Hugging Face's logo

hmbyt5-preliminary
/
flair-hipe-2022-letemps-fr

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
f29e458
•
1 Parent(s): a47e4c9

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. final-model.pt +3 -0
  4. loss.tsv +11 -0
  5. runs/events.out.tfevents.1697242526.c8b2203b18a8.2923.9 +3 -0
  6. test.tsv +0 -0
  7. training.log +265 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e554cb264b012afe642eeee6d6b62fb729fdc148b89303ba7ab57584b656dfc2
3
+ size 870793839
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
final-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc85154abbfc0978f2ddb900650645219d974cc314235f2b1870ee38fc84aa76
3
+ size 870793956
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 00:31:18 0.0002 0.7248 0.1359 0.5052 0.6064 0.5512 0.3941
3
+ 2 00:46:52 0.0001 0.0879 0.1020 0.5286 0.7185 0.6091 0.4441
4
+ 3 01:02:41 0.0001 0.0563 0.1395 0.5762 0.6705 0.6198 0.4575
5
+ 4 01:18:35 0.0001 0.0403 0.1925 0.5312 0.7803 0.6321 0.4703
6
+ 5 01:34:33 0.0001 0.0293 0.2196 0.5591 0.7632 0.6454 0.4847
7
+ 6 01:50:36 0.0001 0.0222 0.2577 0.5683 0.7952 0.6629 0.5040
8
+ 7 02:06:59 0.0001 0.0151 0.2952 0.5705 0.8009 0.6663 0.5072
9
+ 8 02:22:42 0.0000 0.0102 0.3187 0.5738 0.7872 0.6638 0.5044
10
+ 9 02:38:45 0.0000 0.0076 0.3398 0.5818 0.7895 0.6699 0.5111
11
+ 10 02:55:02 0.0000 0.0050 0.3481 0.5750 0.7895 0.6654 0.5074
runs/events.out.tfevents.1697242526.c8b2203b18a8.2923.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e074701b2dcfb2e1422eab625c4614edd9c05294f41445beb76c9f145155224
3
+ size 1018100
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 00:15:26,580 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 00:15:26,582 Model: "SequenceTagger(
3
+ (embeddings): ByT5Embeddings(
4
+ (model): T5EncoderModel(
5
+ (shared): Embedding(384, 1472)
6
+ (encoder): T5Stack(
7
+ (embed_tokens): Embedding(384, 1472)
8
+ (block): ModuleList(
9
+ (0): T5Block(
10
+ (layer): ModuleList(
11
+ (0): T5LayerSelfAttention(
12
+ (SelfAttention): T5Attention(
13
+ (q): Linear(in_features=1472, out_features=384, bias=False)
14
+ (k): Linear(in_features=1472, out_features=384, bias=False)
15
+ (v): Linear(in_features=1472, out_features=384, bias=False)
16
+ (o): Linear(in_features=384, out_features=1472, bias=False)
17
+ (relative_attention_bias): Embedding(32, 6)
18
+ )
19
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (1): T5LayerFF(
23
+ (DenseReluDense): T5DenseGatedActDense(
24
+ (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
25
+ (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
26
+ (wo): Linear(in_features=3584, out_features=1472, bias=False)
27
+ (dropout): Dropout(p=0.1, inplace=False)
28
+ (act): NewGELUActivation()
29
+ )
30
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
31
+ (dropout): Dropout(p=0.1, inplace=False)
32
+ )
33
+ )
34
+ )
35
+ (1-11): 11 x T5Block(
36
+ (layer): ModuleList(
37
+ (0): T5LayerSelfAttention(
38
+ (SelfAttention): T5Attention(
39
+ (q): Linear(in_features=1472, out_features=384, bias=False)
40
+ (k): Linear(in_features=1472, out_features=384, bias=False)
41
+ (v): Linear(in_features=1472, out_features=384, bias=False)
42
