Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
c4522f9
•
1 Parent(s): ba784a5

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 +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79b5bec185c334d2780da078cfed9b6c802cf334294c6c5c98b8025bf0d69db4
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 00:12:33 0.0000 0.3997 0.1048 0.7036 0.6821 0.6927 0.5423
3
+ 2 00:13:35 0.0000 0.1006 0.0931 0.6914 0.7805 0.7333 0.5995
4
+ 3 00:14:38 0.0000 0.0689 0.1079 0.7596 0.7647 0.7621 0.6294
5
+ 4 00:15:41 0.0000 0.0501 0.1288 0.7388 0.7647 0.7515 0.6259
6
+ 5 00:16:44 0.0000 0.0380 0.1583 0.7322 0.7579 0.7449 0.6136
7
+ 6 00:17:47 0.0000 0.0304 0.1779 0.7381 0.7749 0.7561 0.6261
8
+ 7 00:18:49 0.0000 0.0217 0.2032 0.7446 0.7783 0.7611 0.6318
9
+ 8 00:19:52 0.0000 0.0174 0.2113 0.7311 0.7783 0.7540 0.6249
10
+ 9 00:20:55 0.0000 0.0127 0.2239 0.7428 0.7839 0.7628 0.6364
11
+ 10 00:21:58 0.0000 0.0091 0.2320 0.7359 0.7817 0.7581 0.6293
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 00:11:30,961 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 00:11:30,962 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 00:11:30,962 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 00:11:30,962 MultiCorpus: 7936 train + 992 dev + 992 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
53
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 00:11:30,963 Train: 7936 sentences
55
+ 2023-10-14 00:11:30,963 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 00:11:30,963 Training Params:
58
+ 2023-10-14 00:11:30,963 - learning_rate: "3e-05"
59
+ 2023-10-14 00:11:30,963 - mini_batch_size: "8"
60
+ 2023-10-14 00:11:30,963 - max_epochs: "10"
61
+ 2023-10-14 00:11:30,963 - shuffle: "True"
62
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 00:11:30,963 Plugins:
64
+ 2023-10-14 00:11:30,963 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 00:11:30,963 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 00:11:30,963 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 00:11:30,963 Computation:
70
+ 2023-10-14 00:11:30,963 - compute on device: cuda:0
71
+ 2023-10-14 00:11:30,963 - embedding storage: none
72
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 00:11:30,963 Model training base path: "hmbench-icdar/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 00:11:30,963 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 00:11:36,677 epoch 1 - iter 99/992 - loss 2.16888144 - time (sec): 5.71 - samples/sec: 2851.21 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 00:11:42,673 epoch 1 - iter 198/992 - loss 1.30775263 - time (sec): 11.71 - samples/sec: 2765.68 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 00:11:48,333 epoch 1 - iter 297/992 - loss 0.97103134 - time (sec): 17.37 - samples/sec: 2773.68 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 00:11:54,110 epoch 1 - iter 396/992 - loss 0.78162493 - time (sec): 23.15 - samples/sec: 2788.71 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 00:12:00,171 epoch 1 - iter 495/992 - loss 0.65918233 - time (sec): 29.21 - samples/sec: 2785.29 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 00:12:06,385 epoch 1 - iter 594/992 - loss 0.56584426 - time (sec): 35.42 - samples/sec: 2806.86 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 00:12:12,145 epoch 1 - iter 693/992 - loss 0.50932720 - time (sec): 41.18 - samples/sec: 2803.88 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 00:12:17,924 epoch 1 - iter 792/992 - loss 0.46331013 - time (sec): 46.96 - samples/sec: 2807.39 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 00:12:23,883 epoch 1 - iter 891/992 - loss 0.42855665 - time (sec): 52.92 - samples/sec: 2786.75 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 00:12:29,901 epoch 1 - iter 990/992 - loss 0.40009945 - time (sec): 58.94 - samples/sec: 2776.32 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 00:12:30,030 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 00:12:30,030 EPOCH 1 done: loss 0.3997 - lr: 0.000030
88
+ 2023-10-14 00:12:33,137 DEV : loss 0.10478747636079788 - f1-score (micro avg) 0.6927
89
+ 2023-10-14 00:12:33,157 saving best model
90
