Hugging Face's logo

hmbert
/
flair-hipe-2022-hipe2020-fr

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
4e52b9f
·
1 Parent(s): 3676144

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 +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d563ca5dc69e43f79ce87585b4a250ea25eec21467049e07b3b70d7c47960315
3
+ size 443335879
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 16:19:36 0.0000 0.5590 0.1296 0.6740 0.7532 0.7114 0.5860
3
+ 2 16:20:57 0.0000 0.1321 0.1468 0.7100 0.7910 0.7483 0.6323
4
+ 3 16:22:18 0.0000 0.0851 0.1559 0.8000 0.8041 0.8021 0.6982
5
+ 4 16:23:40 0.0000 0.0578 0.1817 0.7629 0.8477 0.8030 0.6942
6
+ 5 16:25:02 0.0000 0.0421 0.1794 0.7954 0.8305 0.8126 0.7132
7
+ 6 16:26:23 0.0000 0.0308 0.1738 0.7855 0.8305 0.8073 0.7070
8
+ 7 16:27:44 0.0000 0.0215 0.2016 0.8082 0.8230 0.8156 0.7225
9
+ 8 16:29:06 0.0000 0.0139 0.2170 0.8241 0.8396 0.8318 0.7374
10
+ 9 16:30:27 0.0000 0.0110 0.2082 0.8231 0.8528 0.8377 0.7449
11
+ 10 16:31:50 0.0000 0.0074 0.2123 0.8242 0.8431 0.8335 0.7416
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 16:18:19,873 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 16:18:19,874 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=21, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-13 16:18:19,874 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 16:18:19,874 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
52
+ - NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
53
+ 2023-10-13 16:18:19,874 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 16:18:19,874 Train: 5901 sentences
55
+ 2023-10-13 16:18:19,874 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 16:18:19,874 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 16:18:19,874 Training Params:
58
+ 2023-10-13 16:18:19,874 - learning_rate: "3e-05"
59
+ 2023-10-13 16:18:19,874 - mini_batch_size: "4"
60
+ 2023-10-13 16:18:19,874 - max_epochs: "10"
61
+ 2023-10-13 16:18:19,874 - shuffle: "True"
62
+ 2023-10-13 16:18:19,874 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 16:18:19,874 Plugins:
64
+ 2023-10-13 16:18:19,874 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 16:18:19,874 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 16:18:19,874 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 16:18:19,874 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 16:18:19,874 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 16:18:19,875 Computation:
70
+ 2023-10-13 16:18:19,875 - compute on device: cuda:0
71
+ 2023-10-13 16:18:19,875 - embedding storage: none
72
+ 2023-10-13 16:18:19,875 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 16:18:19,875 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-13 16:18:19,875 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 16:18:19,875 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 16:18:26,876 epoch 1 - iter 147/1476 - loss 2.68811671 - time (sec): 7.00 - samples/sec: 2530.06 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-13 16:18:33,935 epoch 1 - iter 294/1476 - loss 1.63436810 - time (sec): 14.06 - samples/sec: 2556.09 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-13 16:18:40,951 epoch 1 - iter 441/1476 - loss 1.24746196 - time (sec): 21.07 - samples/sec: 2466.58 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-13 16:18:48,209 epoch 1 - iter 588/1476 - loss 1.01857941 - time (sec): 28.33 - samples/sec: 2425.58 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-13 16:18:55,376 epoch 1 - iter 735/1476 - loss 0.88100047 - time (sec): 35.50 - samples/sec: 2410.76 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-13 16:19:02,343 epoch 1 - iter 882/1476 - loss 0.77829880 - time (sec): 42.47 - samples/sec: 2408.70 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-13 16:19:09,239 epoch 1 - iter 1029/1476 - loss 0.70303474 - time (sec): 49.36 - samples/sec: 2402.58 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-13 16:19:15,979 epoch 1 - iter 1176/1476 - loss 0.64928259 - time (sec): 56.10 - samples/sec: 2376.92 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-13 16:19:22,808 epoch 1 - iter 1323/1476 - loss 0.60030866 - time (sec): 62.93 - samples/sec: 2378.25 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-13 16:19:29,670 epoch 1 - iter 1470/1476 - loss 0.56005280 - time (sec): 69.79 - samples/sec: 2376.48 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-13 16:19:29,929 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 16:19:29,929 EPOCH 1 done: loss 0.5590 - lr: 0.000030