+ (o): Linear(in_features=384, out_features=1472, bias=False)
43
+ )
44
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
45
+ (dropout): Dropout(p=0.1, inplace=False)
46
+ )
47
+ (1): T5LayerFF(
48
+ (DenseReluDense): T5DenseGatedActDense(
49
+ (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
50
+ (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
51
+ (wo): Linear(in_features=3584, out_features=1472, bias=False)
52
+ (dropout): Dropout(p=0.1, inplace=False)
53
+ (act): NewGELUActivation()
54
+ )
55
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
56
+ (dropout): Dropout(p=0.1, inplace=False)
57
+ )
58
+ )
59
+ )
60
+ )
61
+ (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
62
+ (dropout): Dropout(p=0.1, inplace=False)
63
+ )
64
+ )
65
+ )
66
+ (locked_dropout): LockedDropout(p=0.5)
67
+ (linear): Linear(in_features=1472, out_features=13, bias=True)
68
+ (loss_function): CrossEntropyLoss()
69
+ )"
70
+ 2023-10-14 00:15:26,582 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-14 00:15:26,582 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
72
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
73
+ 2023-10-14 00:15:26,583 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-14 00:15:26,583 Train: 14465 sentences
75
+ 2023-10-14 00:15:26,583 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-14 00:15:26,583 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-14 00:15:26,583 Training Params:
78
+ 2023-10-14 00:15:26,583 - learning_rate: "0.00016"
79
+ 2023-10-14 00:15:26,583 - mini_batch_size: "8"
80
+ 2023-10-14 00:15:26,583 - max_epochs: "10"
81
+ 2023-10-14 00:15:26,583 - shuffle: "True"
82
+ 2023-10-14 00:15:26,583 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-14 00:15:26,583 Plugins:
84
+ 2023-10-14 00:15:26,583 - TensorboardLogger
85
+ 2023-10-14 00:15:26,583 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-14 00:15:26,584 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 00:15:26,584 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-14 00:15:26,584 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-14 00:15:26,584 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-14 00:15:26,584 Computation:
91
+ 2023-10-14 00:15:26,584 - compute on device: cuda:0
92
+ 2023-10-14 00:15:26,584 - embedding storage: none
93
+ 2023-10-14 00:15:26,584 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-14 00:15:26,584 Model training base path: "hmbench-letemps/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3"
95
+ 2023-10-14 00:15:26,584 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-14 00:15:26,584 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-14 00:15:26,584 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-14 00:16:57,553 epoch 1 - iter 180/1809 - loss 2.51944389 - time (sec): 90.97 - samples/sec: 413.65 - lr: 0.000016 - momentum: 0.000000
99
+ 2023-10-14 00:18:29,092 epoch 1 - iter 360/1809 - loss 2.26637505 - time (sec): 182.51 - samples/sec: 406.16 - lr: 0.000032 - momentum: 0.000000
100
+ 2023-10-14 00:20:01,815 epoch 1 - iter 540/1809 - loss 1.90816370 - time (sec): 275.23 - samples/sec: 409.55 - lr: 0.000048 - momentum: 0.000000
101
