Hugging Face's logo

stefan-it
/
hmbench-hipe2020-fr-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
679e0aa
·
1 Parent(s): 65af442

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 +243 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62312c273b53722326c31a548ac55aab9d260146d05167473aefab77793b8570
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 18:26:51 0.0000 0.6037 0.1516 0.6483 0.7222 0.6833 0.5504
3
+ 2 18:27:52 0.0000 0.1311 0.1123 0.7531 0.8351 0.7920 0.6816
4
+ 3 18:28:53 0.0000 0.0726 0.1180 0.7924 0.8087 0.8005 0.6932
5
+ 4 18:29:55 0.0000 0.0466 0.1432 0.8055 0.8253 0.8153 0.7127
6
+ 5 18:30:57 0.0000 0.0344 0.1619 0.8276 0.8356 0.8316 0.7376
7
+ 6 18:31:59 0.0000 0.0256 0.1881 0.8017 0.8333 0.8172 0.7182
8
+ 7 18:32:59 0.0000 0.0192 0.2107 0.8125 0.8288 0.8205 0.7239
9
+ 8 18:34:00 0.0000 0.0125 0.1914 0.8178 0.8482 0.8327 0.7379
10
+ 9 18:35:02 0.0000 0.0089 0.2076 0.8109 0.8425 0.8264 0.7344
11
+ 10 18:36:03 0.0000 0.0072 0.2092 0.8234 0.8414 0.8323 0.7412
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 18:25:55,574 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 18:25:55,575 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 18:25:55,575 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 18:25:55,575 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 18:25:55,575 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 18:25:55,575 Train: 5901 sentences
55
+ 2023-10-13 18:25:55,575 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 18:25:55,576 Training Params:
58
+ 2023-10-13 18:25:55,576 - learning_rate: "3e-05"
59
+ 2023-10-13 18:25:55,576 - mini_batch_size: "8"
60
+ 2023-10-13 18:25:55,576 - max_epochs: "10"
61
+ 2023-10-13 18:25:55,576 - shuffle: "True"
62
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 18:25:55,576 Plugins:
64
+ 2023-10-13 18:25:55,576 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 18:25:55,576 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 18:25:55,576 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 18:25:55,576 Computation:
70
+ 2023-10-13 18:25:55,576 - compute on device: cuda:0
71
+ 2023-10-13 18:25:55,576 - embedding storage: none
72
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 18:25:55,576 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 18:25:55,576 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 18:26:00,188 epoch 1 - iter 73/738 - loss 2.77948811 - time (sec): 4.61 - samples/sec: 3504.97 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-13 18:26:04,956 epoch 1 - iter 146/738 - loss 1.83651071 - time (sec): 9.38 - samples/sec: 3464.52 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-13 18:26:09,786 epoch 1 - iter 219/738 - loss 1.38880604 - time (sec): 14.21 - samples/sec: 3420.58 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-13 18:26:14,516 epoch 1 - iter 292/738 - loss 1.13857844 - time (sec): 18.94 - samples/sec: 3407.12 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-13 18:26:19,737 epoch 1 - iter 365/738 - loss 0.98269160 - time (sec): 24.16 - samples/sec: 3360.28 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-13 18:26:24,365 epoch 1 - iter 438/738 - loss 0.87412221 - time (sec): 28.79 - samples/sec: 3348.74 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-13 18:26:29,369 epoch 1 - iter 511/738 - loss 0.78141679 - time (sec): 33.79 - samples/sec: 3367.22 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-13 18:26:34,377 epoch 1 - iter 584/738 - loss 0.70772215 - time (sec): 38.80 - samples/sec: 3383.66 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-13 18:26:39,145 epoch 1 - iter 657/738 - loss 0.65339163 - time (sec): 43.57 - samples/sec: 3387.69 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-13 18:26:44,294 epoch 1 - iter 730/738 - loss 0.60764205 - time (sec): 48.72 - samples/sec: 3380.12 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-13 18:26:44,855 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 18:26:44,856 EPOCH 1 done: loss 0.6037 - lr: 0.000030
88
+ 2023-10-13 18:26:51,028 DEV : loss 0.15163348615169525 - f1-score (micro avg) 0.6833
89
+ 2023-10-13 18:26:51,056 saving best model
90
