:10:12 1 0.0250 0.1292022240736277 0.05762539058923721 0.9705 0.9705 0.9705 0.9487 0.21672436594963074 0.8947 0.8718 0.8831 0.8068
26
+ 25 13:11:23 0 0.0250 0.12257867395525147 0.05489451438188553 0.9762 0.9754 0.9758 0.957 0.20767748355865479 0.9094 0.8828 0.8959 0.8197
27
+ 26 13:12:36 1 0.0250 0.12978441588272005 0.053497862070798874 0.9691 0.977 0.9731 0.9517 0.21015514433383942 0.8848 0.8718 0.8782 0.7987
28
+ 27 13:13:49 2 0.0250 0.12584238060429653 0.05426767095923424 0.974 0.9688 0.9714 0.9494 0.21591384708881378 0.9057 0.8791 0.8922 0.8191
29
+ 28 13:15:05 0 0.0250 0.11048245014825919 0.05138213932514191 0.9753 0.9721 0.9737 0.9534 0.20661891996860504 0.9101 0.8901 0.9 0.8265
30
+ 29 13:16:20 1 0.0250 0.10748581229280035 0.04754020646214485 0.9831 0.9783 0.9807 0.9656 0.20554056763648987 0.8947 0.8718 0.8831 0.8013
31
+ 30 13:17:37 2 0.0250 0.1138213260034737 0.049051374197006226 0.9814 0.9717 0.9765 0.9591 0.21063490211963654 0.9195 0.8791 0.8989 0.8276
32
+ 31 13:18:54 3 0.0250 0.11994299481505574 0.04879188537597656 0.9822 0.9713 0.9767 0.9599 0.21518820524215698 0.9077 0.8645 0.8856 0.811
33
+ 32 13:20:08 4 0.0250 0.113609206908251 0.04602975398302078 0.9762 0.9766 0.9764 0.9593 0.2196006327867508 0.8947 0.8718 0.8831 0.8068
34
+ 33 13:21:25 1 0.0125 0.11010150409611422 0.04103841260075569 0.982 0.9815 0.9818 0.9684 0.19789864122867584 0.8981 0.8718 0.8848 0.8068
35
+ 34 13:22:43 2 0.0125 0.10113046086650729 0.04190446436405182 0.9815 0.9783 0.9799 0.9656 0.20444391667842865 0.9157 0.8755 0.8951 0.8241
36
+ 35 13:24:00 3 0.0125 0.10165476730488562 0.03953753411769867 0.9836 0.9815 0.9826 0.9704 0.20463646948337555 0.9091 0.8791 0.8939 0.8191
37
+ 36 13:25:15 4 0.0125 0.10106634041664465 0.03854582458734512 0.9832 0.9815 0.9824 0.9692 0.20157837867736816 0.9019 0.8755 0.8885 0.8074
38
+ 37 13:26:32 1 0.0063 0.09473008117933748 0.03739665448665619 0.9856 0.9828 0.9842 0.9736 0.20396985113620758 0.916 0.8791 0.8972 0.8276
39
+ 38 13:27:45 0 0.0063 0.0904515941165205 0.03672816604375839 0.9856 0.9824 0.984 0.9724 0.1981133222579956 0.9234 0.8828 0.9026 0.831
40
+ 39 13:28:59 1 0.0063 0.09554681721523668 0.03638274222612381 0.9856 0.9824 0.984 0.9724 0.20241594314575195 0.9122 0.8755 0.8935 0.8185
41
+ 40 13:30:13 2 0.0063 0.09763757313900043 0.03600259870290756 0.9852 0.9824 0.9838 0.972 0.20378927886486053 0.9091 0.8791 0.8939 0.8163
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44f76d94464e679cd6a35070b94ee5efc9fa7d8a75c445cbd751ab10e2a45bc5
3
+ size 682870485
training.log ADDED
@@ -0,0 +1,1095 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-04-05 12:41:26,937 ----------------------------------------------------------------------------------------------------
2
+ 2023-04-05 12:41:26,943 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(105880, 768)
7
+ (position_embeddings): Embedding(512, 768)
8
+ (token_type_embeddings): Embedding(2, 768)
9
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0): BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=768, out_features=768, bias=True)
18
+ (key): Linear(in_features=768, out_features=768, bias=True)
19
+ (value): Linear(in_features=768, out_features=768, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=768, out_features=768, bias=True)
24
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
34
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ (1): BertLayer(
39
+ (attention): BertAttention(
40
+ (self): BertSelfAttention(
41
+ (query): Linear(in_features=768, out_features=768, bias=True)
42
+ (key): Linear(in_features=768, out_features=768, bias=True)
43
+ (value): Linear(in_features=768, out_features=768, bias=True)
44
+ (dropout): Dropout(p=0.1, inplace=False)
45
+ )
46
+ (output): BertSelfOutput(
47
+ (dense): Linear(in_features=768, out_features=768, bias=True)
48
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
49
+ (dropout): Dropout(p=0.1, inplace=False)
50
+ )
51
+ )
52
+ (intermediate): BertIntermediate(
53
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
54
+ (intermediate_act_fn): GELUActivation()
55
+ )
56
+ (output): BertOutput(
57
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
58
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
59
+ (dropout): Dropout(p=0.1, inplace=False)
60
+ )
61
+ )
62
+ (2): BertLayer(
63
+ (attention): BertAttention(
64
+ (self): BertSelfAttention(
65
+ (query): Linear(in_features=768, out_features=768, bias=True)
66
+ (key): Linear(in_features=768, out_features=768, bias=True)
67
+ (value): Linear(in_features=768, out_features=768, bias=True)
68
+ (dropout): Dropout(p=0.1, inplace=False)
69
+ )
70
+ (output): BertSelfOutput(
71
+ (dense): Linear(in_features=768, out_features=768, bias=True)
72
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
73
+ (dropout): Dropout(p=0.1, inplace=False)
74
+ )
75
+ )
76
+ (intermediate): BertIntermediate(
77
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
78
+ (intermediate_act_fn): GELUActivation()
79
+ )
80
+ (output): BertOutput(
81
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
82
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
83
+ (dropout): Dropout(p=0.1, inplace=False)
84
+ )
85
+ )
86
+ (3): BertLayer(
87
+ (attention): BertAttention(
88
+ (self): BertSelfAttention(
89
+ (query): Linear(in_features=768, out_features=768, bias=True)
90
+ (key): Linear(in_features=768, out_features=768, bias=True)
91
+ (value): Linear(in_features=768, out_features=768, bias=True)
92
+ (dropout): Dropout(p=0.1, inplace=False)
93
+ )
94
+ (output): BertSelfOutput(
95
+ (dense): Linear(in_features=768, out_features=768, bias=True)
96
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
97
+ (dropout): Dropout(p=0.1, inplace=False)
98
+ )
99
+ )
100
+ (intermediate): BertIntermediate(
101
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
102
+ (intermediate_act_fn): GELUActivation()
103
+ )
104
+ (output): BertOutput(
105
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
106
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
107
+ (dropout): Dropout(p=0.1, inplace=False)
108
+ )
109
+ )
110
+ (4): BertLayer(
111
+ (attention): BertAttention(
112
+ (self): BertSelfAttention(
113
+ (query): Linear(in_features=768, out_features=768, bias=True)
114
+ (key): Linear(in_features=768, out_features=768, bias=True)
115
+ (value): Linear(in_features=768, out_features=768, bias=True)
116
+ (dropout): Dropout(p=0.1, inplace=False)
117
+ )
118
+ (output): BertSelfOutput(
119
+ (dense): Linear(in_features=768, out_features=768, bias=True)
120
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
121
+ (dropout): Dropout(p=0.1, inplace=False)
122
+ )
123
+ )
124
+ (intermediate): BertIntermediate(
125
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
126
+ (intermediate_act_fn): GELUActivation()
127
+ )
128
+ (output): BertOutput(
129
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
130
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
131
+ (dropout): Dropout(p=0.1, inplace=False)
132
+ )
133
+ )
134
+ (5): BertLayer(
135
+ (attention): BertAttention(
136
+ (self): BertSelfAttention(
137
+ (query): Linear(in_features=768, out_features=768, bias=True)
138
+ (key): Linear(in_features=768, out_features=768, bias=True)
139
+ (value): Linear(in_features=768, out_features=768, bias=True)
140
+ (dropout): Dropout(p=0.1, inplace=False)
141
+ )
142
+ (output): BertSelfOutput(
143
+ (dense): Linear(in_features=768, out_features=768, bias=True)
144
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
145
+ (dropout): Dropout(p=0.1, inplace=False)
146
+ )
147
+ )
148
+ (intermediate): BertIntermediate(
149
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
150
+ (intermediate_act_fn): GELUActivation()
151
+ )
152
+ (output): BertOutput(
153
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
154
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
155
+ (dropout): Dropout(p=0.1, inplace=False)
156
+ )
157
+ )
158
+ (6): BertLayer(
159
+ (attention): BertAttention(
160
+ (self): BertSelfAttention(
161
+ (query): Linear(in_features=768, out_features=768, bias=True)
162
+ (key): Linear(in_features=768, out_features=768, bias=True)
163
+ (value): Linear(in_features=768, out_features=768, bias=True)
164
+ (dropout): Dropout(p=0.1, inplace=False)
165
+ )
166
+ (output): BertSelfOutput(
167
+ (dense): Linear(in_features=768, out_features=768, bias=True)
168
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
169
+ (dropout): Dropout(p=0.1, inplace=False)
170
+ )
171
+ )
172
+ (intermediate): BertIntermediate(
173
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
174
+ (intermediate_act_fn): GELUActivation()
175
+ )
176
+ (output): BertOutput(
177
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
178
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
179
+ (dropout): Dropout(p=0.1, inplace=False)
180
+ )
181
+ )
182
+ (7): BertLayer(
183
+ (attention): BertAttention(
184
+ (self): BertSelfAttention(
185
+ (query): Linear(in_features=768, out_features=768, bias=True)
186
+ (key): Linear(in_features=768, out_features=768, bias=True)
187
+ (value): Linear(in_features=768, out_features=768, bias=True)
188
+ (dropout): Dropout(p=0.1, inplace=False)
189
+ )
190
+ (output): BertSelfOutput(
191
+ (dense): Linear(in_features=768, out_features=768, bias=True)
192
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
193
+ (dropout): Dropout(p=0.1, inplace=False)
194
+ )
195
+ )
196
+ (intermediate): BertIntermediate(
197
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
198
+ (intermediate_act_fn): GELUActivation()
199
+ )
200
+ (output): BertOutput(
201
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
202
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
203
+ (dropout): Dropout(p=0.1, inplace=False)
204
+ )
205
+ )
206
+ (8): BertLayer(
207
+ (attention): BertAttention(
208
+ (self): BertSelfAttention(
209
+ (query): Linear(in_features=768, out_features=768, bias=True)
210
+ (key): Linear(in_features=768, out_features=768, bias=True)
211
+ (value): Linear(in_features=768, out_features=768, bias=True)
212
+ (dropout): Dropout(p=0.1, inplace=False)
213
+ )
214
+ (output): BertSelfOutput(
215
+ (dense): Linear(in_features=768, out_features=768, bias=True)
216
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
217
+ (dropout): Dropout(p=0.1, inplace=False)
218
+ )
219
+ )
220
+ (intermediate): BertIntermediate(
221
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
222
+ (intermediate_act_fn): GELUActivation()
223
+ )
224
+ (output): BertOutput(
225
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
226
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
227
+ (dropout): Dropout(p=0.1, inplace=False)
228
+ )
229
+ )
230
+ (9): BertLayer(
231
+ (attention): BertAttention(
232
+ (self): BertSelfAttention(
233
+ (query): Linear(in_features=768, out_features=768, bias=True)
234
+ (key): Linear(in_features=768, out_features=768, bias=True)
235
+ (value): Linear(in_features=768, out_features=768, bias=True)
236
+ (dropout): Dropout(p=0.1, inplace=False)
237
+ )
238
+ (output): BertSelfOutput(
239
+ (dense): Linear(in_features=768, out_features=768, bias=True)
240
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
241
+ (dropout): Dropout(p=0.1, inplace=False)
242
+ )
243
+ )
244
+ (intermediate): BertIntermediate(
245
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
246
+ (intermediate_act_fn): GELUActivation()
247
+ )
248
+ (output): BertOutput(
249
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
250
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
251
+ (dropout): Dropout(p=0.1, inplace=False)
252
+ )
253
+ )
254
+ (10): BertLayer(
255
+ (attention): BertAttention(
256
+ (self): BertSelfAttention(
257
+ (query): Linear(in_features=768, out_features=768, bias=True)
258
+ (key): Linear(in_features=768, out_features=768, bias=True)
259
+ (value): Linear(in_features=768, out_features=768, bias=True)
260
+ (dropout): Dropout(p=0.1, inplace=False)
261
+ )
262
+ (output): BertSelfOutput(
263
+ (dense): Linear(in_features=768, out_features=768, bias=True)
264
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
265
+ (dropout): Dropout(p=0.1, inplace=False)
266
+ )
267
+ )
268
+ (intermediate): BertIntermediate(
269
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
270
+ (intermediate_act_fn): GELUActivation()
271
+ )
272
+ (output): BertOutput(
273
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
274
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
275
+ (dropout): Dropout(p=0.1, inplace=False)
276
+ )
277
+ )
278
+ (11): BertLayer(
279
+ (attention): BertAttention(
280
+ (self): BertSelfAttention(
281
+ (query): Linear(in_features=768, out_features=768, bias=True)
282
+ (key): Linear(in_features=768, out_features=768, bias=True)
283
+ (value): Linear(in_features=768, out_features=768, bias=True)
284
+ (dropout): Dropout(p=0.1, inplace=False)
285
+ )
286
+ (output): BertSelfOutput(
287
+ (dense): Linear(in_features=768, out_features=768, bias=True)
288
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
289
+ (dropout): Dropout(p=0.1, inplace=False)
290
+ )
291
+ )
292
+ (intermediate): BertIntermediate(
293
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
294
+ (intermediate_act_fn): GELUActivation()
295
+ )
296
+ (output): BertOutput(
297
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
298
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