+ 2023-10-14 00:21:31,625 epoch 1 - iter 720/1809 - loss 1.56194036 - time (sec): 365.04 - samples/sec: 413.28 - lr: 0.000064 - momentum: 0.000000
102
+ 2023-10-14 00:23:01,817 epoch 1 - iter 900/1809 - loss 1.30444465 - time (sec): 455.23 - samples/sec: 414.24 - lr: 0.000080 - momentum: 0.000000
103
+ 2023-10-14 00:24:33,627 epoch 1 - iter 1080/1809 - loss 1.11961360 - time (sec): 547.04 - samples/sec: 413.43 - lr: 0.000095 - momentum: 0.000000
104
+ 2023-10-14 00:26:03,228 epoch 1 - iter 1260/1809 - loss 0.98696147 - time (sec): 636.64 - samples/sec: 412.31 - lr: 0.000111 - momentum: 0.000000
105
+ 2023-10-14 00:27:35,596 epoch 1 - iter 1440/1809 - loss 0.88259280 - time (sec): 729.01 - samples/sec: 411.99 - lr: 0.000127 - momentum: 0.000000
106
+ 2023-10-14 00:29:08,136 epoch 1 - iter 1620/1809 - loss 0.79457894 - time (sec): 821.55 - samples/sec: 413.43 - lr: 0.000143 - momentum: 0.000000
107
+ 2023-10-14 00:30:36,487 epoch 1 - iter 1800/1809 - loss 0.72720679 - time (sec): 909.90 - samples/sec: 415.88 - lr: 0.000159 - momentum: 0.000000
108
+ 2023-10-14 00:30:40,375 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-14 00:30:40,375 EPOCH 1 done: loss 0.7248 - lr: 0.000159
110
+ 2023-10-14 00:31:18,366 DEV : loss 0.13587747514247894 - f1-score (micro avg) 0.5512
111
+ 2023-10-14 00:31:18,430 saving best model
112
+ 2023-10-14 00:31:19,337 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-14 00:32:49,760 epoch 2 - iter 180/1809 - loss 0.10182046 - time (sec): 90.42 - samples/sec: 417.47 - lr: 0.000158 - momentum: 0.000000
114
+ 2023-10-14 00:34:19,933 epoch 2 - iter 360/1809 - loss 0.10191876 - time (sec): 180.59 - samples/sec: 413.42 - lr: 0.000156 - momentum: 0.000000
115
+ 2023-10-14 00:35:49,015 epoch 2 - iter 540/1809 - loss 0.09941428 - time (sec): 269.68 - samples/sec: 429.78 - lr: 0.000155 - momentum: 0.000000
116
+ 2023-10-14 00:37:16,788 epoch 2 - iter 720/1809 - loss 0.09701065 - time (sec): 357.45 - samples/sec: 432.34 - lr: 0.000153 - momentum: 0.000000
117
+ 2023-10-14 00:38:45,501 epoch 2 - iter 900/1809 - loss 0.09626668 - time (sec): 446.16 - samples/sec: 431.73 - lr: 0.000151 - momentum: 0.000000
118
+ 2023-10-14 00:40:13,863 epoch 2 - iter 1080/1809 - loss 0.09449862 - time (sec): 534.52 - samples/sec: 428.08 - lr: 0.000149 - momentum: 0.000000
119
+ 2023-10-14 00:41:41,920 epoch 2 - iter 1260/1809 - loss 0.09221887 - time (sec): 622.58 - samples/sec: 427.87 - lr: 0.000148 - momentum: 0.000000
120
+ 2023-10-14 00:43:10,709 epoch 2 - iter 1440/1809 - loss 0.09076303 - time (sec): 711.37 - samples/sec: 426.55 - lr: 0.000146 - momentum: 0.000000
121
+ 2023-10-14 00:44:41,103 epoch 2 - iter 1620/1809 - loss 0.08931886 - time (sec): 801.76 - samples/sec: 425.61 - lr: 0.000144 - momentum: 0.000000
122
+ 2023-10-14 00:46:10,438 epoch 2 - iter 1800/1809 - loss 0.08787461 - time (sec): 891.10 - samples/sec: 424.53 - lr: 0.000142 - momentum: 0.000000
123
+ 2023-10-14 00:46:14,383 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-14 00:46:14,383 EPOCH 2 done: loss 0.0879 - lr: 0.000142
125
+ 2023-10-14 00:46:52,480 DEV : loss 0.10200479626655579 - f1-score (micro avg) 0.6091
126
+ 2023-10-14 00:46:52,536 saving best model
127
