Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
96e0c59
•
1 Parent(s): 8a23eee

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 +240 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4871e429b94f12eb0c759e74b5cbbbceeafa8451210087889d140880ca4cf4b0
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 21:16:07 0.0000 0.3126 0.1112 0.5319 0.7632 0.6269 0.4622
3
+ 2 21:18:05 0.0000 0.0841 0.1181 0.5075 0.8101 0.6241 0.4637
4
+ 3 21:20:05 0.0000 0.0570 0.1585 0.5338 0.7780 0.6331 0.4732
5
+ 4 21:22:04 0.0000 0.0415 0.2364 0.5237 0.8101 0.6361 0.4758
6
+ 5 21:24:02 0.0000 0.0308 0.2963 0.5377 0.7746 0.6348 0.4748
7
+ 6 21:26:01 0.0000 0.0209 0.3504 0.5268 0.7746 0.6271 0.4682
8
+ 7 21:27:58 0.0000 0.0144 0.3562 0.5558 0.7632 0.6432 0.4833
9
+ 8 21:29:57 0.0000 0.0103 0.3832 0.5518 0.7803 0.6464 0.4858
10
+ 9 21:31:55 0.0000 0.0070 0.3889 0.5571 0.7643 0.6445 0.4841
11
+ 10 21:33:55 0.0000 0.0051 0.4082 0.5465 0.7803 0.6428 0.4823
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 21:14:10,263 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 21:14:10,264 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 21:14:10,264 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 21:14:10,264 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
52
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
53
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 21:14:10,264 Train: 14465 sentences
55
+ 2023-10-14 21:14:10,264 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 21:14:10,264 Training Params:
58
+ 2023-10-14 21:14:10,264 - learning_rate: "3e-05"
59
+ 2023-10-14 21:14:10,264 - mini_batch_size: "8"
60
+ 2023-10-14 21:14:10,264 - max_epochs: "10"
61
+ 2023-10-14 21:14:10,264 - shuffle: "True"
62
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 21:14:10,264 Plugins:
64
+ 2023-10-14 21:14:10,264 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 21:14:10,264 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 21:14:10,264 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 21:14:10,264 Computation:
70
+ 2023-10-14 21:14:10,264 - compute on device: cuda:0
71
+ 2023-10-14 21:14:10,264 - embedding storage: none
72
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 21:14:10,264 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 21:14:10,264 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 21:14:21,346 epoch 1 - iter 180/1809 - loss 1.90381785 - time (sec): 11.08 - samples/sec: 3376.03 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 21:14:32,403 epoch 1 - iter 360/1809 - loss 1.06803482 - time (sec): 22.14 - samples/sec: 3397.48 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 21:14:43,052 epoch 1 - iter 540/1809 - loss 0.77636418 - time (sec): 32.79 - samples/sec: 3390.47 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 21:14:53,957 epoch 1 - iter 720/1809 - loss 0.61914340 - time (sec): 43.69 - samples/sec: 3387.88 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 21:15:05,506 epoch 1 - iter 900/1809 - loss 0.51652229 - time (sec): 55.24 - samples/sec: 3397.47 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 21:15:16,373 epoch 1 - iter 1080/1809 - loss 0.45106548 - time (sec): 66.11 - samples/sec: 3405.24 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 21:15:27,887 epoch 1 - iter 1260/1809 - loss 0.40095933 - time (sec): 77.62 - samples/sec: 3414.66 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 21:15:38,882 epoch 1 - iter 1440/1809 - loss 0.36564921 - time (sec): 88.62 - samples/sec: 3418.04 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 21:15:49,939 epoch 1 - iter 1620/1809 - loss 0.33661742 - time (sec): 99.67 - samples/sec: 3414.98 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 21:16:00,909 epoch 1 - iter 1800/1809 - loss 0.31361967 - time (sec): 110.64 - samples/sec: 3414.89 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 21:16:01,443 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 21:16:01,443 EPOCH 1 done: loss 0.3126 - lr: 0.000030
88
+ 2023-10-14 21:16:07,875 DEV : loss 0.11117860674858093 - f1-score (micro avg) 0.6269
89
+ 2023-10-14 21:16:07,921 saving best model
90