+ 2023-10-14 00:12:33,553 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 00:12:39,339 epoch 2 - iter 99/992 - loss 0.10978478 - time (sec): 5.78 - samples/sec: 2768.41 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 00:12:45,165 epoch 2 - iter 198/992 - loss 0.10476447 - time (sec): 11.61 - samples/sec: 2801.81 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 00:12:51,147 epoch 2 - iter 297/992 - loss 0.11044910 - time (sec): 17.59 - samples/sec: 2782.40 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 00:12:57,135 epoch 2 - iter 396/992 - loss 0.10965807 - time (sec): 23.58 - samples/sec: 2776.96 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 00:13:02,983 epoch 2 - iter 495/992 - loss 0.10888826 - time (sec): 29.43 - samples/sec: 2783.86 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 00:13:08,856 epoch 2 - iter 594/992 - loss 0.10608504 - time (sec): 35.30 - samples/sec: 2790.49 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 00:13:14,595 epoch 2 - iter 693/992 - loss 0.10585318 - time (sec): 41.04 - samples/sec: 2797.08 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 00:13:20,316 epoch 2 - iter 792/992 - loss 0.10509565 - time (sec): 46.76 - samples/sec: 2800.79 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 00:13:26,191 epoch 2 - iter 891/992 - loss 0.10264619 - time (sec): 52.64 - samples/sec: 2804.00 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 00:13:31,994 epoch 2 - iter 990/992 - loss 0.10053276 - time (sec): 58.44 - samples/sec: 2801.72 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 00:13:32,111 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 00:13:32,111 EPOCH 2 done: loss 0.1006 - lr: 0.000027
103
+ 2023-10-14 00:13:35,929 DEV : loss 0.09312836080789566 - f1-score (micro avg) 0.7333
104
+ 2023-10-14 00:13:35,949 saving best model
105
+ 2023-10-14 00:13:36,462 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 00:13:42,202 epoch 3 - iter 99/992 - loss 0.07010417 - time (sec): 5.73 - samples/sec: 2791.40 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 00:13:48,040 epoch 3 - iter 198/992 - loss 0.07383171 - time (sec): 11.57 - samples/sec: 2756.00 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 00:13:54,070 epoch 3 - iter 297/992 - loss 0.07242296 - time (sec): 17.60 - samples/sec: 2773.20 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-14 00:14:00,119 epoch 3 - iter 396/992 - loss 0.07090486 - time (sec): 23.65 - samples/sec: 2789.63 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 00:14:05,834 epoch 3 - iter 495/992 - loss 0.07184703 - time (sec): 29.37 - samples/sec: 2798.87 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 00:14:11,540 epoch 3 - iter 594/992 - loss 0.07267068 - time (sec): 35.07 - samples/sec: 2795.08 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-14 00:14:17,249 epoch 3 - iter 693/992 - loss 0.07128641 - time (sec): 40.78 - samples/sec: 2802.20 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 00:14:23,545 epoch 3 - iter 792/992 - loss 0.06989991 - time (sec): 47.08 - samples/sec: 2785.56 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 00:14:29,379 epoch 3 - iter 891/992 - loss 0.06907457 - time (sec): 52.91 - samples/sec: 2787.11 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-14 00:14:35,168 epoch 3 - iter 990/992 - loss 0.06875863 - time (sec): 58.70 - samples/sec: 2790.50 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-14 00:14:35,276 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 00:14:35,276 EPOCH 3 done: loss 0.0689 - lr: 0.000023
118
+ 2023-10-14 00:14:38,661 DEV : loss 0.10794839262962341 - f1-score (micro avg) 0.7621
119
+ 2023-10-14 00:14:38,682 saving best model
120
+ 2023-10-14 00:14:39,146 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 00:14:44,907 epoch 4 - iter 99/992 - loss 0.05337231 - time (sec): 5.76 - samples/sec: 2764.08 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 00:14:50,927 epoch 4 - iter 198/992 - loss 0.05465630 - time (sec): 11.78 - samples/sec: 2778.52 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-14 00:14:56,532 epoch 4 - iter 297/992 - loss 0.05188037 - time (sec): 17.38 - samples/sec: 2746.41 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 00:15:02,501 epoch 4 - iter 396/992 - loss 0.04937150 - time (sec): 23.35 - samples/sec: 2756.34 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 00:15:08,435 epoch 4 - iter 495/992 - loss 0.04928861 - time (sec): 29.29 - samples/sec: 2764.20 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-14 00:15:14,444 epoch 4 - iter 594/992 - loss 0.05163498 - time (sec): 35.30 - samples/sec: 2772.33 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 00:15:20,289 epoch 4 - iter 693/992 - loss 0.05124858 - time (sec): 41.14 - samples/sec: 2769.38 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 00:15:26,014 epoch 4 - iter 792/992 - loss 0.05078560 - time (sec): 46.87 - samples/sec: 2769.19 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-14 00:15:31,880 epoch 4 - iter 891/992 - loss 0.05119365 - time (sec): 52.73 - samples/sec: 2775.78 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 00:15:38,047 epoch 4 - iter 990/992 - loss 0.05007817 - time (sec): 58.90 - samples/sec: 2779.27 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-14 00:15:38,173 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 00:15:38,173 EPOCH 4 done: loss 0.0501 - lr: 0.000020
133
+ 2023-10-14 00:15:41,595 DEV : loss 0.1288062185049057 - f1-score (micro avg) 0.7515
134
+ 2023-10-14 00:15:41,617 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-14 00:15:48,036 epoch 5 - iter 99/992 - loss 0.03956437 - time (sec): 6.42 - samples/sec: 2515.57 - lr: 0.000020 - momentum: 0.000000
136
+ 2023-10-14 00:15:53,966 epoch 5 - iter 198/992 - loss 0.04134074 - time (sec): 12.35 - samples/sec: 2659.88 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-14 00:15:59,898 epoch 5 - iter 297/992 - loss 0.03664301 - time (sec): 18.28 - samples/sec: 2733.19 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 00:16:05,516 epoch 5 - iter 396/992 - loss 0.03655408 - time (sec): 23.90 - samples/sec: 2747.72 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-14 00:16:11,231 epoch 5 - iter 495/992 - loss 0.03738088 - time (sec): 29.61 - samples/sec: 2751.25 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-14 00:16:17,186 epoch 5 - iter 594/992 - loss 0.03701123 - time (sec): 35.57 - samples/sec: 2758.52 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 00:16:23,114 epoch 5 - iter 693/992 - loss 0.03771113 - time (sec): 41.50 - samples/sec: 2769.18 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-14 00:16:29,178 epoch 5 - iter 792/992 - loss 0.03792863 - time (sec): 47.56 - samples/sec: 2777.70 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-14 00:16:34,887 epoch 5 - iter 891/992 - loss 0.03729411 - time (sec): 53.27 - samples/sec: 2777.67 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 00:16:40,563 epoch 5 - iter 990/992 - loss 0.03803926 - time (sec): 58.94 - samples/sec: 2775.12 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-14 00:16:40,692 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-14 00:16:40,692 EPOCH 5 done: loss 0.0380 - lr: 0.000017
147
+ 2023-10-14 00:16:44,223 DEV : loss 0.15828128159046173 - f1-score (micro avg) 0.7449
148
+ 2023-10-14 00:16:44,247 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-14 00:16:50,655 epoch 6 - iter 99/992 - loss 0.02650741 - time (sec): 6.41 - samples/sec: 2724.73 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-14 00:16:56,568 epoch 6 - iter 198/992 - loss 0.02855856 - time (sec): 12.32 - samples/sec: 2700.12 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-14 00:17:02,403 epoch 6 - iter 297/992 - loss 0.02827081 - time (sec): 18.15 - samples/sec: 2690.41 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-14 00:17:08,357 epoch 6 - iter 396/992 - loss 0.02948653 - time (sec): 24.11 - samples/sec: 2710.97 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-14 00:17:14,052 epoch 6 - iter 495/992 - loss 0.02954967 - time (sec): 29.80 - samples/sec: 2722.51 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-14 00:17:19,690 epoch 6 - iter 594/992 - loss 0.02941991 - time (sec): 35.44 - samples/sec: 2738.41 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-14 00:17:25,780 epoch 6 - iter 693/992 - loss 0.02883939 - time (sec): 41.53 - samples/sec: 2748.31 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-14 00:17:31,687 epoch 6 - iter 792/992 - loss 0.02966393 - time (sec): 47.44 - samples/sec: 2750.43 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-14 00:17:37,543 epoch 6 - iter 891/992 - loss 0.03020474 - time (sec): 53.29 - samples/sec: 2759.23 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-14 00:17:43,521 epoch 6 - iter 990/992 - loss 0.03041067 - time (sec): 59.27 - samples/sec: 2762.40 - lr: 0.000013 - momentum: 0.000000