+ 2023-10-14 00:46:55,099 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-14 00:48:24,885 epoch 3 - iter 180/1809 - loss 0.06314980 - time (sec): 89.78 - samples/sec: 416.42 - lr: 0.000140 - momentum: 0.000000
129
+ 2023-10-14 00:49:56,919 epoch 3 - iter 360/1809 - loss 0.05882268 - time (sec): 181.82 - samples/sec: 422.98 - lr: 0.000139 - momentum: 0.000000
130
+ 2023-10-14 00:51:27,745 epoch 3 - iter 540/1809 - loss 0.05798093 - time (sec): 272.64 - samples/sec: 420.56 - lr: 0.000137 - momentum: 0.000000
131
+ 2023-10-14 00:52:57,778 epoch 3 - iter 720/1809 - loss 0.05728226 - time (sec): 362.67 - samples/sec: 419.92 - lr: 0.000135 - momentum: 0.000000
132
+ 2023-10-14 00:54:27,582 epoch 3 - iter 900/1809 - loss 0.05773776 - time (sec): 452.48 - samples/sec: 423.51 - lr: 0.000133 - momentum: 0.000000
133
+ 2023-10-14 00:55:56,659 epoch 3 - iter 1080/1809 - loss 0.05718530 - time (sec): 541.56 - samples/sec: 424.38 - lr: 0.000132 - momentum: 0.000000
134
+ 2023-10-14 00:57:26,791 epoch 3 - iter 1260/1809 - loss 0.05787219 - time (sec): 631.69 - samples/sec: 420.89 - lr: 0.000130 - momentum: 0.000000
135
+ 2023-10-14 00:58:55,840 epoch 3 - iter 1440/1809 - loss 0.05720897 - time (sec): 720.74 - samples/sec: 420.23 - lr: 0.000128 - momentum: 0.000000
136
+ 2023-10-14 01:00:24,566 epoch 3 - iter 1620/1809 - loss 0.05641677 - time (sec): 809.46 - samples/sec: 419.21 - lr: 0.000126 - momentum: 0.000000
137
+ 2023-10-14 01:01:57,412 epoch 3 - iter 1800/1809 - loss 0.05640597 - time (sec): 902.31 - samples/sec: 418.82 - lr: 0.000125 - momentum: 0.000000
138
+ 2023-10-14 01:02:01,841 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-14 01:02:01,842 EPOCH 3 done: loss 0.0563 - lr: 0.000125
140
+ 2023-10-14 01:02:41,767 DEV : loss 0.13950783014297485 - f1-score (micro avg) 0.6198
141
+ 2023-10-14 01:02:41,831 saving best model
142
+ 2023-10-14 01:02:44,406 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-14 01:04:14,083 epoch 4 - iter 180/1809 - loss 0.04170364 - time (sec): 89.67 - samples/sec: 407.77 - lr: 0.000123 - momentum: 0.000000
144
+ 2023-10-14 01:05:46,581 epoch 4 - iter 360/1809 - loss 0.03858075 - time (sec): 182.17 - samples/sec: 411.86 - lr: 0.000121 - momentum: 0.000000
145
+ 2023-10-14 01:07:17,093 epoch 4 - iter 540/1809 - loss 0.03981754 - time (sec): 272.68 - samples/sec: 412.97 - lr: 0.000119 - momentum: 0.000000
146
+ 2023-10-14 01:08:47,525 epoch 4 - iter 720/1809 - loss 0.03847050 - time (sec): 363.11 - samples/sec: 411.25 - lr: 0.000117 - momentum: 0.000000
147
+ 2023-10-14 01:10:18,340 epoch 4 - iter 900/1809 - loss 0.03892181 - time (sec): 453.93 - samples/sec: 411.93 - lr: 0.000116 - momentum: 0.000000
148
+ 2023-10-14 01:11:50,274 epoch 4 - iter 1080/1809 - loss 0.04023787 - time (sec): 545.86 - samples/sec: 415.14 - lr: 0.000114 - momentum: 0.000000
149
+ 2023-10-14 01:13:22,824 epoch 4 - iter 1260/1809 - loss 0.04131278 - time (sec): 638.41 - samples/sec: 415.03 - lr: 0.000112 - momentum: 0.000000
150
+ 2023-10-14 01:14:51,158 epoch 4 - iter 1440/1809 - loss 0.04134067 - time (sec): 726.75 - samples/sec: 416.20 - lr: 0.000110 - momentum: 0.000000
151
