Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
Dutch
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
83ae4ac
•
1 Parent(s): 3fee925

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. test.tsv +0 -0
  5. training.log +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdf0dcc2b9d7dbd50fc868c384207b98dd3da7855468589b3e46ede2f57930f4
3
+ size 443311111
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 10:31:47 0.0000 0.3543 0.1273 0.7046 0.6136 0.6560 0.4967
3
+ 2 10:33:04 0.0000 0.1025 0.1066 0.8205 0.7273 0.7711 0.6435
4
+ 3 10:34:22 0.0000 0.0693 0.1001 0.8542 0.7686 0.8091 0.6889
5
+ 4 10:35:39 0.0000 0.0484 0.1196 0.8474 0.7800 0.8123 0.6927
6
+ 5 10:36:56 0.0000 0.0373 0.1803 0.8363 0.7180 0.7727 0.6447
7
+ 6 10:38:13 0.0000 0.0264 0.1579 0.8426 0.7355 0.7854 0.6673
8
+ 7 10:39:31 0.0000 0.0206 0.1583 0.8554 0.7820 0.8171 0.7022
9
+ 8 10:40:48 0.0000 0.0146 0.1665 0.8721 0.7820 0.8246 0.7108
10
+ 9 10:42:06 0.0000 0.0097 0.1674 0.8605 0.7903 0.8239 0.7110
11
+ 10 10:43:23 0.0000 0.0075 0.1799 0.8746 0.7779 0.8234 0.7090
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 10:30:31,688 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 10:30:31,689 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 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
+ (0-11): 12 x 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
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=768, out_features=768, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=768, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 10:30:31,690 MultiCorpus: 5777 train + 722 dev + 723 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
53
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 10:30:31,690 Train: 5777 sentences
55
+ 2023-10-14 10:30:31,690 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 10:30:31,690 Training Params:
58
+ 2023-10-14 10:30:31,690 - learning_rate: "3e-05"
59
+ 2023-10-14 10:30:31,690 - mini_batch_size: "4"
60
+ 2023-10-14 10:30:31,690 - max_epochs: "10"
61
+ 2023-10-14 10:30:31,690 - shuffle: "True"
62
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 10:30:31,690 Plugins:
64
+ 2023-10-14 10:30:31,690 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 10:30:31,690 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 10:30:31,690 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 10:30:31,690 Computation:
70
+ 2023-10-14 10:30:31,690 - compute on device: cuda:0
71
+ 2023-10-14 10:30:31,690 - embedding storage: none
72
+ 2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 10:30:31,691 Model training base path: "hmbench-icdar/nl-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-14 10:30:31,691 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 10:30:31,691 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 10:30:39,193 epoch 1 - iter 144/1445 - loss 1.87631987 - time (sec): 7.50 - samples/sec: 2486.02 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 10:30:46,429 epoch 1 - iter 288/1445 - loss 1.10871491 - time (sec): 14.74 - samples/sec: 2450.97 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 10:30:53,808 epoch 1 - iter 432/1445 - loss 0.81766779 - time (sec): 22.12 - samples/sec: 2409.53 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 10:31:00,941 epoch 1 - iter 576/1445 - loss 0.66179866 - time (sec): 29.25 - samples/sec: 2399.89 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 10:31:08,229 epoch 1 - iter 720/1445 - loss 0.56148096 - time (sec): 36.54 - samples/sec: 2416.30 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 10:31:15,427 epoch 1 - iter 864/1445 - loss 0.49511789 - time (sec): 43.74 - samples/sec: 2425.95 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 10:31:22,615 epoch 1 - iter 1008/1445 - loss 0.44567392 - time (sec): 50.92 - samples/sec: 2427.43 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 10:31:30,224 epoch 1 - iter 1152/1445 - loss 0.40668399 - time (sec): 58.53 - samples/sec: 2436.80 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 10:31:37,479 epoch 1 - iter 1296/1445 - loss 0.37622430 - time (sec): 65.79 - samples/sec: 2431.07 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 10:31:44,262 epoch 1 - iter 1440/1445 - loss 0.35511279 - time (sec): 72.57 - samples/sec: 2420.51 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 10:31:44,497 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 10:31:44,497 EPOCH 1 done: loss 0.3543 - lr: 0.000030