299
+ (dropout): Dropout(p=0.1, inplace=False)
300
+ )
301
+ )
302
+ )
303
+ )
304
+ (pooler): BertPooler(
305
+ (dense): Linear(in_features=768, out_features=768, bias=True)
306
+ (activation): Tanh()
307
+ )
308
+ )
309
+ )
310
+ (word_dropout): WordDropout(p=0.05)
311
+ (locked_dropout): LockedDropout(p=0.5)
312
+ (embedding2nn): Linear(in_features=768, out_features=768, bias=True)
313
+ (rnn): LSTM(768, 256, batch_first=True, bidirectional=True)
314
+ (linear): Linear(in_features=512, out_features=11, bias=True)
315
+ (loss_function): ViterbiLoss()
316
+ (crf): CRF()
317
+ )"
318
+ 2023-04-05 12:41:26,945 ----------------------------------------------------------------------------------------------------
319
+ 2023-04-05 12:41:26,947 Corpus: "Corpus: 2253 train + 250 dev + 441 test sentences"
320
+ 2023-04-05 12:41:26,948 ----------------------------------------------------------------------------------------------------
321
+ 2023-04-05 12:41:26,948 Parameters:
322
+ 2023-04-05 12:41:26,949 - learning_rate: "0.100000"
323
+ 2023-04-05 12:41:26,950 - mini_batch_size: "32"
324
+ 2023-04-05 12:41:26,951 - patience: "3"
325
+ 2023-04-05 12:41:26,951 - anneal_factor: "0.5"
326
+ 2023-04-05 12:41:26,953 - max_epochs: "40"
327
+ 2023-04-05 12:41:26,954 - shuffle: "True"
328
+ 2023-04-05 12:41:26,955 - train_with_dev: "False"
329
+ 2023-04-05 12:41:26,955 - batch_growth_annealing: "False"
330
+ 2023-04-05 12:41:26,958 ----------------------------------------------------------------------------------------------------
331
+ 2023-04-05 12:41:26,958 Model training base path: "resources\taggers\addressNER"
332
+ 2023-04-05 12:41:26,959 ----------------------------------------------------------------------------------------------------
333
+ 2023-04-05 12:41:26,960 Device: cuda:0
334
+ 2023-04-05 12:41:26,961 ----------------------------------------------------------------------------------------------------
335
+ 2023-04-05 12:41:26,962 Embeddings storage mode: none
336
+ 2023-04-05 12:41:26,962 ----------------------------------------------------------------------------------------------------
337
+ 2023-04-05 12:41:30,223 epoch 1 - iter 7/71 - loss 2.28721713 - time (sec): 3.26 - samples/sec: 178.21 - lr: 0.100000
338
+ 2023-04-05 12:41:33,767 epoch 1 - iter 14/71 - loss 1.92622932 - time (sec): 6.80 - samples/sec: 171.67 - lr: 0.100000
339
+ 2023-04-05 12:41:37,177 epoch 1 - iter 21/71 - loss 1.73813053 - time (sec): 10.21 - samples/sec: 168.49 - lr: 0.100000
340
+ 2023-04-05 12:41:40,953 epoch 1 - iter 28/71 - loss 1.61830648 - time (sec): 13.99 - samples/sec: 166.33 - lr: 0.100000
341
+ 2023-04-05 12:41:44,268 epoch 1 - iter 35/71 - loss 1.53214432 - time (sec): 17.30 - samples/sec: 168.39 - lr: 0.100000
342
+ 2023-04-05 12:41:48,032 epoch 1 - iter 42/71 - loss 1.45733425 - time (sec): 21.07 - samples/sec: 167.12 - lr: 0.100000
343
+ 2023-04-05 12:41:51,701 epoch 1 - iter 49/71 - loss 1.39384449 - time (sec): 24.74 - samples/sec: 166.14 - lr: 0.100000
344
+ 2023-04-05 12:41:55,370 epoch 1 - iter 56/71 - loss 1.34539887 - time (sec): 28.41 - samples/sec: 164.71 - lr: 0.100000
345
+ 2023-04-05 12:41:59,094 epoch 1 - iter 63/71 - loss 1.30608958 - time (sec): 32.13 - samples/sec: 163.55 - lr: 0.100000
346
+ 2023-04-05 12:42:02,838 epoch 1 - iter 70/71 - loss 1.26400603 - time (sec): 35.87 - samples/sec: 163.93 - lr: 0.100000
347
+ 2023-04-05 12:42:03,627 ----------------------------------------------------------------------------------------------------
348
+ 2023-04-05 12:42:03,628 EPOCH 1 done: loss 1.2625 - lr 0.100000
349
+ 2023-04-05 12:42:34,974 Evaluating as a multi-label problem: False
350
+ 2023-04-05 12:42:34,994 TRAIN : loss 0.861785352230072 - f1-score (micro avg) 0.7261
351
+ 2023-04-05 12:42:38,596 Evaluating as a multi-label problem: False
352
+ 2023-04-05 12:42:38,610 DEV : loss 0.9671951532363892 - f1-score (micro avg) 0.6729
353
+ 2023-04-05 12:42:38,617 BAD EPOCHS (no improvement): 0
354
+ 2023-04-05 12:42:38,618 saving best model
355
+ 2023-04-05 12:42:39,745 ----------------------------------------------------------------------------------------------------
356
+ 2023-04-05 12:42:42,695 epoch 2 - iter 7/71 - loss 0.92334886 - time (sec): 2.95 - samples/sec: 202.11 - lr: 0.100000
357
+ 2023-04-05 12:42:46,576 epoch 2 - iter 14/71 - loss 0.87175200 - time (sec): 6.83 - samples/sec: 177.16 - lr: 0.100000
358
+ 2023-04-05 12:42:50,257 epoch 2 - iter 21/71 - loss 0.81372023 - time (sec): 10.51 - samples/sec: 171.63 - lr: 0.100000
359
+ 2023-04-05 12:42:53,892 epoch 2 - iter 28/71 - loss 0.78629231 - time (sec): 14.15 - samples/sec: 168.38 - lr: 0.100000
360
+ 2023-04-05 12:42:57,549 epoch 2 - iter 35/71 - loss 0.76274145 - time (sec): 17.80 - samples/sec: 166.10 - lr: 0.100000
361
+ 2023-04-05 12:43:01,159 epoch 2 - iter 42/71 - loss 0.73990095 - time (sec): 21.41 - samples/sec: 163.64 - lr: 0.100000
362
+ 2023-04-05 12:43:04,701 epoch 2 - iter 49/71 - loss 0.71935682 - time (sec): 24.96 - samples/sec: 164.09 - lr: 0.100000
363
+ 2023-04-05 12:43:08,315 epoch 2 - iter 56/71 - loss 0.69784735 - time (sec): 28.57 - samples/sec: 163.32 - lr: 0.100000
364
+ 2023-04-05 12:43:11,830 epoch 2 - iter 63/71 - loss 0.68374551 - time (sec): 32.08 - samples/sec: 164.13 - lr: 0.100000
365
+ 2023-04-05 12:43:15,403 epoch 2 - iter 70/71 - loss 0.68318206 - time (sec): 35.66 - samples/sec: 165.13 - lr: 0.100000
366
+ 2023-04-05 12:43:16,197 ----------------------------------------------------------------------------------------------------
367
+ 2023-04-05 12:43:16,199 EPOCH 2 done: loss 0.6825 - lr 0.100000
368
+ 2023-04-05 12:43:47,058 Evaluating as a multi-label problem: False
369
+ 2023-04-05 12:43:47,079 TRAIN : loss 0.5168215036392212 - f1-score (micro avg) 0.7765
370
+ 2023-04-05 12:43:50,605 Evaluating as a multi-label problem: False
371
+ 2023-04-05 12:43:50,618 DEV : loss 0.6507071852684021 - f1-score (micro avg) 0.7109
372
+ 2023-04-05 12:43:50,623 BAD EPOCHS (no improvement): 0
373
+ 2023-04-05 12:43:50,624 saving best model
374
+ 2023-04-05 12:43:51,833 ----------------------------------------------------------------------------------------------------
375
+ 2023-04-05 12:43:54,803 epoch 3 - iter 7/71 - loss 0.54805596 - time (sec): 2.97 - samples/sec: 196.06 - lr: 0.100000
376
+ 2023-04-05 12:43:58,496 epoch 3 - iter 14/71 - loss 0.52917100 - time (sec): 6.66 - samples/sec: 178.19 - lr: 0.100000
377
+ 2023-04-05 12:44:02,157 epoch 3 - iter 21/71 - loss 0.53172923 - time (sec): 10.32 - samples/sec: 173.12 - lr: 0.100000
378
+ 2023-04-05 12:44:05,748 epoch 3 - iter 28/71 - loss 0.53976229 - time (sec): 13.91 - samples/sec: 168.90 - lr: 0.100000
379
+ 2023-04-05 12:44:09,391 epoch 3 - iter 35/71 - loss 0.54268651 - time (sec): 17.56 - samples/sec: 167.34 - lr: 0.100000
380
+ 2023-04-05 12:44:13,039 epoch 3 - iter 42/71 - loss 0.53676146 - time (sec): 21.20 - samples/sec: 165.96 - lr: 0.100000
381
+ 2023-04-05 12:44:16,753 epoch 3 - iter 49/71 - loss 0.52184996 - time (sec): 24.92 - samples/sec: 163.25 - lr: 0.100000
382
+ 2023-04-05 12:44:20,426 epoch 3 - iter 56/71 - loss 0.51321120 - time (sec): 28.59 - samples/sec: 162.14 - lr: 0.100000
383
+ 2023-04-05 12:44:24,224 epoch 3 - iter 63/71 - loss 0.50438788 - time (sec): 32.39 - samples/sec: 161.35 - lr: 0.100000
384
+ 2023-04-05 12:44:27,884 epoch 3 - iter 70/71 - loss 0.51802873 - time (sec): 36.05 - samples/sec: 163.16 - lr: 0.100000
385
+ 2023-04-05 12:44:28,729 ----------------------------------------------------------------------------------------------------
386
+ 2023-04-05 12:44:28,730 EPOCH 3 done: loss 0.5184 - lr 0.100000
387
+ 2023-04-05 12:44:59,957 Evaluating as a multi-label problem: False
388
+ 2023-04-05 12:44:59,978 TRAIN : loss 0.3893236517906189 - f1-score (micro avg) 0.8265
389
+ 2023-04-05 12:45:03,682 Evaluating as a multi-label problem: False
390
+ 2023-04-05 12:45:03,692 DEV : loss 0.4997367858886719 - f1-score (micro avg) 0.7814
391
+ 2023-04-05 12:45:03,700 BAD EPOCHS (no improvement): 0
392
+ 2023-04-05 12:45:03,702 saving best model
393
+ 2023-04-05 12:45:04,923 ----------------------------------------------------------------------------------------------------
394
+ 2023-04-05 12:45:07,851 epoch 4 - iter 7/71 - loss 0.40043816 - time (sec): 2.92 - samples/sec: 198.64 - lr: 0.100000
395
+ 2023-04-05 12:45:11,521 epoch 4 - iter 14/71 - loss 0.43328716 - time (sec): 6.60 - samples/sec: 179.83 - lr: 0.100000
396
+ 2023-04-05 12:45:15,244 epoch 4 - iter 21/71 - loss 0.44871122 - time (sec): 10.32 - samples/sec: 173.97 - lr: 0.100000
397
+ 2023-04-05 12:45:18,967 epoch 4 - iter 28/71 - loss 0.44610987 - time (sec): 14.04 - samples/sec: 167.08 - lr: 0.100000
398
+ 2023-04-05 12:45:22,672 epoch 4 - iter 35/71 - loss 0.46221491 - time (sec): 17.75 - samples/sec: 167.81 - lr: 0.100000
399
+ 2023-04-05 12:45:26,360 epoch 4 - iter 42/71 - loss 0.43613944 - time (sec): 21.43 - samples/sec: 166.04 - lr: 0.100000
400
+ 2023-04-05 12:45:30,026 epoch 4 - iter 49/71 - loss 0.43457310 - time (sec): 25.10 - samples/sec: 163.95 - lr: 0.100000
401
+ 2023-04-05 12:45:33,722 epoch 4 - iter 56/71 - loss 0.42344792 - time (sec): 28.80 - samples/sec: 163.63 - lr: 0.100000
402
+ 2023-04-05 12:45:37,362 epoch 4 - iter 63/71 - loss 0.41981578 - time (sec): 32.44 - samples/sec: 163.15 - lr: 0.100000
403
+ 2023-04-05 12:45:41,136 epoch 4 - iter 70/71 - loss 0.41432375 - time (sec): 36.21 - samples/sec: 162.52 - lr: 0.100000
404
+ 2023-04-05 12:45:41,950 ----------------------------------------------------------------------------------------------------
405
+ 2023-04-05 12:45:41,951 EPOCH 4 done: loss 0.4140 - lr 0.100000
406
+ 2023-04-05 12:46:11,773 Evaluating as a multi-label problem: False
407
+ 2023-04-05 12:46:11,790 TRAIN : loss 0.3005659878253937 - f1-score (micro avg) 0.8763
408
+ 2023-04-05 12:46:15,334 Evaluating as a multi-label problem: False
409
+ 2023-04-05 12:46:15,345 DEV : loss 0.3956667482852936 - f1-score (micro avg) 0.8253
410
+ 2023-04-05 12:46:15,352 BAD EPOCHS (no improvement): 0
411
+ 2023-04-05 12:46:15,353 saving best model
412
+ 2023-04-05 12:46:16,744 ----------------------------------------------------------------------------------------------------
413
+ 2023-04-05 12:46:19,771 epoch 5 - iter 7/71 - loss 0.33104508 - time (sec): 3.03 - samples/sec: 191.03 - lr: 0.100000
414
+ 2023-04-05 12:46:23,484 epoch 5 - iter 14/71 - loss 0.37656699 - time (sec): 6.74 - samples/sec: 178.21 - lr: 0.100000
415
+ 2023-04-05 12:46:27,097 epoch 5 - iter 21/71 - loss 0.37195919 - time (sec): 10.35 - samples/sec: 172.91 - lr: 0.100000
416
+ 2023-04-05 12:46:30,706 epoch 5 - iter 28/71 - loss 0.37480153 - time (sec): 13.96 - samples/sec: 168.97 - lr: 0.100000
417
+ 2023-04-05 12:46:34,278 epoch 5 - iter 35/71 - loss 0.38207478 - time (sec): 17.53 - samples/sec: 169.28 - lr: 0.100000
418
+ 2023-04-05 12:46:38,037 epoch 5 - iter 42/71 - loss 0.37466200 - time (sec): 21.29 - samples/sec: 166.22 - lr: 0.100000
419
+ 2023-04-05 12:46:41,693 epoch 5 - iter 49/71 - loss 0.37114011 - time (sec): 24.95 - samples/sec: 165.58 - lr: 0.100000
420
+ 2023-04-05 12:46:45,303 epoch 5 - iter 56/71 - loss 0.37649024 - time (sec): 28.56 - samples/sec: 165.56 - lr: 0.100000
421
+ 2023-04-05 12:46:49,121 epoch 5 - iter 63/71 - loss 0.37565114 - time (sec): 32.38 - samples/sec: 164.82 - lr: 0.100000
422
+ 2023-04-05 12:46:52,825 epoch 5 - iter 70/71 - loss 0.36875926 - time (sec): 36.08 - samples/sec: 163.00 - lr: 0.100000
423
+ 2023-04-05 12:46:53,641 ----------------------------------------------------------------------------------------------------
424
+ 2023-04-05 12:46:53,642 EPOCH 5 done: loss 0.3682 - lr 0.100000
425
+ 2023-04-05 12:47:23,884 Evaluating as a multi-label problem: False
426
+ 2023-04-05 12:47:23,900 TRAIN : loss 0.24057592451572418 - f1-score (micro avg) 0.8969
427
+ 2023-04-05 12:47:27,486 Evaluating as a multi-label problem: False
428
+ 2023-04-05 12:47:27,497 DEV : loss 0.3635919690132141 - f1-score (micro avg) 0.808
429
+ 2023-04-05 12:47:27,503 BAD EPOCHS (no improvement): 1
430
+ 2023-04-05 12:47:27,509 ----------------------------------------------------------------------------------------------------
431
+ 2023-04-05 12:47:30,597 epoch 6 - iter 7/71 - loss 0.29060467 - time (sec): 3.09 - samples/sec: 191.74 - lr: 0.100000
432
+ 2023-04-05 12:47:34,411 epoch 6 - iter 14/71 - loss 0.33442431 - time (sec): 6.90 - samples/sec: 175.32 - lr: 0.100000
433
+ 2023-04-05 12:47:38,195 epoch 6 - iter 21/71 - loss 0.33081293 - time (sec): 10.69 - samples/sec: 169.02 - lr: 0.100000
434
+ 2023-04-05 12:47:41,883 epoch 6 - iter 28/71 - loss 0.31355855 - time (sec): 14.37 - samples/sec: 164.05 - lr: 0.100000
435
+ 2023-04-05 12:47:45,626 epoch 6 - iter 35/71 - loss 0.30769625 - time (sec): 18.12 - samples/sec: 161.79 - lr: 0.100000
436
+ 2023-04-05 12:47:49,335 epoch 6 - iter 42/71 - loss 0.30399027 - time (sec): 21.83 - samples/sec: 160.77 - lr: 0.100000
437
+ 2023-04-05 12:47:53,093 epoch 6 - iter 49/71 - loss 0.30522447 - time (sec): 25.58 - samples/sec: 161.32 - lr: 0.100000
438
+ 2023-04-05 12:47:56,821 epoch 6 - iter 56/71 - loss 0.30943381 - time (sec): 29.31 - samples/sec: 160.49 - lr: 0.100000
439
+ 2023-04-05 12:48:00,614 epoch 6 - iter 63/71 - loss 0.30689469 - time (sec): 33.10 - samples/sec: 159.74 - lr: 0.100000
440
