Hugging Face's logo

stefan-it
/
flair-modernbert-large-ner-conll03

PyTorch
TensorBoard
English
ner
Model card Files
xet
Metrics Community
stefan-it commited on
Commit
85caa69
·
verified ·
1 Parent(s): c49ad16

feat: add training log

Browse files
Files changed (1) hide show
  1. training.log +251 -0
training.log ADDED
@@ -0,0 +1,251 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
2
+ 2025-05-08 16:35:44,100 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): ModernBertModel(
5
+ (embeddings): ModernBertEmbeddings(
6
+ (tok_embeddings): Embedding(50369, 1024, padding_idx=50283)
7
+ (norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
8
+ (drop): Dropout(p=0.0, inplace=False)
9
+ )
10
+ (layers): ModuleList(
11
+ (0): ModernBertEncoderLayer(
12
+ (attn_norm): Identity()
13
+ (attn): ModernBertAttention(
14
+ (Wqkv): Linear(in_features=1024, out_features=3072, bias=False)
15
+ (rotary_emb): ModernBertRotaryEmbedding()
16
+ (Wo): Linear(in_features=1024, out_features=1024, bias=False)
17
+ (out_drop): Identity()
18
+ )
19
+ (mlp_norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
20
+ (mlp): ModernBertMLP(
21
+ (Wi): Linear(in_features=1024, out_features=5248, bias=False)
22
+ (act): GELUActivation()
23
+ (drop): Dropout(p=0.0, inplace=False)
24
+ (Wo): Linear(in_features=2624, out_features=1024, bias=False)
25
+ )
26
+ )
27
+ (1-27): 27 x ModernBertEncoderLayer(
28
+ (attn_norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
29
+ (attn): ModernBertAttention(
30
+ (Wqkv): Linear(in_features=1024, out_features=3072, bias=False)
31
+ (rotary_emb): ModernBertRotaryEmbedding()
32
+ (Wo): Linear(in_features=1024, out_features=1024, bias=False)
33
+ (out_drop): Identity()
34
+ )
35
+ (mlp_norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
36
+ (mlp): ModernBertMLP(
37
+ (Wi): Linear(in_features=1024, out_features=5248, bias=False)
38
+ (act): GELUActivation()
39
+ (drop): Dropout(p=0.0, inplace=False)
40
+ (Wo): Linear(in_features=2624, out_features=1024, bias=False)
41
+ )
42
+ )
43
+ )
44
+ (final_norm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
45
+ )
46
+ )
47
+ (locked_dropout): LockedDropout(p=0.5)
48
+ (linear): Linear(in_features=2048, out_features=17, bias=True)
49
+ (loss_function): CrossEntropyLoss()
50
+ )"
51
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
52
+ 2025-05-08 16:35:44,100 MultiCorpus: 14987 train + 3466 dev + 3684 test sentences
53
+ - CONLL_03_ENGLISH Corpus: 14987 train + 3466 dev + 3684 test sentences - /home/stefan/.flair/datasets/conll_03_english
54
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
55
+ 2025-05-08 16:35:44,100 Train: 14987 sentences
56
+ 2025-05-08 16:35:44,100 (train_with_dev=False, train_with_test=False)
57
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
58
+ 2025-05-08 16:35:44,100 Training Params:
59
+ 2025-05-08 16:35:44,100 - optimizer: "<class 'torch.optim.adamw.AdamW'>"
60
+ 2025-05-08 16:35:44,100 - learning_rate: "2e-05"
61
+ 2025-05-08 16:35:44,100 - mini_batch_size: "16"
62
+ 2025-05-08 16:35:44,100 - max_epochs: "10"
63
+ 2025-05-08 16:35:44,100 - shuffle: "True"
64
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
65
+ 2025-05-08 16:35:44,100 Plugins:
66
+ 2025-05-08 16:35:44,100 - TensorboardLogger
67
+ 2025-05-08 16:35:44,100 - LinearScheduler | warmup_fraction: '0.1'
68
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
69
+ 2025-05-08 16:35:44,100 Final evaluation on model from best epoch (best-model.pt)
70
+ 2025-05-08 16:35:44,100 - metric: "('micro avg', 'f1-score')"
71
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
72
+ 2025-05-08 16:35:44,100 Computation:
73
+ 2025-05-08 16:35:44,100 - compute on device: cuda:0
74
+ 2025-05-08 16:35:44,100 - embedding storage: none
75
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
76
+ 2025-05-08 16:35:44,100 Model training base path: "flair-ner-conll03_english-modern_bert_large_tokenizer_fix-bs16-e10-cs0-lr2e-05-2"