88
+ 2023-10-13 16:19:36,108 DEV : loss 0.12960338592529297 - f1-score (micro avg) 0.7114
89
+ 2023-10-13 16:19:36,137 saving best model
90
+ 2023-10-13 16:19:36,590 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 16:19:43,259 epoch 2 - iter 147/1476 - loss 0.14371163 - time (sec): 6.67 - samples/sec: 2282.89 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-13 16:19:50,126 epoch 2 - iter 294/1476 - loss 0.14687695 - time (sec): 13.53 - samples/sec: 2321.77 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-13 16:19:56,916 epoch 2 - iter 441/1476 - loss 0.14324943 - time (sec): 20.33 - samples/sec: 2350.40 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-13 16:20:03,828 epoch 2 - iter 588/1476 - loss 0.13800148 - time (sec): 27.24 - samples/sec: 2341.03 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-13 16:20:10,771 epoch 2 - iter 735/1476 - loss 0.14220914 - time (sec): 34.18 - samples/sec: 2310.96 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-13 16:20:17,816 epoch 2 - iter 882/1476 - loss 0.14051112 - time (sec): 41.22 - samples/sec: 2330.35 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-13 16:20:25,068 epoch 2 - iter 1029/1476 - loss 0.13606559 - time (sec): 48.48 - samples/sec: 2361.19 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-13 16:20:31,924 epoch 2 - iter 1176/1476 - loss 0.13105034 - time (sec): 55.33 - samples/sec: 2366.25 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-13 16:20:38,969 epoch 2 - iter 1323/1476 - loss 0.13230912 - time (sec): 62.38 - samples/sec: 2373.34 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-13 16:20:46,152 epoch 2 - iter 1470/1476 - loss 0.13197270 - time (sec): 69.56 - samples/sec: 2380.76 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-13 16:20:46,430 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 16:20:46,431 EPOCH 2 done: loss 0.1321 - lr: 0.000027
103
+ 2023-10-13 16:20:57,635 DEV : loss 0.14676305651664734 - f1-score (micro avg) 0.7483
104
+ 2023-10-13 16:20:57,663 saving best model
105
+ 2023-10-13 16:20:58,251 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 16:21:05,130 epoch 3 - iter 147/1476 - loss 0.07375163 - time (sec): 6.88 - samples/sec: 2256.94 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-13 16:21:12,085 epoch 3 - iter 294/1476 - loss 0.08555357 - time (sec): 13.83 - samples/sec: 2339.15 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-13 16:21:18,976 epoch 3 - iter 441/1476 - loss 0.08707338 - time (sec): 20.72 - samples/sec: 2346.38 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-13 16:21:25,687 epoch 3 - iter 588/1476 - loss 0.08703905 - time (sec): 27.43 - samples/sec: 2337.30 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-13 16:21:32,780 epoch 3 - iter 735/1476 - loss 0.08725903 - time (sec): 34.53 - samples/sec: 2362.45 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-13 16:21:39,920 epoch 3 - iter 882/1476 - loss 0.08502530 - time (sec): 41.67 - samples/sec: 2401.73 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-13 16:21:46,866 epoch 3 - iter 1029/1476 - loss 0.08313622 - time (sec): 48.61 - samples/sec: 2388.85 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-13 16:21:53,843 epoch 3 - iter 1176/1476 - loss 0.08509026 - time (sec): 55.59 - samples/sec: 2410.45 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-13 16:22:00,456 epoch 3 - iter 1323/1476 - loss 0.08340603 - time (sec): 62.20 - samples/sec: 2415.55 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-13 16:22:07,228 epoch 3 - iter 1470/1476 - loss 0.08513356 - time (sec): 68.97 - samples/sec: 2404.08 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-13 16:22:07,489 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 16:22:07,489 EPOCH 3 done: loss 0.0851 - lr: 0.000023
118
+ 2023-10-13 16:22:18,604 DEV : loss 0.1558631807565689 - f1-score (micro avg) 0.8021
119
+ 2023-10-13 16:22:18,632 saving best model
120
