Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
1b26ba7
•
1 Parent(s): a6fcd10

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697663731.46dc0c540dd0.3341.11 +3 -0
  5. test.tsv +0 -0
  6. training.log +246 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de13e0ed63eb8f83594e3b5e671ebe8289c177c488f05eb4f0a60ec21049f7cc
3
+ size 19045922
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 21:15:55 0.0000 0.9409 0.2053 0.3702 0.2421 0.2927 0.1843
3
+ 2 21:16:20 0.0000 0.2593 0.1671 0.3948 0.4287 0.4111 0.2785
4
+ 3 21:16:44 0.0000 0.2139 0.1490 0.5179 0.5079 0.5128 0.3680
5
+ 4 21:17:08 0.0000 0.1897 0.1412 0.5270 0.5633 0.5446 0.4003
6
+ 5 21:17:32 0.0000 0.1701 0.1358 0.5266 0.5939 0.5582 0.4167
7
+ 6 21:17:57 0.0000 0.1597 0.1329 0.5525 0.6075 0.5787 0.4377
8
+ 7 21:18:21 0.0000 0.1508 0.1309 0.5811 0.6120 0.5961 0.4520
9
+ 8 21:18:45 0.0000 0.1451 0.1317 0.5810 0.6165 0.5982 0.4561
10
+ 9 21:19:10 0.0000 0.1393 0.1306 0.5811 0.6324 0.6056 0.4647
11
+ 10 21:19:34 0.0000 0.1383 0.1323 0.5851 0.6222 0.6031 0.4614
runs/events.out.tfevents.1697663731.46dc0c540dd0.3341.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73537e1e2e2edddebd6f2cc143a10124b00e40f34da252ff8af877cdf9e86a13
3
+ size 556612
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 21:15:31,626 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 21:15:31,626 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 21:15:31,627 MultiCorpus: 7936 train + 992 dev + 992 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
53
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 21:15:31,627 Train: 7936 sentences
55
+ 2023-10-18 21:15:31,627 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 21:15:31,627 Training Params:
58
+ 2023-10-18 21:15:31,627 - learning_rate: "5e-05"
59
+ 2023-10-18 21:15:31,627 - mini_batch_size: "8"
60
+ 2023-10-18 21:15:31,627 - max_epochs: "10"
61
+ 2023-10-18 21:15:31,627 - shuffle: "True"
62
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 21:15:31,627 Plugins:
64
+ 2023-10-18 21:15:31,627 - TensorboardLogger
65
+ 2023-10-18 21:15:31,627 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 21:15:31,627 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 21:15:31,627 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 21:15:31,627 Computation:
71
+ 2023-10-18 21:15:31,627 - compute on device: cuda:0
72
+ 2023-10-18 21:15:31,627 - embedding storage: none
73
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 21:15:31,627 Model training base path: "hmbench-icdar/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
75
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 21:15:31,627 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 21:15:31,628 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 21:15:33,871 epoch 1 - iter 99/992 - loss 3.00035148 - time (sec): 2.24 - samples/sec: 7504.52 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-18 21:15:36,122 epoch 1 - iter 198/992 - loss 2.65024826 - time (sec): 4.49 - samples/sec: 7377.52 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-18 21:15:38,453 epoch 1 - iter 297/992 - loss 2.14493534 - time (sec): 6.83 - samples/sec: 7396.19 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-18 21:15:40,761 epoch 1 - iter 396/992 - loss 1.77405864 - time (sec): 9.13 - samples/sec: 7228.32 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-18 21:15:42,985 epoch 1 - iter 495/992 - loss 1.52199448 - time (sec): 11.36 - samples/sec: 7224.42 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-18 21:15:45,245 epoch 1 - iter 594/992 - loss 1.33697784 - time (sec): 13.62 - samples/sec: 7236.13 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-18 21:15:47,493 epoch 1 - iter 693/992 - loss 1.20151051 - time (sec): 15.87 - samples/sec: 7232.93 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-18 21:15:49,727 epoch 1 - iter 792/992 - loss 1.09367873 - time (sec): 18.10 - samples/sec: 7246.04 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-18 21:15:51,954 epoch 1 - iter 891/992 - loss 1.00765596 - time (sec): 20.33 - samples/sec: 7260.10 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-18 21:15:54,200 epoch 1 - iter 990/992 - loss 0.94167931 - time (sec): 22.57 - samples/sec: 7252.23 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-18 21:15:54,247 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 21:15:54,247 EPOCH 1 done: loss 0.9409 - lr: 0.000050
90
+ 2023-10-18 21:15:55,813 DEV : loss 0.20528535544872284 - f1-score (micro avg) 0.2927
91
+ 2023-10-18 21:15:55,832 saving best model
92
+ 2023-10-18 21:15:55,870 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 21:15:58,007 epoch 2 - iter 99/992 - loss 0.29525902 - time (sec): 2.14 - samples/sec: 7453.35 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-18 21:16:00,244 epoch 2 - iter 198/992 - loss 0.30020282 - time (sec): 4.37 - samples/sec: 7681.68 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-18 21:16:02,448 epoch 2 - iter 297/992 - loss 0.29276113 - time (sec): 6.58 - samples/sec: 7608.66 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-18 21:16:04,695 epoch 2 - iter 396/992 - loss 0.28180797 - time (sec): 8.82 - samples/sec: 7520.52 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-18 21:16:06,951 epoch 2 - iter 495/992 - loss 0.27842310 - time (sec): 11.08 - samples/sec: 7417.17 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-18 21:16:09,180 epoch 2 - iter 594/992 - loss 0.27099342 - time (sec): 13.31 - samples/sec: 7460.02 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-18 21:16:11,436 epoch 2 - iter 693/992 - loss 0.26826560 - time (sec): 15.57 - samples/sec: 7433.50 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-18 21:16:13,617 epoch 2 - iter 792/992 - loss 0.26477509 - time (sec): 17.75 - samples/sec: 7393.62 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-18 21:16:15,758 epoch 2 - iter 891/992 - loss 0.26317298 - time (sec): 19.89 - samples/sec: 7333.02 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-18 21:16:17,807 epoch 2 - iter 990/992 - loss 0.25944860 - time (sec): 21.94 - samples/sec: 7465.20 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-18 21:16:17,844 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 21:16:17,845 EPOCH 2 done: loss 0.2593 - lr: 0.000044
105
+ 2023-10-18 21:16:20,056 DEV : loss 0.16714806854724884 - f1-score (micro avg) 0.4111
106
+ 2023-10-18 21:16:20,075 saving best model
107
+ 2023-10-18 21:16:20,110 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-18 21:16:22,218 epoch 3 - iter 99/992 - loss 0.23036403 - time (sec): 2.11 - samples/sec: 7812.27 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-18 21:16:24,469 epoch 3 - iter 198/992 - loss 0.23058276 - time (sec): 4.36 - samples/sec: 7626.43 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-18 21:16:26,711 epoch 3 - iter 297/992 - loss 0.21820133 - time (sec): 6.60 - samples/sec: 7536.57 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-18 21:16:28,916 epoch 3 - iter 396/992 - loss 0.22504416 - time (sec): 8.81 - samples/sec: 7493.40 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-18 21:16:31,138 epoch 3 - iter 495/992 - loss 0.22164956 - time (sec): 11.03 - samples/sec: 7402.51 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-18 21:16:33,390 epoch 3 - iter 594/992 - loss 0.22088483 - time (sec): 13.28 - samples/sec: 7400.22 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-18 