+ 2023-10-14 01:16:20,419 epoch 4 - iter 1620/1809 - loss 0.04065918 - time (sec): 816.01 - samples/sec: 418.02 - lr: 0.000109 - momentum: 0.000000
152
+ 2023-10-14 01:17:50,525 epoch 4 - iter 1800/1809 - loss 0.04036183 - time (sec): 906.11 - samples/sec: 417.55 - lr: 0.000107 - momentum: 0.000000
153
+ 2023-10-14 01:17:54,624 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-14 01:17:54,624 EPOCH 4 done: loss 0.0403 - lr: 0.000107
155
+ 2023-10-14 01:18:35,366 DEV : loss 0.19245745241641998 - f1-score (micro avg) 0.6321
156
+ 2023-10-14 01:18:35,431 saving best model
157
+ 2023-10-14 01:18:38,017 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-14 01:20:09,743 epoch 5 - iter 180/1809 - loss 0.02576443 - time (sec): 91.72 - samples/sec: 418.78 - lr: 0.000105 - momentum: 0.000000
159
+ 2023-10-14 01:21:44,400 epoch 5 - iter 360/1809 - loss 0.02641732 - time (sec): 186.38 - samples/sec: 408.40 - lr: 0.000103 - momentum: 0.000000
160
+ 2023-10-14 01:23:15,834 epoch 5 - iter 540/1809 - loss 0.02651451 - time (sec): 277.81 - samples/sec: 407.70 - lr: 0.000101 - momentum: 0.000000
161
+ 2023-10-14 01:24:43,605 epoch 5 - iter 720/1809 - loss 0.02751988 - time (sec): 365.58 - samples/sec: 409.65 - lr: 0.000100 - momentum: 0.000000
162
+ 2023-10-14 01:26:12,328 epoch 5 - iter 900/1809 - loss 0.02767393 - time (sec): 454.31 - samples/sec: 414.53 - lr: 0.000098 - momentum: 0.000000
163
+ 2023-10-14 01:27:41,085 epoch 5 - iter 1080/1809 - loss 0.02796036 - time (sec): 543.06 - samples/sec: 416.11 - lr: 0.000096 - momentum: 0.000000
164
+ 2023-10-14 01:29:10,393 epoch 5 - iter 1260/1809 - loss 0.02785250 - time (sec): 632.37 - samples/sec: 414.60 - lr: 0.000094 - momentum: 0.000000
165
+ 2023-10-14 01:30:45,505 epoch 5 - iter 1440/1809 - loss 0.02843191 - time (sec): 727.48 - samples/sec: 412.25 - lr: 0.000093 - momentum: 0.000000
166
+ 2023-10-14 01:32:18,477 epoch 5 - iter 1620/1809 - loss 0.02924215 - time (sec): 820.46 - samples/sec: 412.99 - lr: 0.000091 - momentum: 0.000000
167
+ 2023-10-14 01:33:51,099 epoch 5 - iter 1800/1809 - loss 0.02925813 - time (sec): 913.08 - samples/sec: 414.13 - lr: 0.000089 - momentum: 0.000000
168
+ 2023-10-14 01:33:55,154 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-14 01:33:55,155 EPOCH 5 done: loss 0.0293 - lr: 0.000089
170
+ 2023-10-14 01:34:33,142 DEV : loss 0.2196071296930313 - f1-score (micro avg) 0.6454
171
+ 2023-10-14 01:34:33,199 saving best model
172
+ 2023-10-14 01:34:35,762 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-14 01:36:06,097 epoch 6 - iter 180/1809 - loss 0.01696868 - time (sec): 90.33 - samples/sec: 434.02 - lr: 0.000087 - momentum: 0.000000
174
+ 2023-10-14 01:37:36,160 epoch 6 - iter 360/1809 - loss 0.01685237 - time (sec): 180.39 - samples/sec: 426.33 - lr: 0.000085 - momentum: 0.000000
175
+ 2023-10-14 01:39:05,475 epoch 6 - iter 540/1809 - loss 0.02067304 - time (sec): 269.71 - samples/sec: 422.42 - lr: 0.000084 - momentum: 0.000000
176
+ 2023-10-14 01:40:35,631 epoch 6 - iter 720/1809 - loss 0.02166279 - time (sec): 359.87 - samples/sec: 418.94 - lr: 0.000082 - momentum: 0.000000
177
+ 2023-10-14 01:42:08,370 epoch 6 - iter 900/1809 - loss 0.02213544 - time (sec): 452.60 - samples/sec: 412.98 - lr: 0.000080 - momentum: 0.000000