+ 2023-04-05 12:48:04,217 epoch 6 - iter 70/71 - loss 0.31271804 - time (sec): 36.71 - samples/sec: 160.30 - lr: 0.100000
441
+ 2023-04-05 12:48:05,056 ----------------------------------------------------------------------------------------------------
442
+ 2023-04-05 12:48:05,057 EPOCH 6 done: loss 0.3120 - lr 0.100000
443
+ 2023-04-05 12:48:35,310 Evaluating as a multi-label problem: False
444
+ 2023-04-05 12:48:35,330 TRAIN : loss 0.2524365782737732 - f1-score (micro avg) 0.8698
445
+ 2023-04-05 12:48:38,882 Evaluating as a multi-label problem: False
446
+ 2023-04-05 12:48:38,893 DEV : loss 0.3847086429595947 - f1-score (micro avg) 0.7949
447
+ 2023-04-05 12:48:38,897 BAD EPOCHS (no improvement): 2
448
+ 2023-04-05 12:48:38,900 ----------------------------------------------------------------------------------------------------
449
+ 2023-04-05 12:48:41,937 epoch 7 - iter 7/71 - loss 0.37316342 - time (sec): 3.04 - samples/sec: 194.56 - lr: 0.100000
450
+ 2023-04-05 12:48:45,658 epoch 7 - iter 14/71 - loss 0.30328593 - time (sec): 6.76 - samples/sec: 173.42 - lr: 0.100000
451
+ 2023-04-05 12:48:49,422 epoch 7 - iter 21/71 - loss 0.29911219 - time (sec): 10.52 - samples/sec: 168.42 - lr: 0.100000
452
+ 2023-04-05 12:48:53,079 epoch 7 - iter 28/71 - loss 0.29218439 - time (sec): 14.18 - samples/sec: 167.57 - lr: 0.100000
453
+ 2023-04-05 12:48:56,921 epoch 7 - iter 35/71 - loss 0.29301233 - time (sec): 18.02 - samples/sec: 164.19 - lr: 0.100000
454
+ 2023-04-05 12:49:00,741 epoch 7 - iter 42/71 - loss 0.29024371 - time (sec): 21.84 - samples/sec: 162.13 - lr: 0.100000
455
+ 2023-04-05 12:49:04,470 epoch 7 - iter 49/71 - loss 0.29491898 - time (sec): 25.57 - samples/sec: 161.04 - lr: 0.100000
456
+ 2023-04-05 12:49:08,148 epoch 7 - iter 56/71 - loss 0.29485370 - time (sec): 29.25 - samples/sec: 161.27 - lr: 0.100000
457
+ 2023-04-05 12:49:11,798 epoch 7 - iter 63/71 - loss 0.29048646 - time (sec): 32.90 - samples/sec: 161.74 - lr: 0.100000
458
+ 2023-04-05 12:49:15,495 epoch 7 - iter 70/71 - loss 0.28277861 - time (sec): 36.60 - samples/sec: 160.65 - lr: 0.100000
459
+ 2023-04-05 12:49:16,301 ----------------------------------------------------------------------------------------------------
460
+ 2023-04-05 12:49:16,302 EPOCH 7 done: loss 0.2833 - lr 0.100000
461
+ 2023-04-05 12:49:46,527 Evaluating as a multi-label problem: False
462
+ 2023-04-05 12:49:46,546 TRAIN : loss 0.21568651497364044 - f1-score (micro avg) 0.8845
463
+ 2023-04-05 12:49:50,126 Evaluating as a multi-label problem: False
464
+ 2023-04-05 12:49:50,138 DEV : loss 0.3467901051044464 - f1-score (micro avg) 0.8182
465
+ 2023-04-05 12:49:50,144 BAD EPOCHS (no improvement): 3
466
+ 2023-04-05 12:49:50,145 ----------------------------------------------------------------------------------------------------
467
+ 2023-04-05 12:49:53,171 epoch 8 - iter 7/71 - loss 0.25938026 - time (sec): 3.03 - samples/sec: 191.70 - lr: 0.100000
468
+ 2023-04-05 12:49:56,942 epoch 8 - iter 14/71 - loss 0.25775377 - time (sec): 6.80 - samples/sec: 173.18 - lr: 0.100000
469
+ 2023-04-05 12:50:00,558 epoch 8 - iter 21/71 - loss 0.25357329 - time (sec): 10.41 - samples/sec: 168.64 - lr: 0.100000
470
+ 2023-04-05 12:50:04,296 epoch 8 - iter 28/71 - loss 0.24824351 - time (sec): 14.15 - samples/sec: 164.02 - lr: 0.100000
471
+ 2023-04-05 12:50:08,001 epoch 8 - iter 35/71 - loss 0.25876964 - time (sec): 17.86 - samples/sec: 162.19 - lr: 0.100000
472
+ 2023-04-05 12:50:11,740 epoch 8 - iter 42/71 - loss 0.25788824 - time (sec): 21.60 - samples/sec: 162.58 - lr: 0.100000
473
+ 2023-04-05 12:50:15,342 epoch 8 - iter 49/71 - loss 0.25424281 - time (sec): 25.20 - samples/sec: 162.72 - lr: 0.100000
474
+ 2023-04-05 12:50:19,188 epoch 8 - iter 56/71 - loss 0.25450229 - time (sec): 29.04 - samples/sec: 161.73 - lr: 0.100000
475
+ 2023-04-05 12:50:23,093 epoch 8 - iter 63/71 - loss 0.25746436 - time (sec): 32.95 - samples/sec: 160.01 - lr: 0.100000
476
+ 2023-04-05 12:50:27,329 epoch 8 - iter 70/71 - loss 0.25933626 - time (sec): 37.18 - samples/sec: 158.05 - lr: 0.100000
477
+ 2023-04-05 12:50:28,163 ----------------------------------------------------------------------------------------------------
478
+ 2023-04-05 12:50:28,165 EPOCH 8 done: loss 0.2589 - lr 0.100000
479
+ 2023-04-05 12:50:59,296 Evaluating as a multi-label problem: False
480
+ 2023-04-05 12:50:59,312 TRAIN : loss 0.19819645583629608 - f1-score (micro avg) 0.8943
481
+ 2023-04-05 12:51:02,965 Evaluating as a multi-label problem: False
482
+ 2023-04-05 12:51:02,980 DEV : loss 0.33811214566230774 - f1-score (micro avg) 0.8118
483
+ 2023-04-05 12:51:02,988 Epoch 8: reducing learning rate of group 0 to 5.0000e-02.
484
+ 2023-04-05 12:51:02,989 BAD EPOCHS (no improvement): 4
485
+ 2023-04-05 12:51:02,990 ----------------------------------------------------------------------------------------------------
486
+ 2023-04-05 12:51:06,070 epoch 9 - iter 7/71 - loss 0.23812631 - time (sec): 3.08 - samples/sec: 194.17 - lr: 0.050000
487
+ 2023-04-05 12:51:09,867 epoch 9 - iter 14/71 - loss 0.19886070 - time (sec): 6.88 - samples/sec: 169.72 - lr: 0.050000
488
+ 2023-04-05 12:51:13,813 epoch 9 - iter 21/71 - loss 0.22626942 - time (sec): 10.82 - samples/sec: 161.52 - lr: 0.050000
489
+ 2023-04-05 12:51:17,892 epoch 9 - iter 28/71 - loss 0.22291349 - time (sec): 14.90 - samples/sec: 158.04 - lr: 0.050000
490
+ 2023-04-05 12:51:21,762 epoch 9 - iter 35/71 - loss 0.22330927 - time (sec): 18.77 - samples/sec: 155.93 - lr: 0.050000
491
+ 2023-04-05 12:51:25,462 epoch 9 - iter 42/71 - loss 0.22661256 - time (sec): 22.47 - samples/sec: 158.69 - lr: 0.050000
492
+ 2023-04-05 12:51:29,109 epoch 9 - iter 49/71 - loss 0.22209246 - time (sec): 26.12 - samples/sec: 158.89 - lr: 0.050000
493
+ 2023-04-05 12:51:32,660 epoch 9 - iter 56/71 - loss 0.21543228 - time (sec): 29.67 - samples/sec: 158.89 - lr: 0.050000
494
+ 2023-04-05 12:51:36,421 epoch 9 - iter 63/71 - loss 0.22191567 - time (sec): 33.43 - samples/sec: 158.57 - lr: 0.050000
495
+ 2023-04-05 12:51:40,056 epoch 9 - iter 70/71 - loss 0.22014300 - time (sec): 37.06 - samples/sec: 158.83 - lr: 0.050000
496
+ 2023-04-05 12:51:40,907 ----------------------------------------------------------------------------------------------------
497
+ 2023-04-05 12:51:40,908 EPOCH 9 done: loss 0.2208 - lr 0.050000
498
+ 2023-04-05 12:52:11,299 Evaluating as a multi-label problem: False
499
+ 2023-04-05 12:52:11,317 TRAIN : loss 0.14378328621387482 - f1-score (micro avg) 0.9285
500
+ 2023-04-05 12:52:14,963 Evaluating as a multi-label problem: False
501
+ 2023-04-05 12:52:14,976 DEV : loss 0.2700260877609253 - f1-score (micro avg) 0.8339
502
+ 2023-04-05 12:52:14,981 BAD EPOCHS (no improvement): 0
503
+ 2023-04-05 12:52:14,983 saving best model
504
+ 2023-04-05 12:52:16,467 ----------------------------------------------------------------------------------------------------
505
+ 2023-04-05 12:52:19,567 epoch 10 - iter 7/71 - loss 0.17365492 - time (sec): 3.10 - samples/sec: 178.18 - lr: 0.050000
506
+ 2023-04-05 12:52:23,341 epoch 10 - iter 14/71 - loss 0.18889650 - time (sec): 6.87 - samples/sec: 165.16 - lr: 0.050000
507
+ 2023-04-05 12:52:27,018 epoch 10 - iter 21/71 - loss 0.20267585 - time (sec): 10.55 - samples/sec: 165.89 - lr: 0.050000
508
+ 2023-04-05 12:52:30,734 epoch 10 - iter 28/71 - loss 0.19907475 - time (sec): 14.26 - samples/sec: 163.63 - lr: 0.050000
509
+ 2023-04-05 12:52:34,373 epoch 10 - iter 35/71 - loss 0.19695937 - time (sec): 17.90 - samples/sec: 162.87 - lr: 0.050000
510
+ 2023-04-05 12:52:38,176 epoch 10 - iter 42/71 - loss 0.20787867 - time (sec): 21.71 - samples/sec: 163.27 - lr: 0.050000
511
+ 2023-04-05 12:52:41,917 epoch 10 - iter 49/71 - loss 0.21020299 - time (sec): 25.45 - samples/sec: 162.92 - lr: 0.050000
512
+ 2023-04-05 12:52:45,576 epoch 10 - iter 56/71 - loss 0.21217935 - time (sec): 29.11 - samples/sec: 162.53 - lr: 0.050000
513
+ 2023-04-05 12:52:49,330 epoch 10 - iter 63/71 - loss 0.20993717 - time (sec): 32.86 - samples/sec: 161.37 - lr: 0.050000
514
+ 2023-04-05 12:52:53,093 epoch 10 - iter 70/71 - loss 0.20975981 - time (sec): 36.62 - samples/sec: 160.74 - lr: 0.050000
515
+ 2023-04-05 12:52:53,956 ----------------------------------------------------------------------------------------------------
516
+ 2023-04-05 12:52:53,957 EPOCH 10 done: loss 0.2100 - lr 0.050000
517
+ 2023-04-05 12:53:24,907 Evaluating as a multi-label problem: False
518
+ 2023-04-05 12:53:24,923 TRAIN : loss 0.1295960396528244 - f1-score (micro avg) 0.936
519
+ 2023-04-05 12:53:28,577 Evaluating as a multi-label problem: False
520
+ 2023-04-05 12:53:28,590 DEV : loss 0.2524380087852478 - f1-score (micro avg) 0.8561
521
+ 2023-04-05 12:53:28,596 BAD EPOCHS (no improvement): 0
522
+ 2023-04-05 12:53:28,601 saving best model
523
+ 2023-04-05 12:53:29,758 ----------------------------------------------------------------------------------------------------
524
+ 2023-04-05 12:53:32,976 epoch 11 - iter 7/71 - loss 0.23687125 - time (sec): 3.22 - samples/sec: 196.25 - lr: 0.050000
525
+ 2023-04-05 12:53:36,734 epoch 11 - iter 14/71 - loss 0.23011958 - time (sec): 6.97 - samples/sec: 175.94 - lr: 0.050000
526
+ 2023-04-05 12:53:40,400 epoch 11 - iter 21/71 - loss 0.20729973 - time (sec): 10.64 - samples/sec: 170.31 - lr: 0.050000
527
+ 2023-04-05 12:53:44,055 epoch 11 - iter 28/71 - loss 0.20154173 - time (sec): 14.29 - samples/sec: 166.99 - lr: 0.050000
528
+ 2023-04-05 12:53:47,926 epoch 11 - iter 35/71 - loss 0.20303865 - time (sec): 18.17 - samples/sec: 164.55 - lr: 0.050000
529
+ 2023-04-05 12:53:51,746 epoch 11 - iter 42/71 - loss 0.19656997 - time (sec): 21.99 - samples/sec: 162.47 - lr: 0.050000
530
+ 2023-04-05 12:53:55,488 epoch 11 - iter 49/71 - loss 0.20078721 - time (sec): 25.73 - samples/sec: 160.57 - lr: 0.050000
531
+ 2023-04-05 12:53:59,182 epoch 11 - iter 56/71 - loss 0.19665885 - time (sec): 29.42 - samples/sec: 160.26 - lr: 0.050000
532
+ 2023-04-05 12:54:03,005 epoch 11 - iter 63/71 - loss 0.19670097 - time (sec): 33.24 - samples/sec: 158.79 - lr: 0.050000
533
+ 2023-04-05 12:54:06,832 epoch 11 - iter 70/71 - loss 0.19415408 - time (sec): 37.07 - samples/sec: 158.61 - lr: 0.050000
534
+ 2023-04-05 12:54:07,718 ----------------------------------------------------------------------------------------------------
535
+ 2023-04-05 12:54:07,719 EPOCH 11 done: loss 0.1942 - lr 0.050000
536
+ 2023-04-05 12:54:38,401 Evaluating as a multi-label problem: False
537
+ 2023-04-05 12:54:38,416 TRAIN : loss 0.1299719512462616 - f1-score (micro avg) 0.9345
538
+ 2023-04-05 12:54:42,043 Evaluating as a multi-label problem: False
539
+ 2023-04-05 12:54:42,055 DEV : loss 0.2761968970298767 - f1-score (micro avg) 0.8452
540
+ 2023-04-05 12:54:42,061 BAD EPOCHS (no improvement): 1
541
+ 2023-04-05 12:54:42,062 ----------------------------------------------------------------------------------------------------
542
+ 2023-04-05 12:54:45,126 epoch 12 - iter 7/71 - loss 0.18339264 - time (sec): 3.06 - samples/sec: 187.09 - lr: 0.050000
543
+ 2023-04-05 12:54:48,926 epoch 12 - iter 14/71 - loss 0.19237624 - time (sec): 6.86 - samples/sec: 173.27 - lr: 0.050000
544
+ 2023-04-05 12:54:52,633 epoch 12 - iter 21/71 - loss 0.19432209 - time (sec): 10.57 - samples/sec: 166.42 - lr: 0.050000
545
+ 2023-04-05 12:54:56,545 epoch 12 - iter 28/71 - loss 0.20200765 - time (sec): 14.48 - samples/sec: 162.90 - lr: 0.050000
546
+ 2023-04-05 12:55:00,495 epoch 12 - iter 35/71 - loss 0.19446487 - time (sec): 18.43 - samples/sec: 160.60 - lr: 0.050000
547
+ 2023-04-05 12:55:04,256 epoch 12 - iter 42/71 - loss 0.19910943 - time (sec): 22.19 - samples/sec: 160.01 - lr: 0.050000
548
+ 2023-04-05 12:55:08,062 epoch 12 - iter 49/71 - loss 0.19637866 - time (sec): 26.00 - samples/sec: 158.66 - lr: 0.050000
549
+ 2023-04-05 12:55:11,765 epoch 12 - iter 56/71 - loss 0.19106381 - time (sec): 29.70 - samples/sec: 158.17 - lr: 0.050000
550
+ 2023-04-05 12:55:15,677 epoch 12 - iter 63/71 - loss 0.19328764 - time (sec): 33.61 - samples/sec: 157.71 - lr: 0.050000
551
+ 2023-04-05 12:55:19,610 epoch 12 - iter 70/71 - loss 0.18986505 - time (sec): 37.55 - samples/sec: 156.77 - lr: 0.050000
552
+ 2023-04-05 12:55:20,532 ----------------------------------------------------------------------------------------------------
553
+ 2023-04-05 12:55:20,534 EPOCH 12 done: loss 0.1892 - lr 0.050000
554
+ 2023-04-05 12:55:51,520 Evaluating as a multi-label problem: False
555
+ 2023-04-05 12:55:51,541 TRAIN : loss 0.11139164865016937 - f1-score (micro avg) 0.9468
556
+ 2023-04-05 12:55:55,099 Evaluating as a multi-label problem: False
557
+ 2023-04-05 12:55:55,107 DEV : loss 0.24872702360153198 - f1-score (micro avg) 0.8656
558
+ 2023-04-05 12:55:55,114 BAD EPOCHS (no improvement): 0
559
+ 2023-04-05 12:55:55,116 saving best model
560
+ 2023-04-05 12:55:56,261 ----------------------------------------------------------------------------------------------------
561
+ 2023-04-05 12:55:59,215 epoch 13 - iter 7/71 - loss 0.13416757 - time (sec): 2.95 - samples/sec: 191.70 - lr: 0.050000
562
+ 2023-04-05 12:56:02,877 epoch 13 - iter 14/71 - loss 0.14492713 - time (sec): 6.61 - samples/sec: 173.40 - lr: 0.050000
563
+ 2023-04-05 12:56:06,537 epoch 13 - iter 21/71 - loss 0.15026220 - time (sec): 10.27 - samples/sec: 165.55 - lr: 0.050000
564
+ 2023-04-05 12:56:10,099 epoch 13 - iter 28/71 - loss 0.14928537 - time (sec): 13.84 - samples/sec: 166.95 - lr: 0.050000