21:16:35,663 epoch 3 - iter 693/992 - loss 0.21848559 - time (sec): 15.55 - samples/sec: 7360.05 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-18 21:16:38,037 epoch 3 - iter 792/992 - loss 0.21608363 - time (sec): 17.93 - samples/sec: 7370.97 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-18 21:16:40,357 epoch 3 - iter 891/992 - loss 0.21449733 - time (sec): 20.25 - samples/sec: 7305.29 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-18 21:16:42,642 epoch 3 - iter 990/992 - loss 0.21398506 - time (sec): 22.53 - samples/sec: 7266.44 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-18 21:16:42,687 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 21:16:42,688 EPOCH 3 done: loss 0.2139 - lr: 0.000039
120
+ 2023-10-18 21:16:44,501 DEV : loss 0.14900191128253937 - f1-score (micro avg) 0.5128
121
+ 2023-10-18 21:16:44,520 saving best model
122
+ 2023-10-18 21:16:44,555 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-18 21:16:46,843 epoch 4 - iter 99/992 - loss 0.19212402 - time (sec): 2.29 - samples/sec: 7423.04 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-18 21:16:49,021 epoch 4 - iter 198/992 - loss 0.19042800 - time (sec): 4.47 - samples/sec: 7091.02 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-18 21:16:51,197 epoch 4 - iter 297/992 - loss 0.18919111 - time (sec): 6.64 - samples/sec: 7171.20 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-18 21:16:53,390 epoch 4 - iter 396/992 - loss 0.18793688 - time (sec): 8.83 - samples/sec: 7224.86 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-18 21:16:55,585 epoch 4 - iter 495/992 - loss 0.19170755 - time (sec): 11.03 - samples/sec: 7327.72 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-18 21:16:57,801 epoch 4 - iter 594/992 - loss 0.18825759 - time (sec): 13.25 - samples/sec: 7363.35 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-18 21:17:00,013 epoch 4 - iter 693/992 - loss 0.19066190 - time (sec): 15.46 - samples/sec: 7338.86 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-18 21:17:02,303 epoch 4 - iter 792/992 - loss 0.18828028 - time (sec): 17.75 - samples/sec: 7312.02 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-18 21:17:04,532 epoch 4 - iter 891/992 - loss 0.18940494 - time (sec): 19.98 - samples/sec: 7301.53 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-18 21:17:06,861 epoch 4 - iter 990/992 - loss 0.18948850 - time (sec): 22.31 - samples/sec: 7335.54 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-18 21:17:06,913 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-18 21:17:06,913 EPOCH 4 done: loss 0.1897 - lr: 0.000033
135
+ 2023-10-18 21:17:08,764 DEV : loss 0.14117993414402008 - f1-score (micro avg) 0.5446
136
+ 2023-10-18 21:17:08,783 saving best model
137
+ 2023-10-18 21:17:08,817 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-18 21:17:11,009 epoch 5 - iter 99/992 - loss 0.15229802 - time (sec): 2.19 - samples/sec: 7372.22 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-18 21:17:13,239 epoch 5 - iter 198/992 - loss 0.16129571 - time (sec): 4.42 - samples/sec: 7336.97 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-18 21:17:15,485 epoch 5 - iter 297/992 - loss 0.16681954 - time (sec): 6.67 - samples/sec: 7239.47 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-18 21:17:17,603 epoch 5 - iter 396/992 - loss 0.16422500 - time (sec): 8.78 - samples/sec: 7408.94 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-18 21:17:19,858 epoch 5 - iter 495/992 - loss 0.16568909 - time (sec): 11.04 - samples/sec: 7348.81 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-18 21:17:22,037 epoch 5 - iter 594/992 - loss 0.16934167 - time (sec): 13.22 - samples/sec: 