178
+ 2023-10-14 01:43:40,493 epoch 6 - iter 1080/1809 - loss 0.02224410 - time (sec): 544.73 - samples/sec: 412.51 - lr: 0.000078 - momentum: 0.000000
179
+ 2023-10-14 01:45:13,554 epoch 6 - iter 1260/1809 - loss 0.02163785 - time (sec): 637.79 - samples/sec: 414.18 - lr: 0.000077 - momentum: 0.000000
180
+ 2023-10-14 01:46:46,037 epoch 6 - iter 1440/1809 - loss 0.02152121 - time (sec): 730.27 - samples/sec: 415.13 - lr: 0.000075 - momentum: 0.000000
181
+ 2023-10-14 01:48:18,422 epoch 6 - iter 1620/1809 - loss 0.02245358 - time (sec): 822.66 - samples/sec: 412.61 - lr: 0.000073 - momentum: 0.000000
182
+ 2023-10-14 01:49:50,889 epoch 6 - iter 1800/1809 - loss 0.02230601 - time (sec): 915.12 - samples/sec: 413.40 - lr: 0.000071 - momentum: 0.000000
183
+ 2023-10-14 01:49:54,955 ----------------------------------------------------------------------------------------------------
184
+ 2023-10-14 01:49:54,956 EPOCH 6 done: loss 0.0222 - lr: 0.000071
185
+ 2023-10-14 01:50:36,500 DEV : loss 0.25768402218818665 - f1-score (micro avg) 0.6629
186
+ 2023-10-14 01:50:36,559 saving best model
187
+ 2023-10-14 01:50:39,137 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-14 01:52:12,077 epoch 7 - iter 180/1809 - loss 0.00921054 - time (sec): 92.94 - samples/sec: 411.47 - lr: 0.000069 - momentum: 0.000000
189
+ 2023-10-14 01:53:50,733 epoch 7 - iter 360/1809 - loss 0.01168701 - time (sec): 191.59 - samples/sec: 395.89 - lr: 0.000068 - momentum: 0.000000
190
+ 2023-10-14 01:55:21,611 epoch 7 - iter 540/1809 - loss 0.01305153 - time (sec): 282.47 - samples/sec: 400.63 - lr: 0.000066 - momentum: 0.000000
191
+ 2023-10-14 01:56:55,925 epoch 7 - iter 720/1809 - loss 0.01260349 - time (sec): 376.78 - samples/sec: 404.80 - lr: 0.000064 - momentum: 0.000000
192
+ 2023-10-14 01:58:26,714 epoch 7 - iter 900/1809 - loss 0.01297523 - time (sec): 467.57 - samples/sec: 406.34 - lr: 0.000062 - momentum: 0.000000
193
+ 2023-10-14 02:00:02,490 epoch 7 - iter 1080/1809 - loss 0.01344991 - time (sec): 563.35 - samples/sec: 404.29 - lr: 0.000061 - momentum: 0.000000
194
+ 2023-10-14 02:01:34,331 epoch 7 - iter 1260/1809 - loss 0.01433938 - time (sec): 655.19 - samples/sec: 406.81 - lr: 0.000059 - momentum: 0.000000
195
+ 2023-10-14 02:03:06,795 epoch 7 - iter 1440/1809 - loss 0.01477713 - time (sec): 747.65 - samples/sec: 406.47 - lr: 0.000057 - momentum: 0.000000
196
+ 2023-10-14 02:04:40,603 epoch 7 - iter 1620/1809 - loss 0.01477570 - time (sec): 841.46 - samples/sec: 406.11 - lr: 0.000055 - momentum: 0.000000
197
+ 2023-10-14 02:06:15,178 epoch 7 - iter 1800/1809 - loss 0.01511790 - time (sec): 936.04 - samples/sec: 404.12 - lr: 0.000053 - momentum: 0.000000
198
+ 2023-10-14 02:06:19,433 ----------------------------------------------------------------------------------------------------
199
+ 2023-10-14 02:06:19,433 EPOCH 7 done: loss 0.0151 - lr: 0.000053
200
+ 2023-10-14 02:06:59,410 DEV : loss 0.2951534390449524 - f1-score (micro avg) 0.6663
201
+ 2023-10-14 02:06:59,467 saving best model
202
+ 2023-10-14 02:07:02,034 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-14 02:08:31,365 epoch 8 - iter 180/1809 - loss 0.00690787 - time (sec): 89.33 - samples/sec: 414.43 - lr: 0.000052 - momentum: 0.000000