+ 2023-10-13 16:22:19,132 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 16:22:25,971 epoch 4 - iter 147/1476 - loss 0.05321490 - time (sec): 6.84 - samples/sec: 2228.72 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-13 16:22:32,717 epoch 4 - iter 294/1476 - loss 0.05088317 - time (sec): 13.58 - samples/sec: 2289.73 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-13 16:22:39,592 epoch 4 - iter 441/1476 - loss 0.05582560 - time (sec): 20.46 - samples/sec: 2337.16 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-13 16:22:46,278 epoch 4 - iter 588/1476 - loss 0.05853689 - time (sec): 27.14 - samples/sec: 2329.23 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-13 16:22:53,495 epoch 4 - iter 735/1476 - loss 0.05804754 - time (sec): 34.36 - samples/sec: 2325.70 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-13 16:23:00,651 epoch 4 - iter 882/1476 - loss 0.05583338 - time (sec): 41.52 - samples/sec: 2346.35 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-13 16:23:07,936 epoch 4 - iter 1029/1476 - loss 0.05481177 - time (sec): 48.80 - samples/sec: 2384.46 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-13 16:23:14,776 epoch 4 - iter 1176/1476 - loss 0.05475217 - time (sec): 55.64 - samples/sec: 2389.60 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-13 16:23:21,862 epoch 4 - iter 1323/1476 - loss 0.05730543 - time (sec): 62.73 - samples/sec: 2386.95 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-13 16:23:28,584 epoch 4 - iter 1470/1476 - loss 0.05764999 - time (sec): 69.45 - samples/sec: 2386.25 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-13 16:23:28,863 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 16:23:28,864 EPOCH 4 done: loss 0.0578 - lr: 0.000020
133
+ 2023-10-13 16:23:40,076 DEV : loss 0.18168844282627106 - f1-score (micro avg) 0.803
134
+ 2023-10-13 16:23:40,104 saving best model
135
+ 2023-10-13 16:23:40,595 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 16:23:47,639 epoch 5 - iter 147/1476 - loss 0.05535559 - time (sec): 7.04 - samples/sec: 2350.97 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-13 16:23:54,732 epoch 5 - iter 294/1476 - loss 0.04177522 - time (sec): 14.14 - samples/sec: 2354.73 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-13 16:24:01,904 epoch 5 - iter 441/1476 - loss 0.04295503 - time (sec): 21.31 - samples/sec: 2362.66 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-13 16:24:09,007 epoch 5 - iter 588/1476 - loss 0.03870174 - time (sec): 28.41 - samples/sec: 2337.31 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-13 16:24:15,872 epoch 5 - iter 735/1476 - loss 0.03947753 - time (sec): 35.28 - samples/sec: 2340.03 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-13 16:24:22,622 epoch 5 - iter 882/1476 - loss 0.04215224 - time (sec): 42.03 - samples/sec: 2333.54 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-13 16:24:29,595 epoch 5 - iter 1029/1476 - loss 0.04176081 - time (sec): 49.00 - samples/sec: 2334.66 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-13 16:24:36,904 epoch 5 - iter 1176/1476 - loss 0.04176508 - time (sec): 56.31 - samples/sec: 2359.18 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-13 16:24:44,014 epoch 5 - iter 1323/1476 - loss 0.04105383 - time (sec): 63.42 - samples/sec: 2376.53 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-13 16:24:50,709 epoch 5 - iter 1470/1476 - loss 0.04223612 - time (sec): 70.11 - samples/sec: 2365.53 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-13 16:24:50,971 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 16:24:50,972 EPOCH 5 done: loss 0.0421 - lr: 0.000017
148
+ 2023-10-13 16:25:02,159 DEV : loss 0.17941001057624817 - f1-score (micro avg) 0.8126
149
+ 2023-10-13 16:25:02,189 saving best model
150
+ 2023-10-13 16:25:02,695 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-13 16:25:09,753 epoch 6 - iter 147/1476 - loss 0.02653020 - time (sec): 7.06 - samples/sec: 2115.54 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-13 16:25:16,999 epoch 6 - iter 294/1476 - loss 0.02862818 - time (sec): 14.30 - samples/sec: 2386.80 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-13 16:25:23,876 epoch 6 - iter 441/1476 - loss 0.03132242 - time (sec): 21.18 - samples/sec: 2410.71 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-13 16:25:30,766 epoch 6 - iter 588/1476 - loss 0.03209656 - time (sec): 28.07 - samples/sec: 2401.90 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-13 16:25:37,800 epoch 6 - iter 735/1476 - loss 0.03269220 - time (sec): 35.10 - samples/sec: 2419.37 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-13 16:25:44,929 epoch 6 - iter 882/1476 - loss 0.03367211 - time (sec): 42.23 - samples/sec: 2419.30 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-13 16:25:51,539 epoch 6 - iter 1029/1476 - loss 0.03249115 - time (sec): 48.84 - samples/sec: 2403.54 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-13 16:25:58,447 epoch 6 - iter 1176/1476 - loss 0.03140458 - time (sec): 55.75 - samples/sec: 2402.72 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-13 16:26:05,272 epoch 6 - iter 1323/1476 - loss 0.03117712 - time (sec): 62.58 - samples/sec: 2393.92 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-13 16:26:12,144 epoch 6 - iter 1470/1476 - loss 0.03087811 - time (sec): 69.45 - samples/sec: 2389.87 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-13 16:26:12,398 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-13 16:26:12,399 EPOCH 6 done: loss 0.0308 - lr: 0.000013