+ 2023-10-13 18:26:51,480 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 18:26:56,596 epoch 2 - iter 73/738 - loss 0.14905787 - time (sec): 5.11 - samples/sec: 3263.37 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-13 18:27:01,383 epoch 2 - iter 146/738 - loss 0.14670637 - time (sec): 9.90 - samples/sec: 3334.54 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-13 18:27:06,404 epoch 2 - iter 219/738 - loss 0.14929708 - time (sec): 14.92 - samples/sec: 3331.31 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-13 18:27:10,955 epoch 2 - iter 292/738 - loss 0.14176362 - time (sec): 19.47 - samples/sec: 3345.22 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-13 18:27:16,010 epoch 2 - iter 365/738 - loss 0.13682994 - time (sec): 24.53 - samples/sec: 3390.21 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-13 18:27:22,282 epoch 2 - iter 438/738 - loss 0.13774552 - time (sec): 30.80 - samples/sec: 3372.32 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-13 18:27:26,751 epoch 2 - iter 511/738 - loss 0.13433078 - time (sec): 35.27 - samples/sec: 3375.37 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-13 18:27:31,778 epoch 2 - iter 584/738 - loss 0.13450948 - time (sec): 40.30 - samples/sec: 3373.81 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-13 18:27:35,982 epoch 2 - iter 657/738 - loss 0.13257424 - time (sec): 44.50 - samples/sec: 3378.23 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-13 18:27:40,535 epoch 2 - iter 730/738 - loss 0.13177134 - time (sec): 49.05 - samples/sec: 3363.30 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-13 18:27:40,955 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 18:27:40,956 EPOCH 2 done: loss 0.1311 - lr: 0.000027
103
+ 2023-10-13 18:27:52,212 DEV : loss 0.11227148026227951 - f1-score (micro avg) 0.792
104
+ 2023-10-13 18:27:52,241 saving best model
105
+ 2023-10-13 18:27:52,725 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 18:27:57,928 epoch 3 - iter 73/738 - loss 0.07307088 - time (sec): 5.20 - samples/sec: 3558.65 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-13 18:28:03,006 epoch 3 - iter 146/738 - loss 0.06721425 - time (sec): 10.27 - samples/sec: 3411.70 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-13 18:28:07,946 epoch 3 - iter 219/738 - loss 0.07015625 - time (sec): 15.21 - samples/sec: 3402.24 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-13 18:28:13,475 epoch 3 - iter 292/738 - loss 0.07577362 - time (sec): 20.74 - samples/sec: 3372.37 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-13 18:28:18,038 epoch 3 - iter 365/738 - loss 0.07658825 - time (sec): 25.31 - samples/sec: 3364.70 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-13 18:28:22,940 epoch 3 - iter 438/738 - loss 0.07349264 - time (sec): 30.21 - samples/sec: 3352.76 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-13 18:28:27,523 epoch 3 - iter 511/738 - loss 0.07289690 - time (sec): 34.79 - samples/sec: 3357.71 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-13 18:28:32,227 epoch 3 - iter 584/738 - loss 0.07223721 - time (sec): 39.49 - samples/sec: 3361.46 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-13 18:28:37,075 epoch 3 - iter 657/738 - loss 0.07202171 - time (sec): 44.34 - samples/sec: 3354.44 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-13 18:28:41,868 epoch 3 - iter 730/738 - loss 0.07214547 - time (sec): 49.14 - samples/sec: 3356.10 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-13 18:28:42,345 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 18:28:42,345 EPOCH 3 done: loss 0.0726 - lr: 0.000023
118
+ 2023-10-13 18:28:53,535 DEV : loss 0.11804373562335968 - f1-score (micro avg) 0.8005
119
+ 2023-10-13 18:28:53,563 saving best model
120
+ 2023-10-13 18:28:54,108 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 18:28:59,136 epoch 4 - iter 73/738 - loss 0.04640357 - time (sec): 5.02 - samples/sec: 3178.15 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-13 18:29:04,912 epoch 4 - iter 146/738 - loss 0.05188936 - time (sec): 10.80 - samples/sec: 3286.27 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-13 18:29:10,063 epoch 4 - iter 219/738 - loss 0.04957858 - time (sec): 15.95 - samples/sec: 3226.45 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-13 18:29:14,644 epoch 4 - iter 292/738 - loss 0.05013571 - time (sec): 20.53 - samples/sec: 3217.01 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-13 18:29:19,570 epoch 4 - iter 365/738 - loss 0.04848137 - time (sec): 25.46 - samples/sec: 