204
+ 2023-10-14 02:10:01,793 epoch 8 - iter 360/1809 - loss 0.01015270 - time (sec): 179.75 - samples/sec: 417.93 - lr: 0.000050 - momentum: 0.000000
205
+ 2023-10-14 02:11:31,245 epoch 8 - iter 540/1809 - loss 0.01075555 - time (sec): 269.21 - samples/sec: 425.42 - lr: 0.000048 - momentum: 0.000000
206
+ 2023-10-14 02:12:59,192 epoch 8 - iter 720/1809 - loss 0.01026295 - time (sec): 357.15 - samples/sec: 426.20 - lr: 0.000046 - momentum: 0.000000
207
+ 2023-10-14 02:14:29,940 epoch 8 - iter 900/1809 - loss 0.00994762 - time (sec): 447.90 - samples/sec: 425.93 - lr: 0.000044 - momentum: 0.000000
208
+ 2023-10-14 02:16:04,150 epoch 8 - iter 1080/1809 - loss 0.00979174 - time (sec): 542.11 - samples/sec: 420.45 - lr: 0.000043 - momentum: 0.000000
209
+ 2023-10-14 02:17:34,169 epoch 8 - iter 1260/1809 - loss 0.01043605 - time (sec): 632.13 - samples/sec: 420.83 - lr: 0.000041 - momentum: 0.000000
210
+ 2023-10-14 02:19:02,279 epoch 8 - iter 1440/1809 - loss 0.01048305 - time (sec): 720.24 - samples/sec: 420.80 - lr: 0.000039 - momentum: 0.000000
211
+ 2023-10-14 02:20:31,156 epoch 8 - iter 1620/1809 - loss 0.01031134 - time (sec): 809.12 - samples/sec: 421.20 - lr: 0.000037 - momentum: 0.000000
212
+ 2023-10-14 02:21:59,294 epoch 8 - iter 1800/1809 - loss 0.01018123 - time (sec): 897.26 - samples/sec: 421.63 - lr: 0.000036 - momentum: 0.000000
213
+ 2023-10-14 02:22:03,202 ----------------------------------------------------------------------------------------------------
214
+ 2023-10-14 02:22:03,202 EPOCH 8 done: loss 0.0102 - lr: 0.000036
215
+ 2023-10-14 02:22:42,340 DEV : loss 0.31866809725761414 - f1-score (micro avg) 0.6638
216
+ 2023-10-14 02:22:42,399 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-14 02:24:15,293 epoch 9 - iter 180/1809 - loss 0.00817136 - time (sec): 92.89 - samples/sec: 415.41 - lr: 0.000034 - momentum: 0.000000
218
+ 2023-10-14 02:25:48,900 epoch 9 - iter 360/1809 - loss 0.00980844 - time (sec): 186.50 - samples/sec: 417.93 - lr: 0.000032 - momentum: 0.000000
219
+ 2023-10-14 02:27:21,286 epoch 9 - iter 540/1809 - loss 0.00968269 - time (sec): 278.88 - samples/sec: 413.52 - lr: 0.000030 - momentum: 0.000000
220
+ 2023-10-14 02:28:53,866 epoch 9 - iter 720/1809 - loss 0.00851234 - time (sec): 371.46 - samples/sec: 413.77 - lr: 0.000028 - momentum: 0.000000
221
+ 2023-10-14 02:30:25,601 epoch 9 - iter 900/1809 - loss 0.00876138 - time (sec): 463.20 - samples/sec: 412.00 - lr: 0.000027 - momentum: 0.000000
222
+ 2023-10-14 02:31:58,342 epoch 9 - iter 1080/1809 - loss 0.00835601 - time (sec): 555.94 - samples/sec: 411.87 - lr: 0.000025 - momentum: 0.000000
223
+ 2023-10-14 02:33:29,812 epoch 9 - iter 1260/1809 - loss 0.00813399 - time (sec): 647.41 - samples/sec: 411.93 - lr: 0.000023 - momentum: 0.000000
224
+ 2023-10-14 02:35:00,772 epoch 9 - iter 1440/1809 - loss 0.00791238 - time (sec): 738.37 - samples/sec: 411.24 - lr: 0.000021 - momentum: 0.000000
225
+ 2023-10-14 02:36:32,257 epoch 9 - iter 1620/1809 - loss 0.00783074 - time (sec): 829.86 - samples/sec: 409.66 - lr: 0.000020 - momentum: 0.000000
226