163
+ 2023-10-13 16:26:23,570 DEV : loss 0.1737550050020218 - f1-score (micro avg) 0.8073
164
+ 2023-10-13 16:26:23,599 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-13 16:26:30,817 epoch 7 - iter 147/1476 - loss 0.01587476 - time (sec): 7.22 - samples/sec: 2370.42 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-13 16:26:37,420 epoch 7 - iter 294/1476 - loss 0.01435257 - time (sec): 13.82 - samples/sec: 2325.55 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-13 16:26:44,799 epoch 7 - iter 441/1476 - loss 0.02157759 - time (sec): 21.20 - samples/sec: 2384.74 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-13 16:26:52,071 epoch 7 - iter 588/1476 - loss 0.02281126 - time (sec): 28.47 - samples/sec: 2434.79 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-13 16:26:58,793 epoch 7 - iter 735/1476 - loss 0.02402209 - time (sec): 35.19 - samples/sec: 2398.19 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-13 16:27:05,497 epoch 7 - iter 882/1476 - loss 0.02296830 - time (sec): 41.90 - samples/sec: 2381.17 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-13 16:27:12,184 epoch 7 - iter 1029/1476 - loss 0.02320940 - time (sec): 48.58 - samples/sec: 2384.78 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-13 16:27:18,947 epoch 7 - iter 1176/1476 - loss 0.02303719 - time (sec): 55.35 - samples/sec: 2376.63 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-13 16:27:25,778 epoch 7 - iter 1323/1476 - loss 0.02138011 - time (sec): 62.18 - samples/sec: 2373.80 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-13 16:27:32,996 epoch 7 - iter 1470/1476 - loss 0.02155465 - time (sec): 69.40 - samples/sec: 2390.58 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-13 16:27:33,255 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-13 16:27:33,255 EPOCH 7 done: loss 0.0215 - lr: 0.000010
177
+ 2023-10-13 16:27:44,418 DEV : loss 0.20156921446323395 - f1-score (micro avg) 0.8156
178
+ 2023-10-13 16:27:44,448 saving best model
179
+ 2023-10-13 16:27:45,057 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-13 16:27:52,066 epoch 8 - iter 147/1476 - loss 0.01727771 - time (sec): 7.00 - samples/sec: 2398.66 - lr: 0.000010 - momentum: 0.000000
181
+ 2023-10-13 16:27:59,060 epoch 8 - iter 294/1476 - loss 0.01903390 - time (sec): 14.00 - samples/sec: 2412.30 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-13 16:28:06,483 epoch 8 - iter 441/1476 - loss 0.01852869 - time (sec): 21.42 - samples/sec: 2482.55 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-13 16:28:13,207 epoch 8 - iter 588/1476 - loss 0.01894146 - time (sec): 28.14 - samples/sec: 2415.93 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-13 16:28:20,113 epoch 8 - iter 735/1476 - loss 0.01775006 - time (sec): 35.05 - samples/sec: 2390.37 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-13 16:28:26,932 epoch 8 - iter 882/1476 - loss 0.01706730 - time (sec): 41.87 - samples/sec: 2369.20 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-13 16:28:33,847 epoch 8 - iter 1029/1476 - loss 0.01593317 - time (sec): 48.78 - samples/sec: 2361.06 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-13 16:28:40,797 epoch 8 - iter 1176/1476 - loss 0.01516748 - time (sec): 55.73 - samples/sec: 2338.33 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-13 16:28:47,956 epoch 8 - iter 1323/1476 - loss 0.01460520 - time (sec): 62.89 - samples/sec: 2354.33 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-13 16:28:54,942 epoch 8 - iter 1470/1476 - loss 0.01394633 - time (sec): 69.88 - samples/sec: 2372.44 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-13 16:28:55,205 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-13 16:28:55,205 EPOCH 8 done: loss 0.0139 - lr: 0.000007
192
+ 2023-10-13 16:29:06,318 DEV : loss 0.21695148944854736 - f1-score (micro avg) 0.8318
193
+ 2023-10-13 16:29:06,348 saving best model
194
+ 2023-10-13 16:29:06,888 ----------------------------------------------------------------------------------------------------
195