7304.06 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-18 21:17:24,279 epoch 5 - iter 693/992 - loss 0.17025896 - time (sec): 15.46 - samples/sec: 7320.71 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-18 21:17:26,566 epoch 5 - iter 792/992 - loss 0.17179453 - time (sec): 17.75 - samples/sec: 7355.20 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-18 21:17:28,773 epoch 5 - iter 891/992 - loss 0.17143573 - time (sec): 19.96 - samples/sec: 7392.66 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-18 21:17:30,997 epoch 5 - iter 990/992 - loss 0.17025017 - time (sec): 22.18 - samples/sec: 7379.74 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-18 21:17:31,042 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-18 21:17:31,042 EPOCH 5 done: loss 0.1701 - lr: 0.000028
150
+ 2023-10-18 21:17:32,868 DEV : loss 0.1357688307762146 - f1-score (micro avg) 0.5582
151
+ 2023-10-18 21:17:32,887 saving best model
152
+ 2023-10-18 21:17:32,922 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-18 21:17:35,146 epoch 6 - iter 99/992 - loss 0.17300109 - time (sec): 2.22 - samples/sec: 7210.04 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-18 21:17:37,367 epoch 6 - iter 198/992 - loss 0.17503273 - time (sec): 4.44 - samples/sec: 7303.51 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-18 21:17:39,620 epoch 6 - iter 297/992 - loss 0.16777607 - time (sec): 6.70 - samples/sec: 7436.18 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-18 21:17:41,855 epoch 6 - iter 396/992 - loss 0.16541905 - time (sec): 8.93 - samples/sec: 7339.38 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-18 21:17:44,101 epoch 6 - iter 495/992 - loss 0.16309203 - time (sec): 11.18 - samples/sec: 7376.75 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-18 21:17:46,202 epoch 6 - iter 594/992 - loss 0.16325905 - time (sec): 13.28 - samples/sec: 7387.33 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-18 21:17:48,434 epoch 6 - iter 693/992 - loss 0.16011057 - time (sec): 15.51 - samples/sec: 7404.51 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-18 21:17:50,641 epoch 6 - iter 792/992 - loss 0.15712117 - time (sec): 17.72 - samples/sec: 7412.21 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-18 21:17:52,843 epoch 6 - iter 891/992 - loss 0.16041497 - time (sec): 19.92 - samples/sec: 7364.21 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-18 21:17:55,110 epoch 6 - iter 990/992 - loss 0.15981362 - time (sec): 22.19 - samples/sec: 7375.09 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-18 21:17:55,159 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-18 21:17:55,159 EPOCH 6 done: loss 0.1597 - lr: 0.000022
165
+ 2023-10-18 21:17:57,029 DEV : loss 0.13293296098709106 - f1-score (micro avg) 0.5787
166
+ 2023-10-18 21:17:57,048 saving best model
167
+ 2023-10-18 21:17:57,082 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-18 21:17:59,315 epoch 7 - iter 99/992 - loss 0.13942076 - time (sec): 2.23 - samples/sec: 7690.95 - lr: 0.000022 - momentum: 0.000000
169
+ 2023-10-18 21:18:01,494 epoch 7 - iter 198/992 - loss 0.14325448 - time (sec): 4.41 - samples/sec: 7803.96 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-18 21:18:03,759 epoch 7 - iter 297/992 - loss 0.14913377 - time (sec): 6.68 - samples/sec: 7639.00 - lr: 0.000021 - momentum: 0.000000
171