+ 2023-10-14 02:38:03,837 epoch 9 - iter 1800/1809 - loss 0.00756760 - time (sec): 921.44 - samples/sec: 410.66 - lr: 0.000018 - momentum: 0.000000
227
+ 2023-10-14 02:38:07,713 ----------------------------------------------------------------------------------------------------
228
+ 2023-10-14 02:38:07,713 EPOCH 9 done: loss 0.0076 - lr: 0.000018
229
+ 2023-10-14 02:38:45,463 DEV : loss 0.3397609293460846 - f1-score (micro avg) 0.6699
230
+ 2023-10-14 02:38:45,524 saving best model
231
+ 2023-10-14 02:38:48,090 ----------------------------------------------------------------------------------------------------
232
+ 2023-10-14 02:40:17,848 epoch 10 - iter 180/1809 - loss 0.00675340 - time (sec): 89.75 - samples/sec: 418.97 - lr: 0.000016 - momentum: 0.000000
233
+ 2023-10-14 02:41:47,430 epoch 10 - iter 360/1809 - loss 0.00699592 - time (sec): 179.34 - samples/sec: 410.52 - lr: 0.000014 - momentum: 0.000000
234
+ 2023-10-14 02:43:19,155 epoch 10 - iter 540/1809 - loss 0.00674174 - time (sec): 271.06 - samples/sec: 414.12 - lr: 0.000012 - momentum: 0.000000
235
+ 2023-10-14 02:44:53,881 epoch 10 - iter 720/1809 - loss 0.00580321 - time (sec): 365.79 - samples/sec: 407.45 - lr: 0.000011 - momentum: 0.000000
236
+ 2023-10-14 02:46:33,110 epoch 10 - iter 900/1809 - loss 0.00563464 - time (sec): 465.02 - samples/sec: 403.22 - lr: 0.000009 - momentum: 0.000000
237
+ 2023-10-14 02:48:05,921 epoch 10 - iter 1080/1809 - loss 0.00536878 - time (sec): 557.83 - samples/sec: 403.01 - lr: 0.000007 - momentum: 0.000000
238
+ 2023-10-14 02:49:38,498 epoch 10 - iter 1260/1809 - loss 0.00519036 - time (sec): 650.40 - samples/sec: 406.07 - lr: 0.000005 - momentum: 0.000000
239
+ 2023-10-14 02:51:10,923 epoch 10 - iter 1440/1809 - loss 0.00506535 - time (sec): 742.83 - samples/sec: 405.69 - lr: 0.000004 - momentum: 0.000000
240
+ 2023-10-14 02:52:43,275 epoch 10 - iter 1620/1809 - loss 0.00520096 - time (sec): 835.18 - samples/sec: 407.55 - lr: 0.000002 - momentum: 0.000000
241
+ 2023-10-14 02:54:18,302 epoch 10 - iter 1800/1809 - loss 0.00505952 - time (sec): 930.21 - samples/sec: 406.29 - lr: 0.000000 - momentum: 0.000000
242
+ 2023-10-14 02:54:22,857 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-14 02:54:22,857 EPOCH 10 done: loss 0.0050 - lr: 0.000000
244
+ 2023-10-14 02:55:02,618 DEV : loss 0.34813550114631653 - f1-score (micro avg) 0.6654
245
+ 2023-10-14 02:55:03,588 ----------------------------------------------------------------------------------------------------
246
+ 2023-10-14 02:55:03,590 Loading model from best epoch ...
247
+ 2023-10-14 02:55:07,452 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
248
+ 2023-10-14 02:56:07,325
249
+ Results:
250
+ - F-score (micro) 0.6407
251
+ - F-score (macro) 0.4712
252
+ - Accuracy 0.4813
253
+
254
+ By class:
255
+ precision recall f1-score support
256
+
257
+ loc 0.6545 0.7563 0.7017 591
258
+ pers 0.5705 0.7143 0.6343 357
259
+ org 0.1000 0.0633 0.0775 79
260
+
261
+ micro avg 0.5992 0.6884 0.6407 1027
262
+ macro avg 0.4416 0.5113 0.4712 1027
263
+ weighted avg 0.5826 0.6884 0.6303 1027
264
+
265
+ 2023-10-14 02:56:07,325 ----------------------------------------------------------------------------------------------------