+ 2023-10-13 16:29:13,839 epoch 9 - iter 147/1476 - loss 0.01149428 - time (sec): 6.95 - samples/sec: 2241.86 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-13 16:29:20,892 epoch 9 - iter 294/1476 - loss 0.01051718 - time (sec): 14.00 - samples/sec: 2354.76 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-13 16:29:27,592 epoch 9 - iter 441/1476 - loss 0.01109949 - time (sec): 20.70 - samples/sec: 2346.93 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-13 16:29:34,486 epoch 9 - iter 588/1476 - loss 0.01034791 - time (sec): 27.59 - samples/sec: 2365.06 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-13 16:29:41,777 epoch 9 - iter 735/1476 - loss 0.01038753 - time (sec): 34.88 - samples/sec: 2389.31 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-13 16:29:48,663 epoch 9 - iter 882/1476 - loss 0.00977770 - time (sec): 41.77 - samples/sec: 2369.64 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-13 16:29:55,755 epoch 9 - iter 1029/1476 - loss 0.00886349 - time (sec): 48.86 - samples/sec: 2375.51 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-13 16:30:02,562 epoch 9 - iter 1176/1476 - loss 0.00868716 - time (sec): 55.67 - samples/sec: 2364.56 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-13 16:30:09,255 epoch 9 - iter 1323/1476 - loss 0.00810351 - time (sec): 62.36 - samples/sec: 2374.52 - lr: 0.000004 - momentum: 0.000000
204
+ 2023-10-13 16:30:16,370 epoch 9 - iter 1470/1476 - loss 0.01065603 - time (sec): 69.48 - samples/sec: 2383.02 - lr: 0.000003 - momentum: 0.000000
205
+ 2023-10-13 16:30:16,664 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-13 16:30:16,664 EPOCH 9 done: loss 0.0110 - lr: 0.000003
207
+ 2023-10-13 16:30:27,816 DEV : loss 0.20820745825767517 - f1-score (micro avg) 0.8377
208
+ 2023-10-13 16:30:27,845 saving best model
209
+ 2023-10-13 16:30:28,406 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-13 16:30:35,677 epoch 10 - iter 147/1476 - loss 0.00895335 - time (sec): 7.27 - samples/sec: 2440.59 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-13 16:30:42,470 epoch 10 - iter 294/1476 - loss 0.00653471 - time (sec): 14.06 - samples/sec: 2395.44 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-13 16:30:49,027 epoch 10 - iter 441/1476 - loss 0.00745494 - time (sec): 20.62 - samples/sec: 2388.45 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-13 16:30:56,197 epoch 10 - iter 588/1476 - loss 0.00712878 - time (sec): 27.79 - samples/sec: 2366.71 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-13 16:31:03,271 epoch 10 - iter 735/1476 - loss 0.00682681 - time (sec): 34.86 - samples/sec: 2347.22 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-13 16:31:10,608 epoch 10 - iter 882/1476 - loss 0.00780092 - time (sec): 42.20 - samples/sec: 2386.20 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 16:31:17,347 epoch 10 - iter 1029/1476 - loss 0.00743818 - time (sec): 48.94 - samples/sec: 2359.12 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-13 16:31:24,541 epoch 10 - iter 1176/1476 - loss 0.00763687 - time (sec): 56.13 - samples/sec: 2360.28 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-13 16:31:31,528 epoch 10 - iter 1323/1476 - loss 0.00786299 - time (sec): 63.12 - samples/sec: 2367.54 - lr: 0.000000 - momentum: 0.000000
219
+ 2023-10-13 16:31:38,440 epoch 10 - iter 1470/1476 - loss 0.00744089 - time (sec): 70.03 - samples/sec: 2368.59 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-13 16:31:38,703 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-13 16:31:38,703 EPOCH 10 done: loss 0.0074 - lr: 0.000000
222
+ 2023-10-13 16:31:50,374 DEV : loss 0.2123088389635086 - f1-score (micro avg) 0.8335
223
+ 2023-10-13 16:31:50,806 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-13 16:31:50,807 Loading model from best epoch ...
225
+ 2023-10-13 16:31:52,210 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
226
+ 2023-10-13 16:31:58,089
227
+ Results:
228
+ - F-score (micro) 0.7975
229
+ - F-score (macro) 0.703
230
+ - Accuracy 0.6891
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ loc 0.8449 0.8823 0.8632 858
236
+ pers 0.7500 0.8045 0.7763 537
237
+ org 0.5968 0.5606 0.5781 132
238
+ prod 0.6935 0.7049 0.6992 61
239
+ time 0.5556 0.6481 0.5983 54
240
+
241
+ micro avg 0.7792 0.8167 0.7975 1642
242
+ macro avg 0.6881 0.7201 0.7030 1642
243
+ weighted avg 0.7788 0.8167 0.7970 1642
244
+
245
+ 2023-10-13 16:31:58,089 ----------------------------------------------------------------------------------------------------