88
+ 2023-10-14 10:31:47,491 DEV : loss 0.12729743123054504 - f1-score (micro avg) 0.656
89
+ 2023-10-14 10:31:47,518 saving best model
90
+ 2023-10-14 10:31:47,917 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 10:31:55,504 epoch 2 - iter 144/1445 - loss 0.12159020 - time (sec): 7.59 - samples/sec: 2138.39 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 10:32:02,751 epoch 2 - iter 288/1445 - loss 0.11080145 - time (sec): 14.83 - samples/sec: 2278.16 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 10:32:10,121 epoch 2 - iter 432/1445 - loss 0.11375360 - time (sec): 22.20 - samples/sec: 2331.27 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 10:32:17,758 epoch 2 - iter 576/1445 - loss 0.10963141 - time (sec): 29.84 - samples/sec: 2375.48 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 10:32:25,047 epoch 2 - iter 720/1445 - loss 0.10826773 - time (sec): 37.13 - samples/sec: 2392.41 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 10:32:32,192 epoch 2 - iter 864/1445 - loss 0.10632474 - time (sec): 44.27 - samples/sec: 2384.95 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 10:32:39,186 epoch 2 - iter 1008/1445 - loss 0.10768588 - time (sec): 51.27 - samples/sec: 2382.07 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 10:32:46,323 epoch 2 - iter 1152/1445 - loss 0.10485484 - time (sec): 58.40 - samples/sec: 2387.46 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 10:32:53,665 epoch 2 - iter 1296/1445 - loss 0.10330686 - time (sec): 65.75 - samples/sec: 2391.73 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 10:33:01,061 epoch 2 - iter 1440/1445 - loss 0.10268253 - time (sec): 73.14 - samples/sec: 2402.14 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 10:33:01,320 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 10:33:01,321 EPOCH 2 done: loss 0.1025 - lr: 0.000027
103
+ 2023-10-14 10:33:04,810 DEV : loss 0.1065981537103653 - f1-score (micro avg) 0.7711
104
+ 2023-10-14 10:33:04,827 saving best model
105
+ 2023-10-14 10:33:05,296 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 10:33:12,939 epoch 3 - iter 144/1445 - loss 0.06396374 - time (sec): 7.64 - samples/sec: 2373.16 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 10:33:20,190 epoch 3 - iter 288/1445 - loss 0.05986877 - time (sec): 14.89 - samples/sec: 2391.48 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 10:33:27,268 epoch 3 - iter 432/1445 - loss 0.06439752 - time (sec): 21.97 - samples/sec: 2363.30 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-14 10:33:34,315 epoch 3 - iter 576/1445 - loss 0.06506263 - time (sec): 29.02 - samples/sec: 2386.42 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 10:33:41,658 epoch 3 - iter 720/1445 - loss 0.06609539 - time (sec): 36.36 - samples/sec: 2411.92 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 10:33:48,841 epoch 3 - iter 864/1445 - loss 0.06894766 - time (sec): 43.54 - samples/sec: 2414.80 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-14 10:33:56,286 epoch 3 - iter 1008/1445 - loss 0.06961197 - time (sec): 50.99 - samples/sec: 2429.56 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 10:34:03,260 epoch 3 - iter 1152/1445 - loss 0.06912057 - time (sec): 57.96 - samples/sec: 2421.10 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 10:34:10,460 epoch 3 - iter 1296/1445 - loss 0.06848518 - time (sec): 65.16 - samples/sec: 2416.48 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-14 10:34:17,762 epoch 3 - iter 1440/1445 - loss 0.06940007 - time (sec): 72.46 - samples/sec: 2420.46 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-14 10:34:18,075 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 10:34:18,075 EPOCH 3 done: loss 0.0693 - lr: 0.000023
118
+ 2023-10-14 10:34:22,064 DEV : loss 0.10008460283279419 - f1-score (micro avg) 0.8091
119
+ 2023-10-14 10:34:22,083 saving best model
120