+ 2023-10-14 21:16:08,358 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 21:16:19,590 epoch 2 - iter 180/1809 - loss 0.08921046 - time (sec): 11.23 - samples/sec: 3404.09 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 21:16:30,539 epoch 2 - iter 360/1809 - loss 0.08748107 - time (sec): 22.18 - samples/sec: 3399.90 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 21:16:41,559 epoch 2 - iter 540/1809 - loss 0.08786843 - time (sec): 33.20 - samples/sec: 3384.05 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 21:16:52,961 epoch 2 - iter 720/1809 - loss 0.08653904 - time (sec): 44.60 - samples/sec: 3409.28 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 21:17:03,984 epoch 2 - iter 900/1809 - loss 0.08652461 - time (sec): 55.62 - samples/sec: 3406.12 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 21:17:15,172 epoch 2 - iter 1080/1809 - loss 0.08558330 - time (sec): 66.81 - samples/sec: 3414.01 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 21:17:26,791 epoch 2 - iter 1260/1809 - loss 0.08469632 - time (sec): 78.43 - samples/sec: 3409.34 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 21:17:37,861 epoch 2 - iter 1440/1809 - loss 0.08498290 - time (sec): 89.50 - samples/sec: 3405.49 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 21:17:48,517 epoch 2 - iter 1620/1809 - loss 0.08441384 - time (sec): 100.16 - samples/sec: 3398.36 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 21:17:59,613 epoch 2 - iter 1800/1809 - loss 0.08398362 - time (sec): 111.25 - samples/sec: 3399.35 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 21:18:00,196 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 21:18:00,196 EPOCH 2 done: loss 0.0841 - lr: 0.000027
103
+ 2023-10-14 21:18:05,895 DEV : loss 0.1181371659040451 - f1-score (micro avg) 0.6241
104
+ 2023-10-14 21:18:05,930 ----------------------------------------------------------------------------------------------------
105
+ 2023-10-14 21:18:17,254 epoch 3 - iter 180/1809 - loss 0.04597861 - time (sec): 11.32 - samples/sec: 3256.04 - lr: 0.000026 - momentum: 0.000000
106
+ 2023-10-14 21:18:29,594 epoch 3 - iter 360/1809 - loss 0.05194200 - time (sec): 23.66 - samples/sec: 3153.44 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 21:18:41,113 epoch 3 - iter 540/1809 - loss 0.05725339 - time (sec): 35.18 - samples/sec: 3194.67 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 21:18:52,263 epoch 3 - iter 720/1809 - loss 0.05751126 - time (sec): 46.33 - samples/sec: 3234.26 - lr: 0.000025 - momentum: 0.000000
109
+ 2023-10-14 21:19:03,375 epoch 3 - iter 900/1809 - loss 0.05698117 - time (sec): 57.44 - samples/sec: 3266.53 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 21:19:14,364 epoch 3 - iter 1080/1809 - loss 0.05635563 - time (sec): 68.43 - samples/sec: 3296.08 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 21:19:26,004 epoch 3 - iter 1260/1809 - loss 0.05644181 - time (sec): 80.07 - samples/sec: 3306.44 - lr: 0.000024 - momentum: 0.000000
112
+ 2023-10-14 21:19:37,443 epoch 3 - iter 1440/1809 - loss 0.05590284 - time (sec): 91.51 - samples/sec: 3306.55 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 21:19:48,688 epoch 3 - iter 1620/1809 - loss 0.05687337 - time (sec): 102.76 - samples/sec: 3313.21 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 21:19:59,697 epoch 3 - iter 1800/1809 - loss 0.05715098 - time (sec): 113.76 - samples/sec: 3325.13 - lr: 0.000023 - momentum: 0.000000
115
+ 2023-10-14 21:20:00,218 ----------------------------------------------------------------------------------------------------
116
+ 2023-10-14 21:20:00,218 EPOCH 3 done: loss 0.0570 - lr: 0.000023
117
+ 2023-10-14 21:20:05,948 DEV : loss 0.15852448344230652 - f1-score (micro avg) 0.6331
118
+ 2023-10-14 21:20:05,993 saving best model
119
+ 2023-10-14 21:20:06,557 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-14 21:20:17,420 epoch 4 - iter 180/1809 - loss 0.03774158 - time (sec): 10.86 - samples/sec: 3375.73 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-14 21:20:28,751 epoch 4 - iter 360/1809 - loss 0.04107676 - time (sec): 22.19 - samples/sec: 3376.57 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 21:20:39,811 epoch 4 - iter 540/1809 - loss 0.04036546 - time (sec): 33.25 - samples/sec: 3377.69 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-14 21:20:50,772 epoch 4 - iter 720/1809 - loss 0.04111588 - time (sec): 44.21 - samples/sec: 3394.90 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 21:21:02,835 epoch 4 - iter 900/1809 - loss 0.04004334 - time (sec): 56.28 - samples/sec: 