159
+ 2023-10-14 00:17:43,631 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-14 00:17:43,631 EPOCH 6 done: loss 0.0304 - lr: 0.000013
161
+ 2023-10-14 00:17:47,078 DEV : loss 0.17794281244277954 - f1-score (micro avg) 0.7561
162
+ 2023-10-14 00:17:47,099 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-14 00:17:53,451 epoch 7 - iter 99/992 - loss 0.01788615 - time (sec): 6.35 - samples/sec: 2633.95 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-14 00:17:59,328 epoch 7 - iter 198/992 - loss 0.01786951 - time (sec): 12.23 - samples/sec: 2710.07 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-14 00:18:05,363 epoch 7 - iter 297/992 - loss 0.01689469 - time (sec): 18.26 - samples/sec: 2726.74 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-14 00:18:10,955 epoch 7 - iter 396/992 - loss 0.01790618 - time (sec): 23.85 - samples/sec: 2728.73 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-14 00:18:16,903 epoch 7 - iter 495/992 - loss 0.02065770 - time (sec): 29.80 - samples/sec: 2752.54 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-14 00:18:22,478 epoch 7 - iter 594/992 - loss 0.02009396 - time (sec): 35.38 - samples/sec: 2772.49 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-14 00:18:28,300 epoch 7 - iter 693/992 - loss 0.01954440 - time (sec): 41.20 - samples/sec: 2768.82 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-14 00:18:34,153 epoch 7 - iter 792/992 - loss 0.02049321 - time (sec): 47.05 - samples/sec: 2768.46 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-14 00:18:40,247 epoch 7 - iter 891/992 - loss 0.02112247 - time (sec): 53.15 - samples/sec: 2763.57 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-14 00:18:46,278 epoch 7 - iter 990/992 - loss 0.02172067 - time (sec): 59.18 - samples/sec: 2767.03 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-14 00:18:46,396 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-14 00:18:46,397 EPOCH 7 done: loss 0.0217 - lr: 0.000010
175
+ 2023-10-14 00:18:49,841 DEV : loss 0.2032385766506195 - f1-score (micro avg) 0.7611
176
+ 2023-10-14 00:18:49,862 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-14 00:18:56,047 epoch 8 - iter 99/992 - loss 0.02018754 - time (sec): 6.18 - samples/sec: 2756.01 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-14 00:19:01,853 epoch 8 - iter 198/992 - loss 0.02020446 - time (sec): 11.99 - samples/sec: 2781.75 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-14 00:19:07,865 epoch 8 - iter 297/992 - loss 0.01858499 - time (sec): 18.00 - samples/sec: 2803.77 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-14 00:19:13,546 epoch 8 - iter 396/992 - loss 0.01820929 - time (sec): 23.68 - samples/sec: 2818.80 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-14 00:19:19,159 epoch 8 - iter 495/992 - loss 0.01766479 - time (sec): 29.30 - samples/sec: 2801.97 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-14 00:19:25,097 epoch 8 - iter 594/992 - loss 0.01862023 - time (sec): 35.23 - samples/sec: 2794.09 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-14 00:19:30,908 epoch 8 - iter 693/992 - loss 0.01845761 - time (sec): 41.05 - samples/sec: 2797.91 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-14 00:19:36,976 epoch 8 - iter 792/992 - loss 0.01775140 - time (sec): 47.11 - samples/sec: 2790.08 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-14 00:19:42,981 epoch 8 - iter 891/992 - loss 0.01742615 - time (sec): 53.12 - samples/sec: 2788.32 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-14 00:19:48,628 epoch 8 - iter 990/992 - loss 0.01748644 - time (sec): 58.77 - samples/sec: 2784.70 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-14 00:19:48,745 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-14 00:19:48,745 EPOCH 8 done: loss 0.0174 - lr: 0.000007
189
+ 2023-10-14 00:19:52,565 DEV : loss 0.21129660308361053 - f1-score (micro avg) 0.754
190