3251.80 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-13 18:29:24,227 epoch 4 - iter 438/738 - loss 0.04768082 - time (sec): 30.11 - samples/sec: 3283.31 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-13 18:29:28,806 epoch 4 - iter 511/738 - loss 0.04744479 - time (sec): 34.69 - samples/sec: 3284.39 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-13 18:29:33,378 epoch 4 - iter 584/738 - loss 0.04764204 - time (sec): 39.27 - samples/sec: 3303.23 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-13 18:29:38,231 epoch 4 - iter 657/738 - loss 0.04732965 - time (sec): 44.12 - samples/sec: 3300.01 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-13 18:29:43,735 epoch 4 - iter 730/738 - loss 0.04657406 - time (sec): 49.62 - samples/sec: 3319.62 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-13 18:29:44,198 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 18:29:44,198 EPOCH 4 done: loss 0.0466 - lr: 0.000020
133
+ 2023-10-13 18:29:55,341 DEV : loss 0.14324556291103363 - f1-score (micro avg) 0.8153
134
+ 2023-10-13 18:29:55,370 saving best model
135
+ 2023-10-13 18:29:55,914 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 18:30:00,745 epoch 5 - iter 73/738 - loss 0.03463417 - time (sec): 4.83 - samples/sec: 3466.69 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-13 18:30:05,323 epoch 5 - iter 146/738 - loss 0.03424645 - time (sec): 9.40 - samples/sec: 3330.81 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-13 18:30:10,542 epoch 5 - iter 219/738 - loss 0.03250383 - time (sec): 14.62 - samples/sec: 3279.94 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-13 18:30:15,573 epoch 5 - iter 292/738 - loss 0.03238949 - time (sec): 19.65 - samples/sec: 3278.44 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-13 18:30:20,752 epoch 5 - iter 365/738 - loss 0.03263209 - time (sec): 24.83 - samples/sec: 3282.40 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-13 18:30:25,879 epoch 5 - iter 438/738 - loss 0.03210855 - time (sec): 29.96 - samples/sec: 3280.02 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-13 18:30:30,870 epoch 5 - iter 511/738 - loss 0.03349194 - time (sec): 34.95 - samples/sec: 3286.12 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-13 18:30:36,087 epoch 5 - iter 584/738 - loss 0.03354639 - time (sec): 40.17 - samples/sec: 3282.79 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-13 18:30:41,586 epoch 5 - iter 657/738 - loss 0.03422851 - time (sec): 45.67 - samples/sec: 3256.83 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-13 18:30:46,158 epoch 5 - iter 730/738 - loss 0.03467585 - time (sec): 50.24 - samples/sec: 3276.48 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-13 18:30:46,772 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 18:30:46,773 EPOCH 5 done: loss 0.0344 - lr: 0.000017
148
+ 2023-10-13 18:30:57,925 DEV : loss 0.16191978752613068 - f1-score (micro avg) 0.8316
149
+ 2023-10-13 18:30:57,954 saving best model
150
+ 2023-10-13 18:30:58,448 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-13 18:31:03,446 epoch 6 - iter 73/738 - loss 0.03187475 - time (sec): 5.00 - samples/sec: 3395.25 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-13 18:31:08,103 epoch 6 - iter 146/738 - loss 0.02607637 - time (sec): 9.65 - samples/sec: 3340.93 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-13 18:31:12,977 epoch 6 - iter 219/738 - loss 0.02584962 - time (sec): 14.53 - samples/sec: 3330.61 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-13 18:31:17,890 epoch 6 - iter 292/738 - loss 0.02764999 - time (sec): 19.44 - samples/sec: 3322.47 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-13 18:31:22,652 epoch 6 - iter 365/738 - loss 0.02678067 - time (sec): 24.20 - samples/sec: 3311.91 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-13 18:31:27,144 epoch 6 - iter 438/738 - loss 0.02654707 - time (sec): 28.69 - samples/sec: 3322.25 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-13 18:31:32,359 epoch 6 - iter 511/738 - loss 0.02516894 - time (sec): 33.91 - samples/sec: 3334.53 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-13 18:31:37,209 epoch 6 - iter 584/738 - loss 0.02524705 - time (sec): 38.76 - samples/sec: 3337.29 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-13 18:31:42,168 epoch 6 - iter 657/738 - loss 0.02616650 - time (sec): 43.72 - samples/sec: 3335.83 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-13 18:31:47,146 epoch 6 - iter 730/738 - loss 0.02554515 - time (sec): 48.70 - samples/sec: 3369.83 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-13 18:31:47,871 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-13 18:31:47,871 EPOCH 6 done: loss 0.0256 - lr: 0.000013