565
+ 2023-04-05 12:56:13,686 epoch 13 - iter 35/71 - loss 0.14769742 - time (sec): 17.42 - samples/sec: 165.57 - lr: 0.050000
566
+ 2023-04-05 12:56:17,287 epoch 13 - iter 42/71 - loss 0.15688659 - time (sec): 21.02 - samples/sec: 164.09 - lr: 0.050000
567
+ 2023-04-05 12:56:20,996 epoch 13 - iter 49/71 - loss 0.16690754 - time (sec): 24.73 - samples/sec: 164.31 - lr: 0.050000
568
+ 2023-04-05 12:56:24,678 epoch 13 - iter 56/71 - loss 0.17086305 - time (sec): 28.42 - samples/sec: 164.45 - lr: 0.050000
569
+ 2023-04-05 12:56:28,394 epoch 13 - iter 63/71 - loss 0.17395421 - time (sec): 32.13 - samples/sec: 164.13 - lr: 0.050000
570
+ 2023-04-05 12:56:32,037 epoch 13 - iter 70/71 - loss 0.17863766 - time (sec): 35.77 - samples/sec: 164.47 - lr: 0.050000
571
+ 2023-04-05 12:56:32,895 ----------------------------------------------------------------------------------------------------
572
+ 2023-04-05 12:56:32,895 EPOCH 13 done: loss 0.1786 - lr 0.050000
573
+ 2023-04-05 12:57:02,811 Evaluating as a multi-label problem: False
574
+ 2023-04-05 12:57:02,831 TRAIN : loss 0.11495152860879898 - f1-score (micro avg) 0.9423
575
+ 2023-04-05 12:57:06,419 Evaluating as a multi-label problem: False
576
+ 2023-04-05 12:57:06,429 DEV : loss 0.2527526021003723 - f1-score (micro avg) 0.8571
577
+ 2023-04-05 12:57:06,433 BAD EPOCHS (no improvement): 1
578
+ 2023-04-05 12:57:06,435 ----------------------------------------------------------------------------------------------------
579
+ 2023-04-05 12:57:09,526 epoch 14 - iter 7/71 - loss 0.19849680 - time (sec): 3.09 - samples/sec: 198.95 - lr: 0.050000
580
+ 2023-04-05 12:57:13,202 epoch 14 - iter 14/71 - loss 0.18692846 - time (sec): 6.77 - samples/sec: 173.93 - lr: 0.050000
581
+ 2023-04-05 12:57:16,931 epoch 14 - iter 21/71 - loss 0.16410048 - time (sec): 10.50 - samples/sec: 165.11 - lr: 0.050000
582
+ 2023-04-05 12:57:20,568 epoch 14 - iter 28/71 - loss 0.17096232 - time (sec): 14.13 - samples/sec: 164.73 - lr: 0.050000
583
+ 2023-04-05 12:57:24,320 epoch 14 - iter 35/71 - loss 0.17130471 - time (sec): 17.88 - samples/sec: 164.61 - lr: 0.050000
584
+ 2023-04-05 12:57:28,263 epoch 14 - iter 42/71 - loss 0.17276982 - time (sec): 21.83 - samples/sec: 162.55 - lr: 0.050000
585
+ 2023-04-05 12:57:31,855 epoch 14 - iter 49/71 - loss 0.17053371 - time (sec): 25.42 - samples/sec: 161.88 - lr: 0.050000
586
+ 2023-04-05 12:57:35,454 epoch 14 - iter 56/71 - loss 0.17009003 - time (sec): 29.02 - samples/sec: 161.72 - lr: 0.050000
587
+ 2023-04-05 12:57:39,094 epoch 14 - iter 63/71 - loss 0.16765044 - time (sec): 32.66 - samples/sec: 161.28 - lr: 0.050000
588
+ 2023-04-05 12:57:42,812 epoch 14 - iter 70/71 - loss 0.17106035 - time (sec): 36.38 - samples/sec: 161.78 - lr: 0.050000
589
+ 2023-04-05 12:57:43,636 ----------------------------------------------------------------------------------------------------
590
+ 2023-04-05 12:57:43,637 EPOCH 14 done: loss 0.1709 - lr 0.050000
591
+ 2023-04-05 12:58:13,469 Evaluating as a multi-label problem: False
592
+ 2023-04-05 12:58:13,486 TRAIN : loss 0.09600471705198288 - f1-score (micro avg) 0.9574
593
+ 2023-04-05 12:58:17,049 Evaluating as a multi-label problem: False
594
+ 2023-04-05 12:58:17,060 DEV : loss 0.2339187115430832 - f1-score (micro avg) 0.8799
595
+ 2023-04-05 12:58:17,066 BAD EPOCHS (no improvement): 0
596
+ 2023-04-05 12:58:17,068 saving best model
597
+ 2023-04-05 12:58:18,074 ----------------------------------------------------------------------------------------------------
598
+ 2023-04-05 12:58:21,099 epoch 15 - iter 7/71 - loss 0.18573106 - time (sec): 3.02 - samples/sec: 195.41 - lr: 0.050000
599
+ 2023-04-05 12:58:24,823 epoch 15 - iter 14/71 - loss 0.16845787 - time (sec): 6.75 - samples/sec: 173.09 - lr: 0.050000
600
+ 2023-04-05 12:58:28,520 epoch 15 - iter 21/71 - loss 0.17292932 - time (sec): 10.45 - samples/sec: 168.59 - lr: 0.050000
601
+ 2023-04-05 12:58:32,285 epoch 15 - iter 28/71 - loss 0.16194007 - time (sec): 14.21 - samples/sec: 164.53 - lr: 0.050000
602
+ 2023-04-05 12:58:35,962 epoch 15 - iter 35/71 - loss 0.16402796 - time (sec): 17.89 - samples/sec: 163.30 - lr: 0.050000
603
+ 2023-04-05 12:58:39,707 epoch 15 - iter 42/71 - loss 0.16481449 - time (sec): 21.63 - samples/sec: 163.18 - lr: 0.050000
604
+ 2023-04-05 12:58:43,272 epoch 15 - iter 49/71 - loss 0.16905415 - time (sec): 25.20 - samples/sec: 163.51 - lr: 0.050000
605
+ 2023-04-05 12:58:46,892 epoch 15 - iter 56/71 - loss 0.17389761 - time (sec): 28.82 - samples/sec: 162.99 - lr: 0.050000
606
+ 2023-04-05 12:58:50,539 epoch 15 - iter 63/71 - loss 0.18006799 - time (sec): 32.46 - samples/sec: 163.29 - lr: 0.050000
607
+ 2023-04-05 12:58:54,178 epoch 15 - iter 70/71 - loss 0.17852127 - time (sec): 36.10 - samples/sec: 162.76 - lr: 0.050000
608
+ 2023-04-05 12:58:54,989 ----------------------------------------------------------------------------------------------------
609
+ 2023-04-05 12:58:54,990 EPOCH 15 done: loss 0.1795 - lr 0.050000
610
+ 2023-04-05 12:59:25,120 Evaluating as a multi-label problem: False
611
+ 2023-04-05 12:59:25,142 TRAIN : loss 0.10160095989704132 - f1-score (micro avg) 0.9519
612
+ 2023-04-05 12:59:28,704 Evaluating as a multi-label problem: False
613
+ 2023-04-05 12:59:28,715 DEV : loss 0.22547538578510284 - f1-score (micro avg) 0.8848
614
+ 2023-04-05 12:59:28,722 BAD EPOCHS (no improvement): 0
615
+ 2023-04-05 12:59:28,724 saving best model
616
+ 2023-04-05 12:59:29,886 ----------------------------------------------------------------------------------------------------
617
+ 2023-04-05 12:59:32,863 epoch 16 - iter 7/71 - loss 0.18803312 - time (sec): 2.98 - samples/sec: 201.30 - lr: 0.050000
618
+ 2023-04-05 12:59:36,639 epoch 16 - iter 14/71 - loss 0.17472744 - time (sec): 6.75 - samples/sec: 176.98 - lr: 0.050000
619
+ 2023-04-05 12:59:40,239 epoch 16 - iter 21/71 - loss 0.17674174 - time (sec): 10.35 - samples/sec: 173.11 - lr: 0.050000
620
+ 2023-04-05 12:59:43,953 epoch 16 - iter 28/71 - loss 0.18068979 - time (sec): 14.07 - samples/sec: 170.13 - lr: 0.050000
621
+ 2023-04-05 12:59:47,620 epoch 16 - iter 35/71 - loss 0.18222164 - time (sec): 17.73 - samples/sec: 168.22 - lr: 0.050000
622
+ 2023-04-05 12:59:51,255 epoch 16 - iter 42/71 - loss 0.18013268 - time (sec): 21.37 - samples/sec: 167.36 - lr: 0.050000
623
+ 2023-04-05 12:59:54,891 epoch 16 - iter 49/71 - loss 0.17236708 - time (sec): 25.00 - samples/sec: 166.02 - lr: 0.050000
624
+ 2023-04-05 12:59:58,571 epoch 16 - iter 56/71 - loss 0.17486601 - time (sec): 28.68 - samples/sec: 164.55 - lr: 0.050000
625
+ 2023-04-05 13:00:02,189 epoch 16 - iter 63/71 - loss 0.17192697 - time (sec): 32.30 - samples/sec: 164.58 - lr: 0.050000
626
+ 2023-04-05 13:00:05,957 epoch 16 - iter 70/71 - loss 0.16671679 - time (sec): 36.07 - samples/sec: 163.16 - lr: 0.050000
627
+ 2023-04-05 13:00:06,789 ----------------------------------------------------------------------------------------------------
628
+ 2023-04-05 13:00:06,790 EPOCH 16 done: loss 0.1660 - lr 0.050000
629
+ 2023-04-05 13:00:36,903 Evaluating as a multi-label problem: False
630
+ 2023-04-05 13:00:36,919 TRAIN : loss 0.09528940916061401 - f1-score (micro avg) 0.9534
631
+ 2023-04-05 13:00:40,480 Evaluating as a multi-label problem: False
632
+ 2023-04-05 13:00:40,495 DEV : loss 0.2436159998178482 - f1-score (micro avg) 0.8731
633
+ 2023-04-05 13:00:40,500 BAD EPOCHS (no improvement): 1
634
+ 2023-04-05 13:00:40,503 ----------------------------------------------------------------------------------------------------
635
+ 2023-04-05 13:00:43,573 epoch 17 - iter 7/71 - loss 0.16280295 - time (sec): 3.07 - samples/sec: 187.36 - lr: 0.050000
636
+ 2023-04-05 13:00:47,316 epoch 17 - iter 14/71 - loss 0.16491960 - time (sec): 6.81 - samples/sec: 171.74 - lr: 0.050000
637
+ 2023-04-05 13:00:51,066 epoch 17 - iter 21/71 - loss 0.16969325 - time (sec): 10.56 - samples/sec: 171.45 - lr: 0.050000
638
+ 2023-04-05 13:00:54,815 epoch 17 - iter 28/71 - loss 0.16213035 - time (sec): 14.31 - samples/sec: 166.86 - lr: 0.050000
639
+ 2023-04-05 13:00:58,536 epoch 17 - iter 35/71 - loss 0.16020484 - time (sec): 18.03 - samples/sec: 165.03 - lr: 0.050000
640
+ 2023-04-05 13:01:02,186 epoch 17 - iter 42/71 - loss 0.16493772 - time (sec): 21.68 - samples/sec: 162.94 - lr: 0.050000
641
+ 2023-04-05 13:01:05,927 epoch 17 - iter 49/71 - loss 0.16161421 - time (sec): 25.42 - samples/sec: 162.80 - lr: 0.050000
642
+ 2023-04-05 13:01:09,546 epoch 17 - iter 56/71 - loss 0.16247358 - time (sec): 29.04 - samples/sec: 163.49 - lr: 0.050000
643
+ 2023-04-05 13:01:13,292 epoch 17 - iter 63/71 - loss 0.16001124 - time (sec): 32.79 - samples/sec: 161.34 - lr: 0.050000
644
+ 2023-04-05 13:01:17,024 epoch 17 - iter 70/71 - loss 0.16184123 - time (sec): 36.52 - samples/sec: 161.03 - lr: 0.050000
645
+ 2023-04-05 13:01:17,872 ----------------------------------------------------------------------------------------------------
646
+ 2023-04-05 13:01:17,873 EPOCH 17 done: loss 0.1613 - lr 0.050000
647
+ 2023-04-05 13:01:48,038 Evaluating as a multi-label problem: False
648
+ 2023-04-05 13:01:48,055 TRAIN : loss 0.09658616781234741 - f1-score (micro avg) 0.9508
649
+ 2023-04-05 13:01:51,710 Evaluating as a multi-label problem: False
650
+ 2023-04-05 13:01:51,723 DEV : loss 0.25625115633010864 - f1-score (micro avg) 0.8651
651
+ 2023-04-05 13:01:51,730 BAD EPOCHS (no improvement): 2
652
+ 2023-04-05 13:01:51,731 ----------------------------------------------------------------------------------------------------
653
+ 2023-04-05 13:01:54,950 epoch 18 - iter 7/71 - loss 0.14699763 - time (sec): 3.22 - samples/sec: 178.08 - lr: 0.050000
654
+ 2023-04-05 13:01:58,906 epoch 18 - iter 14/71 - loss 0.11720383 - time (sec): 7.17 - samples/sec: 156.81 - lr: 0.050000
655
+ 2023-04-05 13:02:02,544 epoch 18 - iter 21/71 - loss 0.12927531 - time (sec): 10.81 - samples/sec: 160.28 - lr: 0.050000
656
+ 2023-04-05 13:02:06,291 epoch 18 - iter 28/71 - loss 0.13428711 - time (sec): 14.56 - samples/sec: 159.42 - lr: 0.050000
657
+ 2023-04-05 13:02:10,110 epoch 18 - iter 35/71 - loss 0.13574585 - time (sec): 18.38 - samples/sec: 157.52 - lr: 0.050000
658
+ 2023-04-05 13:02:13,768 epoch 18 - iter 42/71 - loss 0.13144409 - time (sec): 22.04 - samples/sec: 158.47 - lr: 0.050000
659
+ 2023-04-05 13:02:17,521 epoch 18 - iter 49/71 - loss 0.13997386 - time (sec): 25.79 - samples/sec: 158.95 - lr: 0.050000
660
+ 2023-04-05 13:02:21,181 epoch 18 - iter 56/71 - loss 0.15194990 - time (sec): 29.45 - samples/sec: 159.70 - lr: 0.050000
661
+ 2023-04-05 13:02:24,954 epoch 18 - iter 63/71 - loss 0.15543297 - time (sec): 33.22 - samples/sec: 159.95 - lr: 0.050000
662
+ 2023-04-05 13:02:28,617 epoch 18 - iter 70/71 - loss 0.15181266 - time (sec): 36.89 - samples/sec: 159.47 - lr: 0.050000
663
+ 2023-04-05 13:02:29,438 ----------------------------------------------------------------------------------------------------
664
+ 2023-04-05 13:02:29,438 EPOCH 18 done: loss 0.1521 - lr 0.050000
665
+ 2023-04-05 13:03:00,408 Evaluating as a multi-label problem: False
666
+ 2023-04-05 13:03:00,426 TRAIN : loss 0.07957068085670471 - f1-score (micro avg) 0.9605
667
+ 2023-04-05 13:03:04,203 Evaluating as a multi-label problem: False
668
+ 2023-04-05 13:03:04,214 DEV : loss 0.24449127912521362 - f1-score (micro avg) 0.8678
669
+ 2023-04-05 13:03:04,220 BAD EPOCHS (no improvement): 3
670
+ 2023-04-05 13:03:04,221 ----------------------------------------------------------------------------------------------------
671
+ 2023-04-05 13:03:07,272 epoch 19 - iter 7/71 - loss 0.14268361 - time (sec): 3.05 - samples/sec: 192.50 - lr: 0.050000
672
+ 2023-04-05 13:03:11,108 epoch 19 - iter 14/71 - loss 0.11705529 - time (sec): 6.89 - samples/sec: 171.38 - lr: 0.050000
673
+ 2023-04-05 13:03:14,913 epoch 19 - iter 21/71 - loss 0.13677063 - time (sec): 10.69 - samples/sec: 166.87 - lr: 0.050000
674
+ 2023-04-05 13:03:18,612 epoch 19 - iter 28/71 - loss 0.14036431 - time (sec): 14.39 - samples/sec: 164.42 - lr: 0.050000
675
+ 2023-04-05 13:03:22,346 epoch 19 - iter 35/71 - loss 0.14069950 - time (sec): 18.12 - samples/sec: 162.99 - lr: 0.050000
676
+ 2023-04-05 13:03:26,137 epoch 19 - iter 42/71 - loss 0.14416925 - time (sec): 21.91 - samples/sec: 161.81 - lr: 0.050000
677
+ 2023-04-05 13:03:29,727 epoch 19 - iter 49/71 - loss 0.14774011 - time (sec): 25.50 - samples/sec: 162.05 - lr: 0.050000
678
+ 2023-04-05 13:03:33,433 epoch 19 - iter 56/71 - loss 0.14539107 - time (sec): 29.21 - samples/sec: 161.11 - lr: 0.050000
679
+ 2023-04-05 13:03:37,072 epoch 19 - iter 63/71 - loss 0.14446964 - time (sec): 32.85 - samples/sec: 160.61 - lr: 0.050000
680
+ 2023-04-05 13:03:40,670 epoch 19 - iter 70/71 - loss 0.14622801 - time (sec): 36.45 - samples/sec: 161.22 - lr: 0.050000
681
+ 2023-04-05 13:03:41,501 ----------------------------------------------------------------------------------------------------
682
+ 2023-04-05 13:03:41,502 EPOCH 19 done: loss 0.1476 - lr 0.050000
683
+ 2023-04-05 13:04:11,752 Evaluating as a multi-label problem: False
684
+ 2023-04-05 13:04:11,772 TRAIN : loss 0.08018826693296432 - f1-score (micro avg) 0.9626
685
+ 2023-04-05 13:04:15,357 Evaluating as a multi-label problem: False
686
+ 2023-04-05 13:04:15,368 DEV : loss 0.2329457700252533 - f1-score (micro avg) 0.8603
687
+ 2023-04-05 13:04:15,373 Epoch 19: reducing learning rate of group 0 to 2.5000e-02.