3360.21 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 21:21:13,868 epoch 4 - iter 1080/1809 - loss 0.04015642 - time (sec): 67.31 - samples/sec: 3384.56 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-14 21:21:24,788 epoch 4 - iter 1260/1809 - loss 0.04109986 - time (sec): 78.23 - samples/sec: 3395.23 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 21:21:35,715 epoch 4 - iter 1440/1809 - loss 0.04150162 - time (sec): 89.16 - samples/sec: 3408.60 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 21:21:47,110 epoch 4 - iter 1620/1809 - loss 0.04171498 - time (sec): 100.55 - samples/sec: 3396.77 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-14 21:21:58,023 epoch 4 - iter 1800/1809 - loss 0.04159954 - time (sec): 111.46 - samples/sec: 3392.39 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 21:21:58,529 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-14 21:21:58,529 EPOCH 4 done: loss 0.0415 - lr: 0.000020
132
+ 2023-10-14 21:22:04,200 DEV : loss 0.23644335567951202 - f1-score (micro avg) 0.6361
133
+ 2023-10-14 21:22:04,231 saving best model
134
+ 2023-10-14 21:22:04,809 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-14 21:22:15,861 epoch 5 - iter 180/1809 - loss 0.03030318 - time (sec): 11.05 - samples/sec: 3333.61 - lr: 0.000020 - momentum: 0.000000
136
+ 2023-10-14 21:22:26,995 epoch 5 - iter 360/1809 - loss 0.03079855 - time (sec): 22.18 - samples/sec: 3422.91 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-14 21:22:38,145 epoch 5 - iter 540/1809 - loss 0.03069607 - time (sec): 33.33 - samples/sec: 3419.90 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 21:22:49,063 epoch 5 - iter 720/1809 - loss 0.03008395 - time (sec): 44.25 - samples/sec: 3414.06 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-14 21:23:00,143 epoch 5 - iter 900/1809 - loss 0.03015912 - time (sec): 55.33 - samples/sec: 3399.24 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-14 21:23:11,578 epoch 5 - iter 1080/1809 - loss 0.03105206 - time (sec): 66.77 - samples/sec: 3409.94 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 21:23:22,723 epoch 5 - iter 1260/1809 - loss 0.03115092 - time (sec): 77.91 - samples/sec: 3411.10 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-14 21:23:33,719 epoch 5 - iter 1440/1809 - loss 0.03091193 - time (sec): 88.91 - samples/sec: 3415.79 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-14 21:23:44,567 epoch 5 - iter 1620/1809 - loss 0.03146358 - time (sec): 99.75 - samples/sec: 3418.58 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 21:23:55,443 epoch 5 - iter 1800/1809 - loss 0.03089906 - time (sec): 110.63 - samples/sec: 3417.50 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-14 21:23:55,964 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-14 21:23:55,964 EPOCH 5 done: loss 0.0308 - lr: 0.000017
147
+ 2023-10-14 21:24:02,349 DEV : loss 0.29626017808914185 - f1-score (micro avg) 0.6348
148
+ 2023-10-14 21:24:02,382 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-14 21:24:13,473 epoch 6 - iter 180/1809 - loss 0.02177454 - time (sec): 11.09 - samples/sec: 3480.30 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-14 21:24:24,551 epoch 6 - iter 360/1809 - loss 0.01815594 - time (sec): 22.17 - samples/sec: 3419.15 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-14 21:24:35,749 epoch 6 - iter 540/1809 - loss 0.01951666 - time (sec): 33.37 - samples/sec: 3413.28 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-14 21:24:46,567 epoch 6 - iter 720/1809 - loss 0.01954928 - time (sec): 44.18 - samples/sec: 3399.04 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-14 21:24:57,651 epoch 6 - iter 900/1809 - loss 0.02065711 - time (sec): 55.27 - samples/sec: 3391.16 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-14 21:25:08,694 epoch 6 - iter 1080/1809 - loss 0.02124727 - time (sec): 66.31 - samples/sec: 3387.12 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-14 21:25:19,606 epoch 6 - iter 1260/1809 - loss 0.02069374 - time (sec): 77.22 - samples/sec: 3394.31 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-14 21:25:30,722 epoch 6 - iter 1440/1809 - loss 0.02117259 - time (sec): 88.34 - samples/sec: 3402.25 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-14 21:25:41,996 epoch 6 - iter 1620/1809 - loss 0.02125642 - time (sec): 99.61 - samples/sec: 3403.99 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-14 21:25:53,227 epoch 6 - iter 1800/1809 - loss 0.02103519 - time (sec): 110.84 - samples/sec: 3409.62 - lr: 0.000013 - momentum: 0.000000