+ 2023-10-18 21:18:05,986 epoch 7 - iter 396/992 - loss 0.15476655 - time (sec): 8.90 - samples/sec: 7534.24 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-18 21:18:08,218 epoch 7 - iter 495/992 - loss 0.15347097 - time (sec): 11.14 - samples/sec: 7513.37 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-18 21:18:10,453 epoch 7 - iter 594/992 - loss 0.15068835 - time (sec): 13.37 - samples/sec: 7485.75 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-18 21:18:12,712 epoch 7 - iter 693/992 - loss 0.14857101 - time (sec): 15.63 - samples/sec: 7470.10 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-18 21:18:14,906 epoch 7 - iter 792/992 - loss 0.14959639 - time (sec): 17.82 - samples/sec: 7457.26 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-18 21:18:17,123 epoch 7 - iter 891/992 - loss 0.14984248 - time (sec): 20.04 - samples/sec: 7383.31 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-18 21:18:19,314 epoch 7 - iter 990/992 - loss 0.15087522 - time (sec): 22.23 - samples/sec: 7355.24 - lr: 0.000017 - momentum: 0.000000
178
+ 2023-10-18 21:18:19,366 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-18 21:18:19,366 EPOCH 7 done: loss 0.1508 - lr: 0.000017
180
+ 2023-10-18 21:18:21,594 DEV : loss 0.13094773888587952 - f1-score (micro avg) 0.5961
181
+ 2023-10-18 21:18:21,612 saving best model
182
+ 2023-10-18 21:18:21,648 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-18 21:18:23,910 epoch 8 - iter 99/992 - loss 0.13743854 - time (sec): 2.26 - samples/sec: 7400.23 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-18 21:18:26,262 epoch 8 - iter 198/992 - loss 0.14110764 - time (sec): 4.61 - samples/sec: 7204.42 - lr: 0.000016 - momentum: 0.000000
185
+ 2023-10-18 21:18:28,496 epoch 8 - iter 297/992 - loss 0.14324894 - time (sec): 6.85 - samples/sec: 7150.71 - lr: 0.000015 - momentum: 0.000000
186
+ 2023-10-18 21:18:30,689 epoch 8 - iter 396/992 - loss 0.14564214 - time (sec): 9.04 - samples/sec: 7201.23 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-18 21:18:32,915 epoch 8 - iter 495/992 - loss 0.14616240 - time (sec): 11.27 - samples/sec: 7330.10 - lr: 0.000014 - momentum: 0.000000
188
+ 2023-10-18 21:18:35,175 epoch 8 - iter 594/992 - loss 0.14778882 - time (sec): 13.53 - samples/sec: 7299.12 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-18 21:18:37,371 epoch 8 - iter 693/992 - loss 0.14760928 - time (sec): 15.72 - samples/sec: 7260.68 - lr: 0.000013 - momentum: 0.000000
190
+ 2023-10-18 21:18:39,585 epoch 8 - iter 792/992 - loss 0.14535005 - time (sec): 17.94 - samples/sec: 7304.22 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-18 21:18:41,814 epoch 8 - iter 891/992 - loss 0.14591531 - time (sec): 20.16 - samples/sec: 7302.19 - lr: 0.000012 - momentum: 0.000000
192
+ 2023-10-18 21:18:44,017 epoch 8 - iter 990/992 - loss 0.14527308 - time (sec): 22.37 - samples/sec: 7313.69 - lr: 0.000011 - momentum: 0.000000
193
+ 2023-10-18 21:18:44,065 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-18 21:18:44,065 EPOCH 8 done: loss 0.1451 - lr: 0.000011
195
+ 2023-10-18 21:18:45,896 DEV : loss 0.13168883323669434 - f1-score (micro avg) 0.5982
196
+ 2023-10-18 21:18:45,915 saving best model
197
+ 2023-10-18 21:18:45,951 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-18 21:18:48,175 epoch 9 - iter 99/992 - loss 0.13619358 - time (sec): 2.22 - samples/sec: 7282.70 - lr: 0.000011 - momentum: 0.000000
199
+ 2023-10-18 21:18:50,442 epoch 9 - iter 198/992 - loss 0.13633010 - time (sec): 4.49 - samples/sec: 7272.94 - lr: 0.000010 - momentum: 0.000000
200
+ 2023-10-18 21:18:52,662 epoch 9 - iter 297/992 - loss 0.13650790 - time (sec): 6.71 - samples/sec: 7255.07 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-18 21:18:54,868 epoch 9 - iter 396/992 - loss 0.13657536 - time (sec): 8.92 - samples/sec: 7228.79 - lr: 0.000009 - momentum: 0.000000
202
+ 2023-10-18 21:18:57,157 epoch 9 - iter 495/992 - loss 0.13625592 - time (sec): 11.21 - samples/sec: 7357.39 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-18 21:18:59,407 epoch 9 - iter 594/992 - loss 0.13653751 - time (sec): 13.46 - samples/sec: 7356.54 - lr: 0.000008 - momentum: 0.000000