688
+ 2023-04-05 13:04:15,374 BAD EPOCHS (no improvement): 4
689
+ 2023-04-05 13:04:15,376 ----------------------------------------------------------------------------------------------------
690
+ 2023-04-05 13:04:18,529 epoch 20 - iter 7/71 - loss 0.15115543 - time (sec): 3.15 - samples/sec: 187.15 - lr: 0.025000
691
+ 2023-04-05 13:04:22,232 epoch 20 - iter 14/71 - loss 0.14745089 - time (sec): 6.86 - samples/sec: 170.06 - lr: 0.025000
692
+ 2023-04-05 13:04:25,979 epoch 20 - iter 21/71 - loss 0.13852706 - time (sec): 10.60 - samples/sec: 162.88 - lr: 0.025000
693
+ 2023-04-05 13:04:29,734 epoch 20 - iter 28/71 - loss 0.13696755 - time (sec): 14.36 - samples/sec: 161.09 - lr: 0.025000
694
+ 2023-04-05 13:04:33,339 epoch 20 - iter 35/71 - loss 0.13835721 - time (sec): 17.96 - samples/sec: 162.89 - lr: 0.025000
695
+ 2023-04-05 13:04:36,924 epoch 20 - iter 42/71 - loss 0.13924535 - time (sec): 21.55 - samples/sec: 163.31 - lr: 0.025000
696
+ 2023-04-05 13:04:40,601 epoch 20 - iter 49/71 - loss 0.13776399 - time (sec): 25.22 - samples/sec: 163.13 - lr: 0.025000
697
+ 2023-04-05 13:04:44,100 epoch 20 - iter 56/71 - loss 0.13600369 - time (sec): 28.72 - samples/sec: 163.35 - lr: 0.025000
698
+ 2023-04-05 13:04:47,772 epoch 20 - iter 63/71 - loss 0.13546503 - time (sec): 32.40 - samples/sec: 163.72 - lr: 0.025000
699
+ 2023-04-05 13:04:51,285 epoch 20 - iter 70/71 - loss 0.13381809 - time (sec): 35.91 - samples/sec: 163.86 - lr: 0.025000
700
+ 2023-04-05 13:04:52,110 ----------------------------------------------------------------------------------------------------
701
+ 2023-04-05 13:04:52,111 EPOCH 20 done: loss 0.1342 - lr 0.025000
702
+ 2023-04-05 13:05:21,762 Evaluating as a multi-label problem: False
703
+ 2023-04-05 13:05:21,781 TRAIN : loss 0.06673520058393478 - f1-score (micro avg) 0.9674
704
+ 2023-04-05 13:05:25,271 Evaluating as a multi-label problem: False
705
+ 2023-04-05 13:05:25,281 DEV : loss 0.22465617954730988 - f1-score (micro avg) 0.8848
706
+ 2023-04-05 13:05:25,286 BAD EPOCHS (no improvement): 0
707
+ 2023-04-05 13:05:25,287 ----------------------------------------------------------------------------------------------------
708
+ 2023-04-05 13:05:28,197 epoch 21 - iter 7/71 - loss 0.14072648 - time (sec): 2.91 - samples/sec: 198.61 - lr: 0.025000
709
+ 2023-04-05 13:05:31,972 epoch 21 - iter 14/71 - loss 0.12496798 - time (sec): 6.68 - samples/sec: 176.52 - lr: 0.025000
710
+ 2023-04-05 13:05:35,617 epoch 21 - iter 21/71 - loss 0.12397679 - time (sec): 10.33 - samples/sec: 172.03 - lr: 0.025000
711
+ 2023-04-05 13:05:39,209 epoch 21 - iter 28/71 - loss 0.12982706 - time (sec): 13.92 - samples/sec: 169.16 - lr: 0.025000
712
+ 2023-04-05 13:05:42,824 epoch 21 - iter 35/71 - loss 0.12812647 - time (sec): 17.54 - samples/sec: 164.51 - lr: 0.025000
713
+ 2023-04-05 13:05:46,347 epoch 21 - iter 42/71 - loss 0.12693924 - time (sec): 21.06 - samples/sec: 163.67 - lr: 0.025000
714
+ 2023-04-05 13:05:49,905 epoch 21 - iter 49/71 - loss 0.12797163 - time (sec): 24.62 - samples/sec: 165.08 - lr: 0.025000
715
+ 2023-04-05 13:05:53,577 epoch 21 - iter 56/71 - loss 0.13077894 - time (sec): 28.29 - samples/sec: 164.40 - lr: 0.025000
716
+ 2023-04-05 13:05:57,191 epoch 21 - iter 63/71 - loss 0.13144443 - time (sec): 31.90 - samples/sec: 163.71 - lr: 0.025000
717
+ 2023-04-05 13:06:00,722 epoch 21 - iter 70/71 - loss 0.13238450 - time (sec): 35.43 - samples/sec: 166.14 - lr: 0.025000
718
+ 2023-04-05 13:06:01,424 ----------------------------------------------------------------------------------------------------
719
+ 2023-04-05 13:06:01,425 EPOCH 21 done: loss 0.1318 - lr 0.025000
720
+ 2023-04-05 13:06:30,981 Evaluating as a multi-label problem: False
721
+ 2023-04-05 13:06:30,997 TRAIN : loss 0.06548392027616501 - f1-score (micro avg) 0.9687
722
+ 2023-04-05 13:06:34,514 Evaluating as a multi-label problem: False
723
+ 2023-04-05 13:06:34,524 DEV : loss 0.22177472710609436 - f1-score (micro avg) 0.8868
724
+ 2023-04-05 13:06:34,529 BAD EPOCHS (no improvement): 0
725
+ 2023-04-05 13:06:34,531 saving best model
726
+ 2023-04-05 13:06:36,118 ----------------------------------------------------------------------------------------------------
727
+ 2023-04-05 13:06:39,070 epoch 22 - iter 7/71 - loss 0.10843301 - time (sec): 2.95 - samples/sec: 193.84 - lr: 0.025000
728
+ 2023-04-05 13:06:42,690 epoch 22 - iter 14/71 - loss 0.12542959 - time (sec): 6.57 - samples/sec: 174.26 - lr: 0.025000
729
+ 2023-04-05 13:06:46,379 epoch 22 - iter 21/71 - loss 0.11852796 - time (sec): 10.26 - samples/sec: 165.88 - lr: 0.025000
730
+ 2023-04-05 13:06:50,145 epoch 22 - iter 28/71 - loss 0.11984452 - time (sec): 14.03 - samples/sec: 165.33 - lr: 0.025000
731
+ 2023-04-05 13:06:53,619 epoch 22 - iter 35/71 - loss 0.12448799 - time (sec): 17.50 - samples/sec: 166.46 - lr: 0.025000
732
+ 2023-04-05 13:06:57,251 epoch 22 - iter 42/71 - loss 0.12873390 - time (sec): 21.13 - samples/sec: 165.81 - lr: 0.025000
733
+ 2023-04-05 13:07:00,920 epoch 22 - iter 49/71 - loss 0.12402659 - time (sec): 24.80 - samples/sec: 164.83 - lr: 0.025000
734
+ 2023-04-05 13:07:04,547 epoch 22 - iter 56/71 - loss 0.12855204 - time (sec): 28.43 - samples/sec: 165.01 - lr: 0.025000
735
+ 2023-04-05 13:07:08,144 epoch 22 - iter 63/71 - loss 0.12747032 - time (sec): 32.02 - samples/sec: 165.53 - lr: 0.025000
736
+ 2023-04-05 13:07:11,868 epoch 22 - iter 70/71 - loss 0.12699574 - time (sec): 35.75 - samples/sec: 164.37 - lr: 0.025000
737
+ 2023-04-05 13:07:12,670 ----------------------------------------------------------------------------------------------------
738
+ 2023-04-05 13:07:12,671 EPOCH 22 done: loss 0.1266 - lr 0.025000
739
+ 2023-04-05 13:07:42,659 Evaluating as a multi-label problem: False
740
+ 2023-04-05 13:07:42,680 TRAIN : loss 0.06098590046167374 - f1-score (micro avg) 0.971
741
+ 2023-04-05 13:07:46,246 Evaluating as a multi-label problem: False
742
+ 2023-04-05 13:07:46,256 DEV : loss 0.21829380095005035 - f1-score (micro avg) 0.8881
743
+ 2023-04-05 13:07:46,263 BAD EPOCHS (no improvement): 0
744
+ 2023-04-05 13:07:46,264 saving best model
745
+ 2023-04-05 13:07:47,600 ----------------------------------------------------------------------------------------------------
746
+ 2023-04-05 13:07:50,798 epoch 23 - iter 7/71 - loss 0.15008917 - time (sec): 3.20 - samples/sec: 187.07 - lr: 0.025000
747
+ 2023-04-05 13:07:54,819 epoch 23 - iter 14/71 - loss 0.14280265 - time (sec): 7.22 - samples/sec: 162.93 - lr: 0.025000
748
+ 2023-04-05 13:07:58,438 epoch 23 - iter 21/71 - loss 0.13304625 - time (sec): 10.84 - samples/sec: 160.38 - lr: 0.025000
749
+ 2023-04-05 13:08:02,077 epoch 23 - iter 28/71 - loss 0.12460526 - time (sec): 14.48 - samples/sec: 161.23 - lr: 0.025000
750
+ 2023-04-05 13:08:05,787 epoch 23 - iter 35/71 - loss 0.11730006 - time (sec): 18.19 - samples/sec: 161.67 - lr: 0.025000
751
+ 2023-04-05 13:08:09,415 epoch 23 - iter 42/71 - loss 0.11912754 - time (sec): 21.81 - samples/sec: 161.64 - lr: 0.025000
752
+ 2023-04-05 13:08:13,061 epoch 23 - iter 49/71 - loss 0.12580741 - time (sec): 25.46 - samples/sec: 161.24 - lr: 0.025000
753
+ 2023-04-05 13:08:16,757 epoch 23 - iter 56/71 - loss 0.12476487 - time (sec): 29.16 - samples/sec: 161.24 - lr: 0.025000
754
+ 2023-04-05 13:08:20,720 epoch 23 - iter 63/71 - loss 0.12545874 - time (sec): 33.12 - samples/sec: 160.03 - lr: 0.025000
755
+ 2023-04-05 13:08:24,409 epoch 23 - iter 70/71 - loss 0.12915321 - time (sec): 36.81 - samples/sec: 159.80 - lr: 0.025000
756
+ 2023-04-05 13:08:25,252 ----------------------------------------------------------------------------------------------------
757
+ 2023-04-05 13:08:25,253 EPOCH 23 done: loss 0.1288 - lr 0.025000
758
+ 2023-04-05 13:08:55,761 Evaluating as a multi-label problem: False
759
+ 2023-04-05 13:08:55,783 TRAIN : loss 0.05855342745780945 - f1-score (micro avg) 0.9743
760
+ 2023-04-05 13:08:59,415 Evaluating as a multi-label problem: False
761
+ 2023-04-05 13:08:59,424 DEV : loss 0.21743902564048767 - f1-score (micro avg) 0.8901
762
+ 2023-04-05 13:08:59,429 BAD EPOCHS (no improvement): 0
763
+ 2023-04-05 13:08:59,431 saving best model
764
+ 2023-04-05 13:09:00,565 ----------------------------------------------------------------------------------------------------
765
+ 2023-04-05 13:09:03,643 epoch 24 - iter 7/71 - loss 0.09932632 - time (sec): 3.08 - samples/sec: 193.76 - lr: 0.025000
766
+ 2023-04-05 13:09:07,439 epoch 24 - iter 14/71 - loss 0.11461498 - time (sec): 6.87 - samples/sec: 173.88 - lr: 0.025000
767
+ 2023-04-05 13:09:11,176 epoch 24 - iter 21/71 - loss 0.12509619 - time (sec): 10.61 - samples/sec: 169.85 - lr: 0.025000
768
+ 2023-04-05 13:09:15,006 epoch 24 - iter 28/71 - loss 0.12628469 - time (sec): 14.44 - samples/sec: 164.13 - lr: 0.025000
769
+ 2023-04-05 13:09:18,758 epoch 24 - iter 35/71 - loss 0.12660201 - time (sec): 18.19 - samples/sec: 162.82 - lr: 0.025000
770
+ 2023-04-05 13:09:22,402 epoch 24 - iter 42/71 - loss 0.13016555 - time (sec): 21.84 - samples/sec: 163.50 - lr: 0.025000
771
+ 2023-04-05 13:09:26,014 epoch 24 - iter 49/71 - loss 0.12703872 - time (sec): 25.45 - samples/sec: 163.55 - lr: 0.025000
772
+ 2023-04-05 13:09:29,654 epoch 24 - iter 56/71 - loss 0.12356562 - time (sec): 29.09 - samples/sec: 162.13 - lr: 0.025000
773
+ 2023-04-05 13:09:33,309 epoch 24 - iter 63/71 - loss 0.12854118 - time (sec): 32.74 - samples/sec: 162.21 - lr: 0.025000
774
+ 2023-04-05 13:09:37,052 epoch 24 - iter 70/71 - loss 0.12844792 - time (sec): 36.49 - samples/sec: 161.19 - lr: 0.025000
775
+ 2023-04-05 13:09:37,848 ----------------------------------------------------------------------------------------------------
776
+ 2023-04-05 13:09:37,849 EPOCH 24 done: loss 0.1292 - lr 0.025000
777
+ 2023-04-05 13:10:08,561 Evaluating as a multi-label problem: False
778
+ 2023-04-05 13:10:08,579 TRAIN : loss 0.05762539058923721 - f1-score (micro avg) 0.9705
779
+ 2023-04-05 13:10:12,126 Evaluating as a multi-label problem: False
780
+ 2023-04-05 13:10:12,136 DEV : loss 0.21672436594963074 - f1-score (micro avg) 0.8831
781
+ 2023-04-05 13:10:12,142 BAD EPOCHS (no improvement): 1
782
+ 2023-04-05 13:10:12,143 ----------------------------------------------------------------------------------------------------
783
+ 2023-04-05 13:10:15,262 epoch 25 - iter 7/71 - loss 0.10395902 - time (sec): 3.12 - samples/sec: 192.28 - lr: 0.025000
784
+ 2023-04-05 13:10:19,012 epoch 25 - iter 14/71 - loss 0.11430174 - time (sec): 6.87 - samples/sec: 175.34 - lr: 0.025000
785
+ 2023-04-05 13:10:22,832 epoch 25 - iter 21/71 - loss 0.11574001 - time (sec): 10.69 - samples/sec: 168.99 - lr: 0.025000
786
+ 2023-04-05 13:10:26,537 epoch 25 - iter 28/71 - loss 0.12338041 - time (sec): 14.39 - samples/sec: 165.38 - lr: 0.025000
787
+ 2023-04-05 13:10:30,355 epoch 25 - iter 35/71 - loss 0.12176332 - time (sec): 18.21 - samples/sec: 161.95 - lr: 0.025000
788
+ 2023-04-05 13:10:34,020 epoch 25 - iter 42/71 - loss 0.11983290 - time (sec): 21.87 - samples/sec: 161.01 - lr: 0.025000
789
+ 2023-04-05 13:10:37,749 epoch 25 - iter 49/71 - loss 0.11602293 - time (sec): 25.60 - samples/sec: 159.00 - lr: 0.025000
790
+ 2023-04-05 13:10:41,445 epoch 25 - iter 56/71 - loss 0.11461970 - time (sec): 29.30 - samples/sec: 158.94 - lr: 0.025000
791
+ 2023-04-05 13:10:45,185 epoch 25 - iter 63/71 - loss 0.11841378 - time (sec): 33.04 - samples/sec: 159.87 - lr: 0.025000