+ 2023-10-14 10:34:22,596 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 10:34:30,131 epoch 4 - iter 144/1445 - loss 0.04151479 - time (sec): 7.53 - samples/sec: 2332.12 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 10:34:37,775 epoch 4 - iter 288/1445 - loss 0.05738197 - time (sec): 15.18 - samples/sec: 2362.39 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-14 10:34:44,979 epoch 4 - iter 432/1445 - loss 0.05769172 - time (sec): 22.38 - samples/sec: 2358.88 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 10:34:52,330 epoch 4 - iter 576/1445 - loss 0.05226711 - time (sec): 29.73 - samples/sec: 2376.58 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 10:34:59,385 epoch 4 - iter 720/1445 - loss 0.05024125 - time (sec): 36.79 - samples/sec: 2369.38 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-14 10:35:06,722 epoch 4 - iter 864/1445 - loss 0.04812138 - time (sec): 44.12 - samples/sec: 2395.51 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 10:35:13,970 epoch 4 - iter 1008/1445 - loss 0.04874861 - time (sec): 51.37 - samples/sec: 2389.48 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 10:35:21,254 epoch 4 - iter 1152/1445 - loss 0.04918823 - time (sec): 58.66 - samples/sec: 2395.34 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-14 10:35:28,544 epoch 4 - iter 1296/1445 - loss 0.04838955 - time (sec): 65.95 - samples/sec: 2404.60 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 10:35:35,780 epoch 4 - iter 1440/1445 - loss 0.04831869 - time (sec): 73.18 - samples/sec: 2399.55 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-14 10:35:36,022 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 10:35:36,023 EPOCH 4 done: loss 0.0484 - lr: 0.000020
133
+ 2023-10-14 10:35:39,548 DEV : loss 0.11963574588298798 - f1-score (micro avg) 0.8123
134
+ 2023-10-14 10:35:39,565 saving best model
135
+ 2023-10-14 10:35:40,073 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-14 10:35:47,545 epoch 5 - iter 144/1445 - loss 0.04127449 - time (sec): 7.47 - samples/sec: 2465.38 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-14 10:35:54,619 epoch 5 - iter 288/1445 - loss 0.04126193 - time (sec): 14.54 - samples/sec: 2436.83 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 10:36:02,056 epoch 5 - iter 432/1445 - loss 0.03693959 - time (sec): 21.98 - samples/sec: 2462.92 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-14 10:36:09,257 epoch 5 - iter 576/1445 - loss 0.03990160 - time (sec): 29.18 - samples/sec: 2427.30 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-14 10:36:16,307 epoch 5 - iter 720/1445 - loss 0.03748764 - time (sec): 36.23 - samples/sec: 2419.21 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 10:36:23,621 epoch 5 - iter 864/1445 - loss 0.03765664 - time (sec): 43.54 - samples/sec: 2383.06 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-14 10:36:30,856 epoch 5 - iter 1008/1445 - loss 0.03711077 - time (sec): 50.78 - samples/sec: 2401.78 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-14 10:36:38,122 epoch 5 - iter 1152/1445 - loss 0.03800064 - time (sec): 58.05 - samples/sec: 2407.38 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 10:36:45,564 epoch 5 - iter 1296/1445 - loss 0.03872554 - time (sec): 65.49 - samples/sec: 2416.72 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-14 10:36:52,852 epoch 5 - iter 1440/1445 - loss 0.03722040 - time (sec): 72.78 - samples/sec: 2414.26 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-14 10:36:53,079 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-14 10:36:53,079 EPOCH 5 done: loss 0.0373 - lr: 0.000017
148
+ 2023-10-14 10:36:56,597 DEV : loss 0.18025827407836914 - f1-score (micro avg) 0.7727
149
+ 2023-10-14 10:36:56,614 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-14 10:37:03,761 epoch 6 - iter 144/1445 - loss 0.02549194 - time (sec): 7.15 - samples/sec: 2428.23 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-14 10:37:11,095 epoch 6 - iter 288/1445 - loss 0.02825162 - time (sec): 14.48 - samples/sec: 2422.36 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-14 10:37:18,192 epoch 6 - iter 432/1445 - loss 0.02593624 - time (sec): 21.58 - samples/sec: 2432.82 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-14 10:37:25,601 epoch 6 - iter 576/1445 - loss 0.02696454 - time (sec): 28.99 - samples/sec: 2433.41 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-14 10:37:32,807 epoch 6 - iter 720/1445 - loss 0.02662243 - time (sec): 36.19 - samples/sec: 2415.50 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-14 10:37:40,250 epoch 6 - iter 864/1445 - loss 0.02600658 - time (sec): 43.64 - samples/sec: 2407.28 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-14 10:37:47,675 epoch 6 - iter 1008/1445 - loss 0.02574166 - time (sec): 51.06 - samples/sec: 2406.40 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-14 10:37:55,224 epoch 6 - iter 1152/1445 - loss 0.02422581 - time (sec): 58.61 - samples/sec: 2427.06 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-14 10:38:02,318 epoch 6 - iter 1296/1445 - loss 0.02622967 - time (sec): 65.70 - samples/sec: 2421.76 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-14 10:38:09,361 epoch 6 - iter 1440/1445 - loss 0.02636942 - time (sec): 72.75 - samples/sec: 2416.00 - lr: 0.000013 - momentum: 0.000000