163
+ 2023-10-13 18:31:59,141 DEV : loss 0.18813824653625488 - f1-score (micro avg) 0.8172
164
+ 2023-10-13 18:31:59,170 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-13 18:32:04,119 epoch 7 - iter 73/738 - loss 0.01945020 - time (sec): 4.95 - samples/sec: 3505.84 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-13 18:32:09,953 epoch 7 - iter 146/738 - loss 0.02190146 - time (sec): 10.78 - samples/sec: 3316.79 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-13 18:32:15,106 epoch 7 - iter 219/738 - loss 0.01858043 - time (sec): 15.93 - samples/sec: 3324.70 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-13 18:32:20,277 epoch 7 - iter 292/738 - loss 0.01910198 - time (sec): 21.11 - samples/sec: 3364.95 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-13 18:32:24,500 epoch 7 - iter 365/738 - loss 0.01972075 - time (sec): 25.33 - samples/sec: 3391.31 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-13 18:32:29,452 epoch 7 - iter 438/738 - loss 0.02062901 - time (sec): 30.28 - samples/sec: 3382.30 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-13 18:32:34,183 epoch 7 - iter 511/738 - loss 0.01984974 - time (sec): 35.01 - samples/sec: 3373.91 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-13 18:32:38,775 epoch 7 - iter 584/738 - loss 0.01906639 - time (sec): 39.60 - samples/sec: 3372.83 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-13 18:32:43,455 epoch 7 - iter 657/738 - loss 0.01891734 - time (sec): 44.28 - samples/sec: 3373.12 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-13 18:32:48,038 epoch 7 - iter 730/738 - loss 0.01895225 - time (sec): 48.87 - samples/sec: 3368.29 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-13 18:32:48,534 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-13 18:32:48,534 EPOCH 7 done: loss 0.0192 - lr: 0.000010
177
+ 2023-10-13 18:32:59,738 DEV : loss 0.21069413423538208 - f1-score (micro avg) 0.8205
178
+ 2023-10-13 18:32:59,768 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-13 18:33:04,738 epoch 8 - iter 73/738 - loss 0.01208919 - time (sec): 4.97 - samples/sec: 3546.99 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-13 18:33:09,349 epoch 8 - iter 146/738 - loss 0.01011562 - time (sec): 9.58 - samples/sec: 3428.49 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-13 18:33:14,711 epoch 8 - iter 219/738 - loss 0.01197427 - time (sec): 14.94 - samples/sec: 3453.24 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-13 18:33:19,729 epoch 8 - iter 292/738 - loss 0.01073791 - time (sec): 19.96 - samples/sec: 3356.40 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-13 18:33:24,003 epoch 8 - iter 365/738 - loss 0.01273368 - time (sec): 24.23 - samples/sec: 3359.97 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-13 18:33:28,829 epoch 8 - iter 438/738 - loss 0.01246504 - time (sec): 29.06 - samples/sec: 3347.32 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-13 18:33:33,827 epoch 8 - iter 511/738 - loss 0.01250396 - time (sec): 34.06 - samples/sec: 3368.55 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-13 18:33:38,771 epoch 8 - iter 584/738 - loss 0.01223467 - time (sec): 39.00 - samples/sec: 3356.88 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-13 18:33:43,688 epoch 8 - iter 657/738 - loss 0.01262949 - time (sec): 43.92 - samples/sec: 3350.21 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-13 18:33:48,706 epoch 8 - iter 730/738 - loss 0.01263386 - time (sec): 48.94 - samples/sec: 3352.69 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-13 18:33:49,394 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-13 18:33:49,395 EPOCH 8 done: loss 0.0125 - lr: 0.000007
191
+ 2023-10-13 18:34:00,541 DEV : loss 0.1914064884185791 - f1-score (micro avg) 0.8327
192
+ 2023-10-13 18:34:00,571 saving best model
193