792
+ 2023-04-05 13:10:48,840 epoch 25 - iter 70/71 - loss 0.12186183 - time (sec): 36.70 - samples/sec: 160.21 - lr: 0.025000
793
+ 2023-04-05 13:10:49,657 ----------------------------------------------------------------------------------------------------
794
+ 2023-04-05 13:10:49,659 EPOCH 25 done: loss 0.1226 - lr 0.025000
795
+ 2023-04-05 13:11:20,148 Evaluating as a multi-label problem: False
796
+ 2023-04-05 13:11:20,167 TRAIN : loss 0.05489451438188553 - f1-score (micro avg) 0.9758
797
+ 2023-04-05 13:11:23,819 Evaluating as a multi-label problem: False
798
+ 2023-04-05 13:11:23,833 DEV : loss 0.20767748355865479 - f1-score (micro avg) 0.8959
799
+ 2023-04-05 13:11:23,837 BAD EPOCHS (no improvement): 0
800
+ 2023-04-05 13:11:23,840 saving best model
801
+ 2023-04-05 13:11:25,356 ----------------------------------------------------------------------------------------------------
802
+ 2023-04-05 13:11:28,428 epoch 26 - iter 7/71 - loss 0.14960103 - time (sec): 3.07 - samples/sec: 195.35 - lr: 0.025000
803
+ 2023-04-05 13:11:32,086 epoch 26 - iter 14/71 - loss 0.13459322 - time (sec): 6.73 - samples/sec: 176.72 - lr: 0.025000
804
+ 2023-04-05 13:11:35,752 epoch 26 - iter 21/71 - loss 0.14211700 - time (sec): 10.39 - samples/sec: 168.54 - lr: 0.025000
805
+ 2023-04-05 13:11:39,370 epoch 26 - iter 28/71 - loss 0.13695156 - time (sec): 14.01 - samples/sec: 166.77 - lr: 0.025000
806
+ 2023-04-05 13:11:43,047 epoch 26 - iter 35/71 - loss 0.13044682 - time (sec): 17.69 - samples/sec: 164.66 - lr: 0.025000
807
+ 2023-04-05 13:11:46,632 epoch 26 - iter 42/71 - loss 0.12209705 - time (sec): 21.27 - samples/sec: 162.92 - lr: 0.025000
808
+ 2023-04-05 13:11:50,270 epoch 26 - iter 49/71 - loss 0.12045467 - time (sec): 24.91 - samples/sec: 163.36 - lr: 0.025000
809
+ 2023-04-05 13:11:53,914 epoch 26 - iter 56/71 - loss 0.12313593 - time (sec): 28.56 - samples/sec: 163.85 - lr: 0.025000
810
+ 2023-04-05 13:11:57,552 epoch 26 - iter 63/71 - loss 0.12808456 - time (sec): 32.20 - samples/sec: 164.06 - lr: 0.025000
811
+ 2023-04-05 13:12:01,370 epoch 26 - iter 70/71 - loss 0.13030809 - time (sec): 36.01 - samples/sec: 163.25 - lr: 0.025000
812
+ 2023-04-05 13:12:02,206 ----------------------------------------------------------------------------------------------------
813
+ 2023-04-05 13:12:02,207 EPOCH 26 done: loss 0.1298 - lr 0.025000
814
+ 2023-04-05 13:12:32,718 Evaluating as a multi-label problem: False
815
+ 2023-04-05 13:12:32,738 TRAIN : loss 0.053497862070798874 - f1-score (micro avg) 0.9731
816
+ 2023-04-05 13:12:36,375 Evaluating as a multi-label problem: False
817
+ 2023-04-05 13:12:36,388 DEV : loss 0.21015514433383942 - f1-score (micro avg) 0.8782
818
+ 2023-04-05 13:12:36,393 BAD EPOCHS (no improvement): 1
819
+ 2023-04-05 13:12:36,395 ----------------------------------------------------------------------------------------------------
820
+ 2023-04-05 13:12:39,594 epoch 27 - iter 7/71 - loss 0.14380847 - time (sec): 3.20 - samples/sec: 190.03 - lr: 0.025000
821
+ 2023-04-05 13:12:43,352 epoch 27 - iter 14/71 - loss 0.14111434 - time (sec): 6.96 - samples/sec: 173.92 - lr: 0.025000
822
+ 2023-04-05 13:12:47,121 epoch 27 - iter 21/71 - loss 0.12589741 - time (sec): 10.73 - samples/sec: 168.18 - lr: 0.025000
823
+ 2023-04-05 13:12:51,064 epoch 27 - iter 28/71 - loss 0.12663826 - time (sec): 14.67 - samples/sec: 163.67 - lr: 0.025000
824
+ 2023-04-05 13:12:54,988 epoch 27 - iter 35/71 - loss 0.12494789 - time (sec): 18.59 - samples/sec: 160.22 - lr: 0.025000
825
+ 2023-04-05 13:12:58,720 epoch 27 - iter 42/71 - loss 0.12722136 - time (sec): 22.33 - samples/sec: 158.56 - lr: 0.025000
826
+ 2023-04-05 13:13:02,624 epoch 27 - iter 49/71 - loss 0.12614770 - time (sec): 26.23 - samples/sec: 157.34 - lr: 0.025000
827
+ 2023-04-05 13:13:06,387 epoch 27 - iter 56/71 - loss 0.12524578 - time (sec): 29.99 - samples/sec: 156.77 - lr: 0.025000
828
+ 2023-04-05 13:13:10,205 epoch 27 - iter 63/71 - loss 0.12175985 - time (sec): 33.81 - samples/sec: 156.61 - lr: 0.025000
829
+ 2023-04-05 13:13:14,135 epoch 27 - iter 70/71 - loss 0.12627498 - time (sec): 37.74 - samples/sec: 155.75 - lr: 0.025000
830
+ 2023-04-05 13:13:15,072 ----------------------------------------------------------------------------------------------------
831
+ 2023-04-05 13:13:15,073 EPOCH 27 done: loss 0.1258 - lr 0.025000
832
+ 2023-04-05 13:13:46,132 Evaluating as a multi-label problem: False
833
+ 2023-04-05 13:13:46,148 TRAIN : loss 0.05426767095923424 - f1-score (micro avg) 0.9714
834
+ 2023-04-05 13:13:49,886 Evaluating as a multi-label problem: False
835
+ 2023-04-05 13:13:49,900 DEV : loss 0.21591384708881378 - f1-score (micro avg) 0.8922
836
+ 2023-04-05 13:13:49,905 BAD EPOCHS (no improvement): 2
837
+ 2023-04-05 13:13:49,907 ----------------------------------------------------------------------------------------------------
838
+ 2023-04-05 13:13:53,123 epoch 28 - iter 7/71 - loss 0.14167046 - time (sec): 3.21 - samples/sec: 183.52 - lr: 0.025000
839
+ 2023-04-05 13:13:56,811 epoch 28 - iter 14/71 - loss 0.12129582 - time (sec): 6.90 - samples/sec: 174.86 - lr: 0.025000
840
+ 2023-04-05 13:14:00,650 epoch 28 - iter 21/71 - loss 0.11742410 - time (sec): 10.74 - samples/sec: 170.64 - lr: 0.025000
841
+ 2023-04-05 13:14:04,744 epoch 28 - iter 28/71 - loss 0.11524606 - time (sec): 14.84 - samples/sec: 163.59 - lr: 0.025000
842
+ 2023-04-05 13:14:08,572 epoch 28 - iter 35/71 - loss 0.11837341 - time (sec): 18.66 - samples/sec: 161.11 - lr: 0.025000
843
+ 2023-04-05 13:14:12,510 epoch 28 - iter 42/71 - loss 0.11728808 - time (sec): 22.60 - samples/sec: 159.06 - lr: 0.025000
844
+ 2023-04-05 13:14:16,455 epoch 28 - iter 49/71 - loss 0.11521758 - time (sec): 26.55 - samples/sec: 154.90 - lr: 0.025000
845
+ 2023-04-05 13:14:20,473 epoch 28 - iter 56/71 - loss 0.11345810 - time (sec): 30.57 - samples/sec: 152.56 - lr: 0.025000
846
+ 2023-04-05 13:14:24,545 epoch 28 - iter 63/71 - loss 0.11271746 - time (sec): 34.64 - samples/sec: 152.47 - lr: 0.025000
847
+ 2023-04-05 13:14:28,525 epoch 28 - iter 70/71 - loss 0.10973027 - time (sec): 38.62 - samples/sec: 152.11 - lr: 0.025000
848
+ 2023-04-05 13:14:29,417 ----------------------------------------------------------------------------------------------------
849
+ 2023-04-05 13:14:29,418 EPOCH 28 done: loss 0.1105 - lr 0.025000
850
+ 2023-04-05 13:15:01,734 Evaluating as a multi-label problem: False
851
+ 2023-04-05 13:15:01,755 TRAIN : loss 0.05138213932514191 - f1-score (micro avg) 0.9737
852
+ 2023-04-05 13:15:05,435 Evaluating as a multi-label problem: False
853
+ 2023-04-05 13:15:05,449 DEV : loss 0.20661891996860504 - f1-score (micro avg) 0.9
854
+ 2023-04-05 13:15:05,454 BAD EPOCHS (no improvement): 0
855
+ 2023-04-05 13:15:05,456 saving best model
856
+ 2023-04-05 13:15:06,556 ----------------------------------------------------------------------------------------------------
857
+ 2023-04-05 13:15:09,647 epoch 29 - iter 7/71 - loss 0.09828948 - time (sec): 3.09 - samples/sec: 191.35 - lr: 0.025000
858
+ 2023-04-05 13:15:13,489 epoch 29 - iter 14/71 - loss 0.09758705 - time (sec): 6.93 - samples/sec: 169.24 - lr: 0.025000
859
+ 2023-04-05 13:15:17,592 epoch 29 - iter 21/71 - loss 0.10076913 - time (sec): 11.03 - samples/sec: 159.06 - lr: 0.025000
860
+ 2023-04-05 13:15:21,325 epoch 29 - iter 28/71 - loss 0.10712337 - time (sec): 14.77 - samples/sec: 159.01 - lr: 0.025000
861
+ 2023-04-05 13:15:25,125 epoch 29 - iter 35/71 - loss 0.10798193 - time (sec): 18.57 - samples/sec: 159.10 - lr: 0.025000
862
+ 2023-04-05 13:15:28,807 epoch 29 - iter 42/71 - loss 0.10437713 - time (sec): 22.25 - samples/sec: 159.56 - lr: 0.025000
863
+ 2023-04-05 13:15:32,624 epoch 29 - iter 49/71 - loss 0.10833207 - time (sec): 26.07 - samples/sec: 158.25 - lr: 0.025000
864
+ 2023-04-05 13:15:36,363 epoch 29 - iter 56/71 - loss 0.10749686 - time (sec): 29.80 - samples/sec: 158.10 - lr: 0.025000
865
+ 2023-04-05 13:15:40,194 epoch 29 - iter 63/71 - loss 0.10740193 - time (sec): 33.64 - samples/sec: 157.39 - lr: 0.025000
866
+ 2023-04-05 13:15:43,973 epoch 29 - iter 70/71 - loss 0.10763063 - time (sec): 37.41 - samples/sec: 157.24 - lr: 0.025000
867
+ 2023-04-05 13:15:44,840 ----------------------------------------------------------------------------------------------------
868
+ 2023-04-05 13:15:44,841 EPOCH 29 done: loss 0.1075 - lr 0.025000
869
+ 2023-04-05 13:16:16,518 Evaluating as a multi-label problem: False
870
+ 2023-04-05 13:16:16,538 TRAIN : loss 0.04754020646214485 - f1-score (micro avg) 0.9807
871
+ 2023-04-05 13:16:20,428 Evaluating as a multi-label problem: False
872
+ 2023-04-05 13:16:20,441 DEV : loss 0.20554056763648987 - f1-score (micro avg) 0.8831
873
+ 2023-04-05 13:16:20,447 BAD EPOCHS (no improvement): 1
874
+ 2023-04-05 13:16:20,449 ----------------------------------------------------------------------------------------------------
875
+ 2023-04-05 13:16:23,644 epoch 30 - iter 7/71 - loss 0.12214543 - time (sec): 3.19 - samples/sec: 190.64 - lr: 0.025000
876
+ 2023-04-05 13:16:27,504 epoch 30 - iter 14/71 - loss 0.10450958 - time (sec): 7.05 - samples/sec: 169.69 - lr: 0.025000
877
+ 2023-04-05 13:16:31,376 epoch 30 - iter 21/71 - loss 0.10887640 - time (sec): 10.93 - samples/sec: 163.00 - lr: 0.025000
878
+ 2023-04-05 13:16:35,226 epoch 30 - iter 28/71 - loss 0.10731823 - time (sec): 14.78 - samples/sec: 160.73 - lr: 0.025000
879
+ 2023-04-05 13:16:39,102 epoch 30 - iter 35/71 - loss 0.10467509 - time (sec): 18.65 - samples/sec: 158.64 - lr: 0.025000
880
+ 2023-04-05 13:16:43,120 epoch 30 - iter 42/71 - loss 0.11094270 - time (sec): 22.67 - samples/sec: 156.95 - lr: 0.025000
881
+ 2023-04-05 13:16:46,982 epoch 30 - iter 49/71 - loss 0.10634089 - time (sec): 26.53 - samples/sec: 157.10 - lr: 0.025000
882
+ 2023-04-05 13:16:51,073 epoch 30 - iter 56/71 - loss 0.10685095 - time (sec): 30.62 - samples/sec: 154.23 - lr: 0.025000
883
+ 2023-04-05 13:16:55,184 epoch 30 - iter 63/71 - loss 0.11220058 - time (sec): 34.73 - samples/sec: 152.73 - lr: 0.025000
884
+ 2023-04-05 13:16:59,355 epoch 30 - iter 70/71 - loss 0.11420165 - time (sec): 38.91 - samples/sec: 151.32 - lr: 0.025000
885
+ 2023-04-05 13:17:00,290 ----------------------------------------------------------------------------------------------------
886
+ 2023-04-05 13:17:00,291 EPOCH 30 done: loss 0.1138 - lr 0.025000
887
+ 2023-04-05 13:17:33,935 Evaluating as a multi-label problem: False
888
+ 2023-04-05 13:17:33,955 TRAIN : loss 0.049051374197006226 - f1-score (micro avg) 0.9765
889
+ 2023-04-05 13:17:37,925 Evaluating as a multi-label problem: False
890
+ 2023-04-05 13:17:37,937 DEV : loss 0.21063490211963654 - f1-score (micro avg) 0.8989
891
+ 2023-04-05 13:17:37,943 BAD EPOCHS (no improvement): 2
892
+ 2023-04-05 13:17:37,944 ----------------------------------------------------------------------------------------------------
893
+ 2023-04-05 13:17:41,197 epoch 31 - iter 7/71 - loss 0.12085872 - time (sec): 3.25 - samples/sec: 182.37 - lr: 0.025000
894
+ 2023-04-05 13:17:44,963 epoch 31 - iter 14/71 - loss 0.12129786 - time (sec): 7.02 - samples/sec: 164.16 - lr: 0.025000
895
+ 2023-04-05 13:17:48,794 epoch 31 - iter 21/71 - loss 0.11010256 - time (sec): 10.85 - samples/sec: 160.39 - lr: 0.025000
896
+ 2023-04-05 13:17:52,660 epoch 31 - iter 28/71 - loss 0.11869391 - time (sec): 14.71 - samples/sec: 157.67 - lr: 0.025000
897