160
+ 2023-10-14 10:38:09,579 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-14 10:38:09,580 EPOCH 6 done: loss 0.0264 - lr: 0.000013
162
+ 2023-10-14 10:38:13,698 DEV : loss 0.1579323559999466 - f1-score (micro avg) 0.7854
163
+ 2023-10-14 10:38:13,726 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-14 10:38:21,834 epoch 7 - iter 144/1445 - loss 0.01472983 - time (sec): 8.11 - samples/sec: 2164.44 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-14 10:38:30,227 epoch 7 - iter 288/1445 - loss 0.01529246 - time (sec): 16.50 - samples/sec: 2221.27 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-14 10:38:37,324 epoch 7 - iter 432/1445 - loss 0.01675210 - time (sec): 23.60 - samples/sec: 2268.28 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-14 10:38:44,693 epoch 7 - iter 576/1445 - loss 0.01805192 - time (sec): 30.97 - samples/sec: 2322.73 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-14 10:38:51,730 epoch 7 - iter 720/1445 - loss 0.01822939 - time (sec): 38.00 - samples/sec: 2328.47 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-14 10:38:58,764 epoch 7 - iter 864/1445 - loss 0.01917789 - time (sec): 45.04 - samples/sec: 2333.32 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-14 10:39:06,276 epoch 7 - iter 1008/1445 - loss 0.02084014 - time (sec): 52.55 - samples/sec: 2356.77 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-14 10:39:13,611 epoch 7 - iter 1152/1445 - loss 0.02033218 - time (sec): 59.88 - samples/sec: 2369.19 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-14 10:39:20,678 epoch 7 - iter 1296/1445 - loss 0.02070489 - time (sec): 66.95 - samples/sec: 2376.83 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-14 10:39:27,817 epoch 7 - iter 1440/1445 - loss 0.02027042 - time (sec): 74.09 - samples/sec: 2372.88 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-14 10:39:28,051 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-14 10:39:28,051 EPOCH 7 done: loss 0.0206 - lr: 0.000010
176
+ 2023-10-14 10:39:31,558 DEV : loss 0.15825796127319336 - f1-score (micro avg) 0.8171
177
+ 2023-10-14 10:39:31,574 saving best model
178
+ 2023-10-14 10:39:32,037 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-14 10:39:39,139 epoch 8 - iter 144/1445 - loss 0.02396519 - time (sec): 7.10 - samples/sec: 2354.23 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-14 10:39:46,793 epoch 8 - iter 288/1445 - loss 0.01698943 - time (sec): 14.75 - samples/sec: 2344.64 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-14 10:39:54,032 epoch 8 - iter 432/1445 - loss 0.01654060 - time (sec): 21.99 - samples/sec: 2379.55 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-14 10:40:01,390 epoch 8 - iter 576/1445 - loss 0.01471133 - time (sec): 29.35 - samples/sec: 2374.05 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-14 10:40:08,625 epoch 8 - iter 720/1445 - loss 0.01430743 - time (sec): 36.59 - samples/sec: 2403.83 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-14 10:40:15,611 epoch 8 - iter 864/1445 - loss 0.01506483 - time (sec): 43.57 - samples/sec: 2399.74 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-14 10:40:23,044 epoch 8 - iter 1008/1445 - loss 0.01564489 - time (sec): 51.01 - samples/sec: 2406.42 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-14 10:40:30,274 epoch 8 - iter 1152/1445 - loss 0.01510179 - time (sec): 58.24 - samples/sec: 2410.96 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-14 10:40:37,318 epoch 8 - iter 1296/1445 - loss 0.01417517 - time (sec): 65.28 - samples/sec: 2413.79 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-14 10:40:44,723 epoch 8 - iter 1440/1445 - loss 0.01465237 - time (sec): 72.68 - samples/sec: 2413.51 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-14 10:40:44,975 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-14 10:40:44,975 EPOCH 8 done: loss 0.0146 - lr: 0.000007
191
+ 2023-10-14 10:40:48,530 DEV : loss 0.16648930311203003 - f1-score (micro avg) 0.8246
192
+ 2023-10-14 10:40:48,550 saving best model
193