+ 2023-10-13 18:34:01,118 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-13 18:34:05,943 epoch 9 - iter 73/738 - loss 0.01636345 - time (sec): 4.82 - samples/sec: 3310.78 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-13 18:34:10,703 epoch 9 - iter 146/738 - loss 0.01343032 - time (sec): 9.58 - samples/sec: 3354.67 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-13 18:34:15,064 epoch 9 - iter 219/738 - loss 0.01024467 - time (sec): 13.94 - samples/sec: 3361.07 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-13 18:34:20,316 epoch 9 - iter 292/738 - loss 0.01040151 - time (sec): 19.20 - samples/sec: 3322.59 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-13 18:34:25,011 epoch 9 - iter 365/738 - loss 0.00980650 - time (sec): 23.89 - samples/sec: 3318.50 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-13 18:34:29,967 epoch 9 - iter 438/738 - loss 0.00914193 - time (sec): 28.85 - samples/sec: 3308.67 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-13 18:34:35,040 epoch 9 - iter 511/738 - loss 0.00900502 - time (sec): 33.92 - samples/sec: 3333.78 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-13 18:34:40,548 epoch 9 - iter 584/738 - loss 0.00995733 - time (sec): 39.43 - samples/sec: 3334.23 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-13 18:34:45,124 epoch 9 - iter 657/738 - loss 0.00904693 - time (sec): 44.00 - samples/sec: 3336.14 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-13 18:34:49,897 epoch 9 - iter 730/738 - loss 0.00901017 - time (sec): 48.78 - samples/sec: 3355.24 - lr: 0.000003 - momentum: 0.000000
204
+ 2023-10-13 18:34:50,765 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-13 18:34:50,765 EPOCH 9 done: loss 0.0089 - lr: 0.000003
206
+ 2023-10-13 18:35:02,006 DEV : loss 0.20760603249073029 - f1-score (micro avg) 0.8264
207
+ 2023-10-13 18:35:02,036 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-13 18:35:06,611 epoch 10 - iter 73/738 - loss 0.00505717 - time (sec): 4.57 - samples/sec: 3318.85 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-13 18:35:11,830 epoch 10 - iter 146/738 - loss 0.00742873 - time (sec): 9.79 - samples/sec: 3315.90 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-13 18:35:17,652 epoch 10 - iter 219/738 - loss 0.00650006 - time (sec): 15.62 - samples/sec: 3154.32 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-13 18:35:22,973 epoch 10 - iter 292/738 - loss 0.00677815 - time (sec): 20.94 - samples/sec: 3170.26 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-13 18:35:28,293 epoch 10 - iter 365/738 - loss 0.00621081 - time (sec): 26.26 - samples/sec: 3208.44 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-13 18:35:32,788 epoch 10 - iter 438/738 - loss 0.00657852 - time (sec): 30.75 - samples/sec: 3245.23 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-13 18:35:37,205 epoch 10 - iter 511/738 - loss 0.00821402 - time (sec): 35.17 - samples/sec: 3273.58 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-13 18:35:42,271 epoch 10 - iter 584/738 - loss 0.00795187 - time (sec): 40.23 - samples/sec: 3268.92 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 18:35:47,325 epoch 10 - iter 657/738 - loss 0.00749875 - time (sec): 45.29 - samples/sec: 3296.73 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-13 18:35:51,858 epoch 10 - iter 730/738 - loss 0.00725227 - time (sec): 49.82 - samples/sec: 3308.48 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-13 18:35:52,320 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-13 18:35:52,320 EPOCH 10 done: loss 0.0072 - lr: 0.000000
220
+ 2023-10-13 18:36:03,499 DEV : loss 0.20923016965389252 - f1-score (micro avg) 0.8323
221
+ 2023-10-13 18:36:03,920 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-13 18:36:03,922 Loading model from best epoch ...
223
+ 2023-10-13 18:36:05,272 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
224
+ 2023-10-13 18:36:11,234
225
+ Results:
226
+ - F-score (micro) 0.8013
227
+ - F-score (macro) 0.7115
228
+ - Accuracy 0.6915
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ loc 0.8692 0.8671 0.8681 858
234
+ pers 0.7301 0.8212 0.7730 537
235
+ org 0.6170 0.6591 0.6374 132
236
+ prod 0.7119 0.6885 0.7000 61
237
+ time 0.5500 0.6111 0.5789 54
238
+
239
+ micro avg 0.7831 0.8203 0.8013 1642
240
+ macro avg 0.6956 0.7294 0.7115 1642
241
+ weighted avg 0.7871 0.8203 0.8027 1642
242
+
243
+ 2023-10-13 18:36:11,234 ----------------------------------------------------------------------------------------------------