159
+ 2023-10-14 21:25:53,744 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-14 21:25:53,744 EPOCH 6 done: loss 0.0209 - lr: 0.000013
161
+ 2023-10-14 21:26:01,401 DEV : loss 0.35041970014572144 - f1-score (micro avg) 0.6271
162
+ 2023-10-14 21:26:01,436 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-14 21:26:12,224 epoch 7 - iter 180/1809 - loss 0.01038516 - time (sec): 10.79 - samples/sec: 3579.72 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-14 21:26:23,538 epoch 7 - iter 360/1809 - loss 0.01077422 - time (sec): 22.10 - samples/sec: 3448.80 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-14 21:26:34,673 epoch 7 - iter 540/1809 - loss 0.01201565 - time (sec): 33.24 - samples/sec: 3417.81 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-14 21:26:45,591 epoch 7 - iter 720/1809 - loss 0.01213645 - time (sec): 44.15 - samples/sec: 3443.25 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-14 21:26:56,659 epoch 7 - iter 900/1809 - loss 0.01412555 - time (sec): 55.22 - samples/sec: 3438.25 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-14 21:27:07,537 epoch 7 - iter 1080/1809 - loss 0.01408455 - time (sec): 66.10 - samples/sec: 3435.92 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-14 21:27:18,660 epoch 7 - iter 1260/1809 - loss 0.01332679 - time (sec): 77.22 - samples/sec: 3422.80 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-14 21:27:30,101 epoch 7 - iter 1440/1809 - loss 0.01394835 - time (sec): 88.66 - samples/sec: 3428.58 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-14 21:27:41,000 epoch 7 - iter 1620/1809 - loss 0.01439966 - time (sec): 99.56 - samples/sec: 3424.09 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-14 21:27:51,982 epoch 7 - iter 1800/1809 - loss 0.01437537 - time (sec): 110.54 - samples/sec: 3422.81 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-14 21:27:52,474 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-14 21:27:52,474 EPOCH 7 done: loss 0.0144 - lr: 0.000010
175
+ 2023-10-14 21:27:58,730 DEV : loss 0.3561807870864868 - f1-score (micro avg) 0.6432
176
+ 2023-10-14 21:27:58,762 saving best model
177
+ 2023-10-14 21:27:59,385 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-14 21:28:10,412 epoch 8 - iter 180/1809 - loss 0.00796504 - time (sec): 11.03 - samples/sec: 3440.41 - lr: 0.000010 - momentum: 0.000000
179
+ 2023-10-14 21:28:21,842 epoch 8 - iter 360/1809 - loss 0.00775403 - time (sec): 22.46 - samples/sec: 3402.77 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-14 21:28:32,860 epoch 8 - iter 540/1809 - loss 0.00765121 - time (sec): 33.47 - samples/sec: 3380.13 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-14 21:28:43,863 epoch 8 - iter 720/1809 - loss 0.00901281 - time (sec): 44.48 - samples/sec: 3403.24 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-14 21:28:54,924 epoch 8 - iter 900/1809 - loss 0.00858084 - time (sec): 55.54 - samples/sec: 3420.72 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-14 21:29:06,065 epoch 8 - iter 1080/1809 - loss 0.00865634 - time (sec): 66.68 - samples/sec: 3409.39 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-14 21:29:16,924 epoch 8 - iter 1260/1809 - loss 0.00989641 - time (sec): 77.54 - samples/sec: 3417.94 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-14 21:29:27,925 epoch 8 - iter 1440/1809 - loss 0.01016223 - time (sec): 88.54 - samples/sec: 3422.99 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-14 21:29:38,969 epoch 8 - iter 1620/1809 - loss 0.01064998 - time (sec): 99.58 - samples/sec: 3422.54 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-14 21:29:50,129 epoch 8 - iter 1800/1809 - loss 0.01033000 - time (sec): 110.74 - samples/sec: 3417.04 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-14 21:29:50,633 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-14 21:29:50,633 EPOCH 8 done: loss 0.0103 - lr: 0.000007
190
+ 2023-10-14 21:29:56,979 DEV : loss 0.38320016860961914 - f1-score (micro avg) 0.6464
191
+ 2023-10-14 21:29:57,010 saving best model
192