204
+ 2023-10-18 21:19:01,641 epoch 9 - iter 693/992 - loss 0.13850946 - time (sec): 15.69 - samples/sec: 7338.11 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-18 21:19:03,933 epoch 9 - iter 792/992 - loss 0.13733639 - time (sec): 17.98 - samples/sec: 7322.44 - lr: 0.000007 - momentum: 0.000000
206
+ 2023-10-18 21:19:06,261 epoch 9 - iter 891/992 - loss 0.13746495 - time (sec): 20.31 - samples/sec: 7268.68 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-18 21:19:08,482 epoch 9 - iter 990/992 - loss 0.13949435 - time (sec): 22.53 - samples/sec: 7265.99 - lr: 0.000006 - momentum: 0.000000
208
+ 2023-10-18 21:19:08,526 ----------------------------------------------------------------------------------------------------
209
+ 2023-10-18 21:19:08,527 EPOCH 9 done: loss 0.1393 - lr: 0.000006
210
+ 2023-10-18 21:19:10,353 DEV : loss 0.13056860864162445 - f1-score (micro avg) 0.6056
211
+ 2023-10-18 21:19:10,372 saving best model
212
+ 2023-10-18 21:19:10,406 ----------------------------------------------------------------------------------------------------
213
+ 2023-10-18 21:19:12,669 epoch 10 - iter 99/992 - loss 0.13940439 - time (sec): 2.26 - samples/sec: 7075.75 - lr: 0.000005 - momentum: 0.000000
214
+ 2023-10-18 21:19:14,890 epoch 10 - iter 198/992 - loss 0.13866219 - time (sec): 4.48 - samples/sec: 7328.81 - lr: 0.000004 - momentum: 0.000000
215
+ 2023-10-18 21:19:17,089 epoch 10 - iter 297/992 - loss 0.13654053 - time (sec): 6.68 - samples/sec: 7366.53 - lr: 0.000004 - momentum: 0.000000
216
+ 2023-10-18 21:19:19,280 epoch 10 - iter 396/992 - loss 0.13952155 - time (sec): 8.87 - samples/sec: 7422.76 - lr: 0.000003 - momentum: 0.000000
217
+ 2023-10-18 21:19:21,607 epoch 10 - iter 495/992 - loss 0.13874725 - time (sec): 11.20 - samples/sec: 7386.53 - lr: 0.000003 - momentum: 0.000000
218
+ 2023-10-18 21:19:23,650 epoch 10 - iter 594/992 - loss 0.14103061 - time (sec): 13.24 - samples/sec: 7453.94 - lr: 0.000002 - momentum: 0.000000
219
+ 2023-10-18 21:19:25,737 epoch 10 - iter 693/992 - loss 0.14107427 - time (sec): 15.33 - samples/sec: 7458.99 - lr: 0.000002 - momentum: 0.000000
220
+ 2023-10-18 21:19:27,975 epoch 10 - iter 792/992 - loss 0.13930647 - time (sec): 17.57 - samples/sec: 7445.01 - lr: 0.000001 - momentum: 0.000000
221
+ 2023-10-18 21:19:30,292 epoch 10 - iter 891/992 - loss 0.14078958 - time (sec): 19.89 - samples/sec: 7377.90 - lr: 0.000001 - momentum: 0.000000
222
+ 2023-10-18 21:19:32,519 epoch 10 - iter 990/992 - loss 0.13836888 - time (sec): 22.11 - samples/sec: 7400.86 - lr: 0.000000 - momentum: 0.000000
223
+ 2023-10-18 21:19:32,565 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-18 21:19:32,565 EPOCH 10 done: loss 0.1383 - lr: 0.000000
225
+ 2023-10-18 21:19:34,388 DEV : loss 0.13230597972869873 - f1-score (micro avg) 0.6031
226
+ 2023-10-18 21:19:34,434 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-18 21:19:34,435 Loading model from best epoch ...
228
+ 2023-10-18 21:19:34,517 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG
229
+ 2023-10-18 21:19:36,025
230
+ Results:
231
+ - F-score (micro) 0.6226
232
+ - F-score (macro) 0.4557
233
+ - Accuracy 0.4912
234
+
235
+ By class:
236
+ precision recall f1-score support
237
+
238
+ LOC 0.7189 0.7496 0.7339 655
239
+ PER 0.4162 0.6233 0.4991 223
240
+ ORG 0.2973 0.0866 0.1341 127
241
+
242
+ micro avg 0.6082 0.6378 0.6226 1005
243
+ macro avg 0.4775 0.4865 0.4557 1005
244
+ weighted avg 0.5984 0.6378 0.6060 1005
245
+
246
+ 2023-10-18 21:19:36,025 ----------------------------------------------------------------------------------------------------