+ 2023-04-05 13:17:56,463 epoch 31 - iter 35/71 - loss 0.11978621 - time (sec): 18.52 - samples/sec: 157.63 - lr: 0.025000
898
+ 2023-04-05 13:18:00,669 epoch 31 - iter 42/71 - loss 0.12158608 - time (sec): 22.72 - samples/sec: 153.76 - lr: 0.025000
899
+ 2023-04-05 13:18:04,928 epoch 31 - iter 49/71 - loss 0.12205383 - time (sec): 26.98 - samples/sec: 151.10 - lr: 0.025000
900
+ 2023-04-05 13:18:08,749 epoch 31 - iter 56/71 - loss 0.12407999 - time (sec): 30.80 - samples/sec: 152.35 - lr: 0.025000
901
+ 2023-04-05 13:18:12,993 epoch 31 - iter 63/71 - loss 0.12182435 - time (sec): 35.05 - samples/sec: 151.14 - lr: 0.025000
902
+ 2023-04-05 13:18:17,045 epoch 31 - iter 70/71 - loss 0.11982468 - time (sec): 39.10 - samples/sec: 150.39 - lr: 0.025000
903
+ 2023-04-05 13:18:17,906 ----------------------------------------------------------------------------------------------------
904
+ 2023-04-05 13:18:17,907 EPOCH 31 done: loss 0.1199 - lr 0.025000
905
+ 2023-04-05 13:18:50,399 Evaluating as a multi-label problem: False
906
+ 2023-04-05 13:18:50,415 TRAIN : loss 0.04879188537597656 - f1-score (micro avg) 0.9767
907
+ 2023-04-05 13:18:54,298 Evaluating as a multi-label problem: False
908
+ 2023-04-05 13:18:54,311 DEV : loss 0.21518820524215698 - f1-score (micro avg) 0.8856
909
+ 2023-04-05 13:18:54,319 BAD EPOCHS (no improvement): 3
910
+ 2023-04-05 13:18:54,320 ----------------------------------------------------------------------------------------------------
911
+ 2023-04-05 13:18:57,672 epoch 32 - iter 7/71 - loss 0.09686100 - time (sec): 3.35 - samples/sec: 182.98 - lr: 0.025000
912
+ 2023-04-05 13:19:01,687 epoch 32 - iter 14/71 - loss 0.11264203 - time (sec): 7.36 - samples/sec: 163.07 - lr: 0.025000
913
+ 2023-04-05 13:19:05,453 epoch 32 - iter 21/71 - loss 0.11587441 - time (sec): 11.13 - samples/sec: 160.90 - lr: 0.025000
914
+ 2023-04-05 13:19:09,495 epoch 32 - iter 28/71 - loss 0.11953699 - time (sec): 15.17 - samples/sec: 154.68 - lr: 0.025000
915
+ 2023-04-05 13:19:13,613 epoch 32 - iter 35/71 - loss 0.11792902 - time (sec): 19.29 - samples/sec: 151.52 - lr: 0.025000
916
+ 2023-04-05 13:19:17,553 epoch 32 - iter 42/71 - loss 0.11791943 - time (sec): 23.23 - samples/sec: 151.60 - lr: 0.025000
917
+ 2023-04-05 13:19:21,512 epoch 32 - iter 49/71 - loss 0.11984092 - time (sec): 27.19 - samples/sec: 151.85 - lr: 0.025000
918
+ 2023-04-05 13:19:25,273 epoch 32 - iter 56/71 - loss 0.12028672 - time (sec): 30.95 - samples/sec: 152.37 - lr: 0.025000
919
+ 2023-04-05 13:19:29,003 epoch 32 - iter 63/71 - loss 0.11505568 - time (sec): 34.68 - samples/sec: 153.28 - lr: 0.025000
920
+ 2023-04-05 13:19:32,783 epoch 32 - iter 70/71 - loss 0.11360413 - time (sec): 38.46 - samples/sec: 152.91 - lr: 0.025000
921
+ 2023-04-05 13:19:33,667 ----------------------------------------------------------------------------------------------------
922
+ 2023-04-05 13:19:33,668 EPOCH 32 done: loss 0.1136 - lr 0.025000
923
+ 2023-04-05 13:20:05,299 Evaluating as a multi-label problem: False
924
+ 2023-04-05 13:20:05,317 TRAIN : loss 0.04602975398302078 - f1-score (micro avg) 0.9764
925
+ 2023-04-05 13:20:08,959 Evaluating as a multi-label problem: False
926
+ 2023-04-05 13:20:08,971 DEV : loss 0.2196006327867508 - f1-score (micro avg) 0.8831
927
+ 2023-04-05 13:20:08,977 Epoch 32: reducing learning rate of group 0 to 1.2500e-02.
928
+ 2023-04-05 13:20:08,978 BAD EPOCHS (no improvement): 4
929
+ 2023-04-05 13:20:08,980 ----------------------------------------------------------------------------------------------------
930
+ 2023-04-05 13:20:12,222 epoch 33 - iter 7/71 - loss 0.11581283 - time (sec): 3.24 - samples/sec: 177.97 - lr: 0.012500
931
+ 2023-04-05 13:20:16,008 epoch 33 - iter 14/71 - loss 0.11978144 - time (sec): 7.03 - samples/sec: 169.61 - lr: 0.012500
932
+ 2023-04-05 13:20:19,936 epoch 33 - iter 21/71 - loss 0.11537349 - time (sec): 10.96 - samples/sec: 159.65 - lr: 0.012500
933
+ 2023-04-05 13:20:23,770 epoch 33 - iter 28/71 - loss 0.10736190 - time (sec): 14.79 - samples/sec: 160.05 - lr: 0.012500
934
+ 2023-04-05 13:20:27,878 epoch 33 - iter 35/71 - loss 0.11021785 - time (sec): 18.90 - samples/sec: 156.10 - lr: 0.012500
935
+ 2023-04-05 13:20:31,966 epoch 33 - iter 42/71 - loss 0.10713205 - time (sec): 22.99 - samples/sec: 152.88 - lr: 0.012500
936
+ 2023-04-05 13:20:35,718 epoch 33 - iter 49/71 - loss 0.10934260 - time (sec): 26.74 - samples/sec: 153.53 - lr: 0.012500
937
+ 2023-04-05 13:20:39,467 epoch 33 - iter 56/71 - loss 0.10654440 - time (sec): 30.49 - samples/sec: 153.51 - lr: 0.012500
938
+ 2023-04-05 13:20:43,388 epoch 33 - iter 63/71 - loss 0.10765648 - time (sec): 34.41 - samples/sec: 153.34 - lr: 0.012500
939
+ 2023-04-05 13:20:47,348 epoch 33 - iter 70/71 - loss 0.11051452 - time (sec): 38.37 - samples/sec: 153.15 - lr: 0.012500
940
+ 2023-04-05 13:20:48,204 ----------------------------------------------------------------------------------------------------
941
+ 2023-04-05 13:20:48,205 EPOCH 33 done: loss 0.1101 - lr 0.012500
942
+ 2023-04-05 13:21:21,733 Evaluating as a multi-label problem: False
943
+ 2023-04-05 13:21:21,755 TRAIN : loss 0.04103841260075569 - f1-score (micro avg) 0.9818
944
+ 2023-04-05 13:21:25,443 Evaluating as a multi-label problem: False
945
+ 2023-04-05 13:21:25,453 DEV : loss 0.19789864122867584 - f1-score (micro avg) 0.8848
946
+ 2023-04-05 13:21:25,460 BAD EPOCHS (no improvement): 1
947
+ 2023-04-05 13:21:25,462 ----------------------------------------------------------------------------------------------------
948
+ 2023-04-05 13:21:28,952 epoch 34 - iter 7/71 - loss 0.11239772 - time (sec): 3.49 - samples/sec: 171.41 - lr: 0.012500
949
+ 2023-04-05 13:21:33,108 epoch 34 - iter 14/71 - loss 0.09980787 - time (sec): 7.64 - samples/sec: 155.67 - lr: 0.012500
950
+ 2023-04-05 13:21:37,298 epoch 34 - iter 21/71 - loss 0.09221232 - time (sec): 11.84 - samples/sec: 150.74 - lr: 0.012500
951
+ 2023-04-05 13:21:41,356 epoch 34 - iter 28/71 - loss 0.09881858 - time (sec): 15.89 - samples/sec: 148.74 - lr: 0.012500
952
+ 2023-04-05 13:21:45,307 epoch 34 - iter 35/71 - loss 0.10014088 - time (sec): 19.84 - samples/sec: 147.75 - lr: 0.012500
953
+ 2023-04-05 13:21:49,356 epoch 34 - iter 42/71 - loss 0.09659220 - time (sec): 23.89 - samples/sec: 147.12 - lr: 0.012500
954
+ 2023-04-05 13:21:53,323 epoch 34 - iter 49/71 - loss 0.09546800 - time (sec): 27.86 - samples/sec: 147.34 - lr: 0.012500
955
+ 2023-04-05 13:21:57,491 epoch 34 - iter 56/71 - loss 0.09803050 - time (sec): 32.03 - samples/sec: 146.53 - lr: 0.012500
956
+ 2023-04-05 13:22:01,642 epoch 34 - iter 63/71 - loss 0.10373384 - time (sec): 36.18 - samples/sec: 147.24 - lr: 0.012500
957
+ 2023-04-05 13:22:05,766 epoch 34 - iter 70/71 - loss 0.10094599 - time (sec): 40.30 - samples/sec: 145.94 - lr: 0.012500
958
+ 2023-04-05 13:22:06,689 ----------------------------------------------------------------------------------------------------
959
+ 2023-04-05 13:22:06,690 EPOCH 34 done: loss 0.1011 - lr 0.012500
960
+ 2023-04-05 13:22:39,510 Evaluating as a multi-label problem: False
961
+ 2023-04-05 13:22:39,527 TRAIN : loss 0.04190446436405182 - f1-score (micro avg) 0.9799
962
+ 2023-04-05 13:22:43,323 Evaluating as a multi-label problem: False
963
+ 2023-04-05 13:22:43,335 DEV : loss 0.20444391667842865 - f1-score (micro avg) 0.8951
964
+ 2023-04-05 13:22:43,341 BAD EPOCHS (no improvement): 2
965
+ 2023-04-05 13:22:43,342 ----------------------------------------------------------------------------------------------------
966
+ 2023-04-05 13:22:46,687 epoch 35 - iter 7/71 - loss 0.12740153 - time (sec): 3.34 - samples/sec: 179.39 - lr: 0.012500
967
+ 2023-04-05 13:22:50,605 epoch 35 - iter 14/71 - loss 0.09989261 - time (sec): 7.26 - samples/sec: 162.61 - lr: 0.012500
968
+ 2023-04-05 13:22:54,708 epoch 35 - iter 21/71 - loss 0.09380928 - time (sec): 11.37 - samples/sec: 158.03 - lr: 0.012500
969
+ 2023-04-05 13:22:58,737 epoch 35 - iter 28/71 - loss 0.10283196 - time (sec): 15.39 - samples/sec: 154.02 - lr: 0.012500
970
+ 2023-04-05 13:23:02,706 epoch 35 - iter 35/71 - loss 0.10197240 - time (sec): 19.36 - samples/sec: 153.54 - lr: 0.012500
971
+ 2023-04-05 13:23:06,476 epoch 35 - iter 42/71 - loss 0.09712460 - time (sec): 23.13 - samples/sec: 152.98 - lr: 0.012500
972
+ 2023-04-05 13:23:10,261 epoch 35 - iter 49/71 - loss 0.09975067 - time (sec): 26.92 - samples/sec: 152.68 - lr: 0.012500
973
+ 2023-04-05 13:23:14,485 epoch 35 - iter 56/71 - loss 0.09870986 - time (sec): 31.14 - samples/sec: 150.82 - lr: 0.012500
974
+ 2023-04-05 13:23:18,733 epoch 35 - iter 63/71 - loss 0.10022879 - time (sec): 35.39 - samples/sec: 149.50 - lr: 0.012500
975
+ 2023-04-05 13:23:22,647 epoch 35 - iter 70/71 - loss 0.10151940 - time (sec): 39.30 - samples/sec: 149.76 - lr: 0.012500
976
+ 2023-04-05 13:23:23,519 ----------------------------------------------------------------------------------------------------
977
+ 2023-04-05 13:23:23,520 EPOCH 35 done: loss 0.1017 - lr 0.012500
978
+ 2023-04-05 13:23:56,251 Evaluating as a multi-label problem: False
979
+ 2023-04-05 13:23:56,267 TRAIN : loss 0.03953753411769867 - f1-score (micro avg) 0.9826
980
+ 2023-04-05 13:24:00,070 Evaluating as a multi-label problem: False
981
+ 2023-04-05 13:24:00,083 DEV : loss 0.20463646948337555 - f1-score (micro avg) 0.8939
982
+ 2023-04-05 13:24:00,089 BAD EPOCHS (no improvement): 3
983
+ 2023-04-05 13:24:00,091 ----------------------------------------------------------------------------------------------------
984
+ 2023-04-05 13:24:03,512 epoch 36 - iter 7/71 - loss 0.09733479 - time (sec): 3.42 - samples/sec: 169.82 - lr: 0.012500
985
+ 2023-04-05 13:24:07,519 epoch 36 - iter 14/71 - loss 0.09929472 - time (sec): 7.43 - samples/sec: 155.63 - lr: 0.012500
986
+ 2023-04-05 13:24:11,244 epoch 36 - iter 21/71 - loss 0.09236051 - time (sec): 11.15 - samples/sec: 156.73 - lr: 0.012500
987
+ 2023-04-05 13:24:14,927 epoch 36 - iter 28/71 - loss 0.09717889 - time (sec): 14.84 - samples/sec: 158.54 - lr: 0.012500
988
+ 2023-04-05 13:24:18,720 epoch 36 - iter 35/71 - loss 0.09814890 - time (sec): 18.63 - samples/sec: 157.23 - lr: 0.012500
989
+ 2023-04-05 13:24:22,810 epoch 36 - iter 42/71 - loss 0.09793933 - time (sec): 22.72 - samples/sec: 154.50 - lr: 0.012500
990
+ 2023-04-05 13:24:26,643 epoch 36 - iter 49/71 - loss 0.09698078 - time (sec): 26.55 - samples/sec: 154.15 - lr: 0.012500
991
+ 2023-04-05 13:24:30,395 epoch 36 - iter 56/71 - loss 0.09628057 - time (sec): 30.30 - samples/sec: 155.19 - lr: 0.012500
992
+ 2023-04-05 13:24:34,471 epoch 36 - iter 63/71 - loss 0.09878813 - time (sec): 34.38 - samples/sec: 154.22 - lr: 0.012500
993
+ 2023-04-05 13:24:38,425 epoch 36 - iter 70/71 - loss 0.10012306 - time (sec): 38.33 - samples/sec: 153.20 - lr: 0.012500
994
+ 2023-04-05 13:24:39,322 ----------------------------------------------------------------------------------------------------
995
+ 2023-04-05 13:24:39,323 EPOCH 36 done: loss 0.1011 - lr 0.012500
996
+ 2023-04-05 13:25:11,618 Evaluating as a multi-label problem: False
997
+ 2023-04-05 13:25:11,634 TRAIN : loss 0.03854582458734512 - f1-score (micro avg) 0.9824
998
+ 2023-04-05 13:25:15,735 Evaluating as a multi-label problem: False
999
+ 2023-04-05 13:25:15,747 DEV : loss 0.20157837867736816 - f1-score (micro avg) 0.8885
1000
+ 2023-04-05 13:25:15,751 Epoch 36: reducing learning rate of group 0 to 6.2500e-03.