+ 2023-10-14 00:19:52,586 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-14 00:19:58,233 epoch 9 - iter 99/992 - loss 0.01582445 - time (sec): 5.65 - samples/sec: 2818.03 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-14 00:20:04,200 epoch 9 - iter 198/992 - loss 0.01645995 - time (sec): 11.61 - samples/sec: 2778.98 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-14 00:20:10,220 epoch 9 - iter 297/992 - loss 0.01506388 - time (sec): 17.63 - samples/sec: 2801.59 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-14 00:20:16,226 epoch 9 - iter 396/992 - loss 0.01438323 - time (sec): 23.64 - samples/sec: 2789.87 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-14 00:20:22,248 epoch 9 - iter 495/992 - loss 0.01331515 - time (sec): 29.66 - samples/sec: 2767.81 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-14 00:20:28,138 epoch 9 - iter 594/992 - loss 0.01331908 - time (sec): 35.55 - samples/sec: 2771.62 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-14 00:20:33,994 epoch 9 - iter 693/992 - loss 0.01304711 - time (sec): 41.41 - samples/sec: 2765.58 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-14 00:20:39,654 epoch 9 - iter 792/992 - loss 0.01359041 - time (sec): 47.07 - samples/sec: 2781.33 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-14 00:20:45,656 epoch 9 - iter 891/992 - loss 0.01300277 - time (sec): 53.07 - samples/sec: 2769.11 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-14 00:20:51,582 epoch 9 - iter 990/992 - loss 0.01274266 - time (sec): 58.99 - samples/sec: 2773.76 - lr: 0.000003 - momentum: 0.000000
201
+ 2023-10-14 00:20:51,708 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-14 00:20:51,708 EPOCH 9 done: loss 0.0127 - lr: 0.000003
203
+ 2023-10-14 00:20:55,153 DEV : loss 0.2239181101322174 - f1-score (micro avg) 0.7628
204
+ 2023-10-14 00:20:55,174 saving best model
205
+ 2023-10-14 00:20:55,710 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-14 00:21:02,160 epoch 10 - iter 99/992 - loss 0.01038855 - time (sec): 6.45 - samples/sec: 2660.63 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-14 00:21:07,910 epoch 10 - iter 198/992 - loss 0.00863248 - time (sec): 12.20 - samples/sec: 2721.99 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-14 00:21:13,805 epoch 10 - iter 297/992 - loss 0.00888013 - time (sec): 18.09 - samples/sec: 2752.19 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-14 00:21:19,853 epoch 10 - iter 396/992 - loss 0.00922199 - time (sec): 24.14 - samples/sec: 2764.40 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-14 00:21:25,836 epoch 10 - iter 495/992 - loss 0.00894249 - time (sec): 30.12 - samples/sec: 2789.05 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-14 00:21:31,545 epoch 10 - iter 594/992 - loss 0.00915085 - time (sec): 35.83 - samples/sec: 2792.86 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-14 00:21:37,055 epoch 10 - iter 693/992 - loss 0.00962406 - time (sec): 41.34 - samples/sec: 2793.08 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-14 00:21:42,693 epoch 10 - iter 792/992 - loss 0.00971077 - time (sec): 46.98 - samples/sec: 2786.25 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 00:21:48,490 epoch 10 - iter 891/992 - loss 0.00941827 - time (sec): 52.78 - samples/sec: 2782.46 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-14 00:21:54,375 epoch 10 - iter 990/992 - loss 0.00906870 - time (sec): 58.66 - samples/sec: 2790.27 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-14 00:21:54,482 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-14 00:21:54,482 EPOCH 10 done: loss 0.0091 - lr: 0.000000
218
+ 2023-10-14 00:21:58,105 DEV : loss 0.23203293979167938 - f1-score (micro avg) 0.7581
219
+ 2023-10-14 00:21:58,572 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-14 00:21:58,574 Loading model from best epoch ...
221
+ 2023-10-14 00:22:00,398 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG
222
+ 2023-10-14 00:22:03,335
223
+ Results:
224
+ - F-score (micro) 0.7764
225
+ - F-score (macro) 0.692
226
+ - Accuracy 0.656
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ LOC 0.8143 0.8504 0.8320 655
232
+ PER 0.7244 0.8251 0.7715 223
233
+ ORG 0.5091 0.4409 0.4726 127
234
+
235
+ micro avg 0.7605 0.7930 0.7764 1005
236
+ macro avg 0.6826 0.7055 0.6920 1005
237
+ weighted avg 0.7558 0.7930 0.7731 1005
238
+
239
+ 2023-10-14 00:22:03,335 ----------------------------------------------------------------------------------------------------