+ 2023-10-14 21:29:57,602 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-14 21:30:08,728 epoch 9 - iter 180/1809 - loss 0.00972321 - time (sec): 11.12 - samples/sec: 3412.05 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-14 21:30:19,700 epoch 9 - iter 360/1809 - loss 0.00656536 - time (sec): 22.10 - samples/sec: 3439.93 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-14 21:30:31,056 epoch 9 - iter 540/1809 - loss 0.00671814 - time (sec): 33.45 - samples/sec: 3452.06 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-14 21:30:42,103 epoch 9 - iter 720/1809 - loss 0.00664463 - time (sec): 44.50 - samples/sec: 3426.80 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-14 21:30:52,969 epoch 9 - iter 900/1809 - loss 0.00691173 - time (sec): 55.37 - samples/sec: 3419.30 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-14 21:31:03,885 epoch 9 - iter 1080/1809 - loss 0.00690807 - time (sec): 66.28 - samples/sec: 3415.64 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-14 21:31:14,931 epoch 9 - iter 1260/1809 - loss 0.00726721 - time (sec): 77.33 - samples/sec: 3419.26 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-14 21:31:25,828 epoch 9 - iter 1440/1809 - loss 0.00726244 - time (sec): 88.22 - samples/sec: 3421.98 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-14 21:31:36,948 epoch 9 - iter 1620/1809 - loss 0.00705247 - time (sec): 99.34 - samples/sec: 3423.83 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-14 21:31:48,785 epoch 9 - iter 1800/1809 - loss 0.00696297 - time (sec): 111.18 - samples/sec: 3402.85 - lr: 0.000003 - momentum: 0.000000
203
+ 2023-10-14 21:31:49,342 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-14 21:31:49,342 EPOCH 9 done: loss 0.0070 - lr: 0.000003
205
+ 2023-10-14 21:31:54,979 DEV : loss 0.38886281847953796 - f1-score (micro avg) 0.6445
206
+ 2023-10-14 21:31:55,023 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-14 21:32:08,159 epoch 10 - iter 180/1809 - loss 0.00786543 - time (sec): 13.14 - samples/sec: 2875.89 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-14 21:32:19,712 epoch 10 - iter 360/1809 - loss 0.00565639 - time (sec): 24.69 - samples/sec: 3046.75 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-14 21:32:31,193 epoch 10 - iter 540/1809 - loss 0.00526643 - time (sec): 36.17 - samples/sec: 3136.28 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-14 21:32:42,536 epoch 10 - iter 720/1809 - loss 0.00464233 - time (sec): 47.51 - samples/sec: 3170.36 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-14 21:32:54,160 epoch 10 - iter 900/1809 - loss 0.00541164 - time (sec): 59.14 - samples/sec: 3196.61 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 21:33:04,934 epoch 10 - iter 1080/1809 - loss 0.00573972 - time (sec): 69.91 - samples/sec: 3225.60 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-14 21:33:15,716 epoch 10 - iter 1260/1809 - loss 0.00553462 - time (sec): 80.69 - samples/sec: 3260.00 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 21:33:27,092 epoch 10 - iter 1440/1809 - loss 0.00551375 - time (sec): 92.07 - samples/sec: 3286.62 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-14 21:33:37,913 epoch 10 - iter 1620/1809 - loss 0.00525782 - time (sec): 102.89 - samples/sec: 3308.15 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-14 21:33:49,470 epoch 10 - iter 1800/1809 - loss 0.00511494 - time (sec): 114.45 - samples/sec: 3304.67 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-14 21:33:49,989 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-14 21:33:49,990 EPOCH 10 done: loss 0.0051 - lr: 0.000000
219
+ 2023-10-14 21:33:55,618 DEV : loss 0.4081748127937317 - f1-score (micro avg) 0.6428
220
+ 2023-10-14 21:33:56,145 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-14 21:33:56,146 Loading model from best epoch ...
222
+ 2023-10-14 21:33:57,658 SequenceTagger predicts: Dictionary with 13 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
223
+ 2023-10-14 21:34:05,332
224
+ Results:
225
+ - F-score (micro) 0.6527
226
+ - F-score (macro) 0.5176
227
+ - Accuracy 0.4984
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ loc 0.6280 0.7970 0.7025 591
233
+ pers 0.5703 0.7843 0.6604 357
234
+ org 0.1899 0.1899 0.1899 79
235
+
236
+ micro avg 0.5803 0.7459 0.6527 1027
237
+ macro avg 0.4627 0.5904 0.5176 1027
238
+ weighted avg 0.5742 0.7459 0.6484 1027
239
+
240
+ 2023-10-14 21:34:05,332 ----------------------------------------------------------------------------------------------------