1001
+ 2023-04-05 13:25:15,752 BAD EPOCHS (no improvement): 4
1002
+ 2023-04-05 13:25:15,754 ----------------------------------------------------------------------------------------------------
1003
+ 2023-04-05 13:25:19,364 epoch 37 - iter 7/71 - loss 0.10420912 - time (sec): 3.61 - samples/sec: 165.95 - lr: 0.006250
1004
+ 2023-04-05 13:25:23,503 epoch 37 - iter 14/71 - loss 0.09970839 - time (sec): 7.75 - samples/sec: 153.57 - lr: 0.006250
1005
+ 2023-04-05 13:25:27,746 epoch 37 - iter 21/71 - loss 0.09589796 - time (sec): 11.99 - samples/sec: 149.02 - lr: 0.006250
1006
+ 2023-04-05 13:25:31,896 epoch 37 - iter 28/71 - loss 0.09015041 - time (sec): 16.14 - samples/sec: 145.40 - lr: 0.006250
1007
+ 2023-04-05 13:25:35,703 epoch 37 - iter 35/71 - loss 0.09060108 - time (sec): 19.95 - samples/sec: 145.42 - lr: 0.006250
1008
+ 2023-04-05 13:25:39,605 epoch 37 - iter 42/71 - loss 0.08912907 - time (sec): 23.85 - samples/sec: 145.99 - lr: 0.006250
1009
+ 2023-04-05 13:25:43,386 epoch 37 - iter 49/71 - loss 0.09042087 - time (sec): 27.63 - samples/sec: 148.27 - lr: 0.006250
1010
+ 2023-04-05 13:25:47,332 epoch 37 - iter 56/71 - loss 0.08986784 - time (sec): 31.58 - samples/sec: 148.59 - lr: 0.006250
1011
+ 2023-04-05 13:25:51,296 epoch 37 - iter 63/71 - loss 0.09147423 - time (sec): 35.54 - samples/sec: 148.87 - lr: 0.006250
1012
+ 2023-04-05 13:25:55,383 epoch 37 - iter 70/71 - loss 0.09523919 - time (sec): 39.63 - samples/sec: 148.30 - lr: 0.006250
1013
+ 2023-04-05 13:25:56,305 ----------------------------------------------------------------------------------------------------
1014
+ 2023-04-05 13:25:56,307 EPOCH 37 done: loss 0.0947 - lr 0.006250
1015
+ 2023-04-05 13:26:28,852 Evaluating as a multi-label problem: False
1016
+ 2023-04-05 13:26:28,871 TRAIN : loss 0.03739665448665619 - f1-score (micro avg) 0.9842
1017
+ 2023-04-05 13:26:32,646 Evaluating as a multi-label problem: False
1018
+ 2023-04-05 13:26:32,656 DEV : loss 0.20396985113620758 - f1-score (micro avg) 0.8972
1019
+ 2023-04-05 13:26:32,663 BAD EPOCHS (no improvement): 1
1020
+ 2023-04-05 13:26:32,664 ----------------------------------------------------------------------------------------------------
1021
+ 2023-04-05 13:26:35,733 epoch 38 - iter 7/71 - loss 0.05286205 - time (sec): 3.07 - samples/sec: 184.49 - lr: 0.006250
1022
+ 2023-04-05 13:26:39,585 epoch 38 - iter 14/71 - loss 0.07231140 - time (sec): 6.92 - samples/sec: 168.21 - lr: 0.006250
1023
+ 2023-04-05 13:26:43,417 epoch 38 - iter 21/71 - loss 0.07400087 - time (sec): 10.75 - samples/sec: 163.88 - lr: 0.006250
1024
+ 2023-04-05 13:26:47,164 epoch 38 - iter 28/71 - loss 0.08128836 - time (sec): 14.50 - samples/sec: 161.87 - lr: 0.006250
1025
+ 2023-04-05 13:26:50,941 epoch 38 - iter 35/71 - loss 0.08052234 - time (sec): 18.28 - samples/sec: 161.08 - lr: 0.006250
1026
+ 2023-04-05 13:26:54,710 epoch 38 - iter 42/71 - loss 0.08768679 - time (sec): 22.04 - samples/sec: 158.99 - lr: 0.006250
1027
+ 2023-04-05 13:26:58,468 epoch 38 - iter 49/71 - loss 0.08866679 - time (sec): 25.80 - samples/sec: 159.44 - lr: 0.006250
1028
+ 2023-04-05 13:27:02,174 epoch 38 - iter 56/71 - loss 0.09098401 - time (sec): 29.51 - samples/sec: 159.68 - lr: 0.006250
1029
+ 2023-04-05 13:27:06,037 epoch 38 - iter 63/71 - loss 0.09004818 - time (sec): 33.37 - samples/sec: 158.34 - lr: 0.006250
1030
+ 2023-04-05 13:27:09,739 epoch 38 - iter 70/71 - loss 0.09029994 - time (sec): 37.07 - samples/sec: 158.63 - lr: 0.006250
1031
+ 2023-04-05 13:27:10,609 ----------------------------------------------------------------------------------------------------
1032
+ 2023-04-05 13:27:10,610 EPOCH 38 done: loss 0.0905 - lr 0.006250
1033
+ 2023-04-05 13:27:41,464 Evaluating as a multi-label problem: False
1034
+ 2023-04-05 13:27:41,484 TRAIN : loss 0.03672816604375839 - f1-score (micro avg) 0.984
1035
+ 2023-04-05 13:27:45,153 Evaluating as a multi-label problem: False
1036
+ 2023-04-05 13:27:45,162 DEV : loss 0.1981133222579956 - f1-score (micro avg) 0.9026
1037
+ 2023-04-05 13:27:45,168 BAD EPOCHS (no improvement): 0
1038
+ 2023-04-05 13:27:45,170 saving best model
1039
+ 2023-04-05 13:27:46,871 ----------------------------------------------------------------------------------------------------
1040
+ 2023-04-05 13:27:50,071 epoch 39 - iter 7/71 - loss 0.09828755 - time (sec): 3.20 - samples/sec: 178.58 - lr: 0.006250
1041
+ 2023-04-05 13:27:53,862 epoch 39 - iter 14/71 - loss 0.08947759 - time (sec): 6.99 - samples/sec: 166.98 - lr: 0.006250
1042
+ 2023-04-05 13:27:57,588 epoch 39 - iter 21/71 - loss 0.09770654 - time (sec): 10.71 - samples/sec: 164.63 - lr: 0.006250
1043
+ 2023-04-05 13:28:01,143 epoch 39 - iter 28/71 - loss 0.09769007 - time (sec): 14.27 - samples/sec: 162.79 - lr: 0.006250
1044
+ 2023-04-05 13:28:04,815 epoch 39 - iter 35/71 - loss 0.09508572 - time (sec): 17.94 - samples/sec: 161.85 - lr: 0.006250
1045
+ 2023-04-05 13:28:08,491 epoch 39 - iter 42/71 - loss 0.09602471 - time (sec): 21.62 - samples/sec: 160.52 - lr: 0.006250
1046
+ 2023-04-05 13:28:12,205 epoch 39 - iter 49/71 - loss 0.09330580 - time (sec): 25.33 - samples/sec: 158.85 - lr: 0.006250
1047
+ 2023-04-05 13:28:15,838 epoch 39 - iter 56/71 - loss 0.09344274 - time (sec): 28.96 - samples/sec: 160.30 - lr: 0.006250
1048
+ 2023-04-05 13:28:19,582 epoch 39 - iter 63/71 - loss 0.09424575 - time (sec): 32.71 - samples/sec: 161.12 - lr: 0.006250
1049
+ 2023-04-05 13:28:23,266 epoch 39 - iter 70/71 - loss 0.09570928 - time (sec): 36.39 - samples/sec: 161.54 - lr: 0.006250
1050
+ 2023-04-05 13:28:24,136 ----------------------------------------------------------------------------------------------------
1051
+ 2023-04-05 13:28:24,137 EPOCH 39 done: loss 0.0955 - lr 0.006250
1052
+ 2023-04-05 13:28:55,230 Evaluating as a multi-label problem: False
1053
+ 2023-04-05 13:28:55,254 TRAIN : loss 0.03638274222612381 - f1-score (micro avg) 0.984
1054
+ 2023-04-05 13:28:59,307 Evaluating as a multi-label problem: False
1055
+ 2023-04-05 13:28:59,322 DEV : loss 0.20241594314575195 - f1-score (micro avg) 0.8935
1056
+ 2023-04-05 13:28:59,327 BAD EPOCHS (no improvement): 1
1057
+ 2023-04-05 13:28:59,329 ----------------------------------------------------------------------------------------------------
1058
+ 2023-04-05 13:29:02,529 epoch 40 - iter 7/71 - loss 0.10842314 - time (sec): 3.20 - samples/sec: 172.19 - lr: 0.006250
1059
+ 2023-04-05 13:29:06,624 epoch 40 - iter 14/71 - loss 0.10954390 - time (sec): 7.29 - samples/sec: 154.36 - lr: 0.006250
1060
+ 2023-04-05 13:29:10,512 epoch 40 - iter 21/71 - loss 0.11209631 - time (sec): 11.18 - samples/sec: 155.68 - lr: 0.006250
1061
+ 2023-04-05 13:29:14,381 epoch 40 - iter 28/71 - loss 0.10212289 - time (sec): 15.05 - samples/sec: 153.20 - lr: 0.006250
1062
+ 2023-04-05 13:29:18,226 epoch 40 - iter 35/71 - loss 0.09686748 - time (sec): 18.90 - samples/sec: 154.84 - lr: 0.006250
1063
+ 2023-04-05 13:29:22,011 epoch 40 - iter 42/71 - loss 0.09428059 - time (sec): 22.68 - samples/sec: 155.41 - lr: 0.006250
1064
+ 2023-04-05 13:29:25,738 epoch 40 - iter 49/71 - loss 0.09207549 - time (sec): 26.41 - samples/sec: 156.04 - lr: 0.006250
1065
+ 2023-04-05 13:29:29,548 epoch 40 - iter 56/71 - loss 0.09320325 - time (sec): 30.22 - samples/sec: 156.72 - lr: 0.006250
1066
+ 2023-04-05 13:29:33,362 epoch 40 - iter 63/71 - loss 0.09517989 - time (sec): 34.03 - samples/sec: 155.61 - lr: 0.006250
1067
+ 2023-04-05 13:29:37,104 epoch 40 - iter 70/71 - loss 0.09762175 - time (sec): 37.78 - samples/sec: 155.84 - lr: 0.006250
1068
+ 2023-04-05 13:29:37,944 ----------------------------------------------------------------------------------------------------
1069
+ 2023-04-05 13:29:37,944 EPOCH 40 done: loss 0.0976 - lr 0.006250
1070
+ 2023-04-05 13:30:09,218 Evaluating as a multi-label problem: False
1071
+ 2023-04-05 13:30:09,236 TRAIN : loss 0.03600259870290756 - f1-score (micro avg) 0.9838
1072
+ 2023-04-05 13:30:13,253 Evaluating as a multi-label problem: False
1073
+ 2023-04-05 13:30:13,263 DEV : loss 0.20378927886486053 - f1-score (micro avg) 0.8939
1074
+ 2023-04-05 13:30:13,269 BAD EPOCHS (no improvement): 2
1075
+ 2023-04-05 13:30:14,367 ----------------------------------------------------------------------------------------------------
1076
+ 2023-04-05 13:30:18,413 SequenceTagger predicts: Dictionary with 11 tags: O, S-PERSON, B-PERSON, E-PERSON, I-PERSON, S-ORG, B-ORG, E-ORG, I-ORG, <START>, <STOP>
1077
+ 2023-04-05 13:30:24,734 Evaluating as a multi-label problem: False
1078
+ 2023-04-05 13:30:24,746 0.9125 0.905 0.9087 0.8456
1079
+ 2023-04-05 13:30:24,747
1080
+ Results:
1081
+ - F-score (micro) 0.9087
1082
+ - F-score (macro) 0.8766
1083
+ - Accuracy 0.8456
1084
+
1085
+ By class:
1086
+ precision recall f1-score support
1087
+
1088
+ PERSON 0.9391 0.9443 0.9417 359
1089
+ ORG 0.8319 0.7920 0.8115 125
1090
+
1091
+ micro avg 0.9125 0.9050 0.9087 484
1092
+ macro avg 0.8855 0.8681 0.8766 484
1093
+ weighted avg 0.9114 0.9050 0.9080 484
1094
+
1095
+ 2023-04-05 13:30:24,748 ----------------------------------------------------------------------------------------------------