+ 2023-10-14 10:40:49,056 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-14 10:40:56,452 epoch 9 - iter 144/1445 - loss 0.00629566 - time (sec): 7.39 - samples/sec: 2457.52 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-14 10:41:03,691 epoch 9 - iter 288/1445 - loss 0.00790835 - time (sec): 14.63 - samples/sec: 2414.93 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-14 10:41:11,244 epoch 9 - iter 432/1445 - loss 0.00782712 - time (sec): 22.18 - samples/sec: 2349.41 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-14 10:41:18,904 epoch 9 - iter 576/1445 - loss 0.00916480 - time (sec): 29.84 - samples/sec: 2391.33 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-14 10:41:26,049 epoch 9 - iter 720/1445 - loss 0.00886500 - time (sec): 36.99 - samples/sec: 2399.86 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-14 10:41:33,379 epoch 9 - iter 864/1445 - loss 0.00941845 - time (sec): 44.32 - samples/sec: 2406.98 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-14 10:41:40,390 epoch 9 - iter 1008/1445 - loss 0.00912052 - time (sec): 51.33 - samples/sec: 2406.29 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-14 10:41:47,769 epoch 9 - iter 1152/1445 - loss 0.00928227 - time (sec): 58.71 - samples/sec: 2409.76 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-14 10:41:54,865 epoch 9 - iter 1296/1445 - loss 0.00960617 - time (sec): 65.81 - samples/sec: 2408.28 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-14 10:42:02,054 epoch 9 - iter 1440/1445 - loss 0.00965262 - time (sec): 72.99 - samples/sec: 2407.34 - lr: 0.000003 - momentum: 0.000000
204
+ 2023-10-14 10:42:02,314 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-14 10:42:02,314 EPOCH 9 done: loss 0.0097 - lr: 0.000003
206
+ 2023-10-14 10:42:06,175 DEV : loss 0.1673649400472641 - f1-score (micro avg) 0.8239
207
+ 2023-10-14 10:42:06,191 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-14 10:42:13,336 epoch 10 - iter 144/1445 - loss 0.00886764 - time (sec): 7.14 - samples/sec: 2336.97 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-14 10:42:20,685 epoch 10 - iter 288/1445 - loss 0.00772641 - time (sec): 14.49 - samples/sec: 2396.61 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-14 10:42:28,082 epoch 10 - iter 432/1445 - loss 0.00818643 - time (sec): 21.89 - samples/sec: 2393.56 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-14 10:42:35,619 epoch 10 - iter 576/1445 - loss 0.00943870 - time (sec): 29.43 - samples/sec: 2417.73 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 10:42:43,034 epoch 10 - iter 720/1445 - loss 0.00885679 - time (sec): 36.84 - samples/sec: 2432.46 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-14 10:42:50,269 epoch 10 - iter 864/1445 - loss 0.00944317 - time (sec): 44.08 - samples/sec: 2424.76 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 10:42:57,240 epoch 10 - iter 1008/1445 - loss 0.00869228 - time (sec): 51.05 - samples/sec: 2401.38 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-14 10:43:04,381 epoch 10 - iter 1152/1445 - loss 0.00816640 - time (sec): 58.19 - samples/sec: 2394.05 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-14 10:43:11,807 epoch 10 - iter 1296/1445 - loss 0.00783806 - time (sec): 65.61 - samples/sec: 2404.43 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-14 10:43:19,049 epoch 10 - iter 1440/1445 - loss 0.00751126 - time (sec): 72.86 - samples/sec: 2408.74 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-14 10:43:19,315 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-14 10:43:19,316 EPOCH 10 done: loss 0.0075 - lr: 0.000000
220
+ 2023-10-14 10:43:22,979 DEV : loss 0.17992718517780304 - f1-score (micro avg) 0.8234
221
+ 2023-10-14 10:43:23,439 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-14 10:43:23,441 Loading model from best epoch ...
223
+ 2023-10-14 10:43:25,159 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG
224
+ 2023-10-14 10:43:28,390
225
+ Results:
226
+ - F-score (micro) 0.8056
227
+ - F-score (macro) 0.7149
228
+ - Accuracy 0.6863
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ PER 0.8326 0.8050 0.8186 482
234
+ LOC 0.8916 0.7904 0.8380 458
235
+ ORG 0.5345 0.4493 0.4882 69
236
+
237
+ micro avg 0.8398 0.7740 0.8056 1009
238
+ macro avg 0.7529 0.6815 0.7149 1009
239
+ weighted avg 0.8390 0.7740 0.8048 1009
240
+
241
+ 2023-10-14 10:43:28,390 ----------------------------------------------------------------------------------------------------