77
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
78
+ 2025-05-08 16:35:44,100 ----------------------------------------------------------------------------------------------------
79
+ 2025-05-08 16:35:44,100 Logging anything other than scalars to TensorBoard is currently not supported.
80
+ 2025-05-08 16:35:56,453 epoch 1 - iter 93/937 - loss 51.00117641 - time (sec): 12.35 - samples/sec: 1732.91 - lr: 0.000002 - momentum: 0.000000
81
+ 2025-05-08 16:36:08,727 epoch 1 - iter 186/937 - loss 39.31915124 - time (sec): 24.63 - samples/sec: 1689.29 - lr: 0.000004 - momentum: 0.000000
82
+ 2025-05-08 16:36:21,012 epoch 1 - iter 279/937 - loss 28.34790128 - time (sec): 36.91 - samples/sec: 1696.57 - lr: 0.000006 - momentum: 0.000000
83
+ 2025-05-08 16:36:33,294 epoch 1 - iter 372/937 - loss 21.50534985 - time (sec): 49.19 - samples/sec: 1692.10 - lr: 0.000008 - momentum: 0.000000
84
+ 2025-05-08 16:36:45,533 epoch 1 - iter 465/937 - loss 17.43941279 - time (sec): 61.43 - samples/sec: 1676.80 - lr: 0.000010 - momentum: 0.000000
85
+ 2025-05-08 16:36:57,843 epoch 1 - iter 558/937 - loss 14.61276887 - time (sec): 73.74 - samples/sec: 1671.44 - lr: 0.000012 - momentum: 0.000000
86
+ 2025-05-08 16:37:10,132 epoch 1 - iter 651/937 - loss 12.61916178 - time (sec): 86.03 - samples/sec: 1664.38 - lr: 0.000014 - momentum: 0.000000
87
+ 2025-05-08 16:37:22,323 epoch 1 - iter 744/937 - loss 11.09955475 - time (sec): 98.22 - samples/sec: 1660.09 - lr: 0.000016 - momentum: 0.000000
88
+ 2025-05-08 16:37:34,541 epoch 1 - iter 837/937 - loss 9.91430791 - time (sec): 110.44 - samples/sec: 1656.84 - lr: 0.000018 - momentum: 0.000000
89
+ 2025-05-08 16:37:46,710 epoch 1 - iter 930/937 - loss 8.93883090 - time (sec): 122.61 - samples/sec: 1657.42 - lr: 0.000020 - momentum: 0.000000
90
+ 2025-05-08 16:37:47,602 ----------------------------------------------------------------------------------------------------
91
+ 2025-05-08 16:37:47,602 EPOCH 1 done: loss 8.8805 - lr: 0.000020
92
+ 2025-05-08 16:37:54,144 DEV : loss 0.1118580624461174 - f1-score (micro avg) 0.9036
93
+ 2025-05-08 16:37:54,170 saving best model
94
+ 2025-05-08 16:37:54,784 ----------------------------------------------------------------------------------------------------
95
+ 2025-05-08 16:38:06,997 epoch 2 - iter 93/937 - loss 0.11137172 - time (sec): 12.21 - samples/sec: 1652.27 - lr: 0.000020 - momentum: 0.000000
96
+ 2025-05-08 16:38:19,287 epoch 2 - iter 186/937 - loss 0.11052976 - time (sec): 24.50 - samples/sec: 1627.98 - lr: 0.000020 - momentum: 0.000000
97
+ 2025-05-08 16:38:31,513 epoch 2 - iter 279/937 - loss 0.10145703 - time (sec): 36.73 - samples/sec: 1641.78 - lr: 0.000019 - momentum: 0.000000
98
+ 2025-05-08 16:38:43,868 epoch 2 - iter 372/937 - loss 0.09382547 - time (sec): 49.08 - samples/sec: 1653.09 - lr: 0.000019 - momentum: 0.000000
99
+ 2025-05-08 16:38:56,121 epoch 2 - iter 465/937 - loss 0.09283270 - time (sec): 61.34 - samples/sec: 1656.97 - lr: 0.000019 - momentum: 0.000000
100
+ 2025-05-08 16:39:08,413 epoch 2 - iter 558/937 - loss 0.09469460 - time (sec): 73.63 - samples/sec: 1657.89 - lr: 0.000019 - momentum: 0.000000
101
+ 2025-05-08 16:39:20,753 epoch 2 - iter 651/937 - loss 0.09219174 - time (sec): 85.97 - samples/sec: 1655.83 - lr: 0.000018 - momentum: 0.000000
102
+ 2025-05-08 16:39:33,056 epoch 2 - iter 744/937 - loss 0.09000350 - time (sec): 98.27 - samples/sec: 1655.73 - lr: 0.000018 - momentum: 0.000000
103
+ 2025-05-08 16:39:45,277 epoch 2 - iter 837/937 - loss 0.09094705 - time (sec): 110.49 - samples/sec: 1651.81 - lr: 0.000018 - momentum: 0.000000
104
+ 2025-05-08 16:39:57,613 epoch 2 - iter 930/937 - loss 0.08852163 - time (sec): 122.83 - samples/sec: 1653.86 - lr: 0.000018 - momentum: 0.000000
105
+ 2025-05-08 16:39:58,500 ----------------------------------------------------------------------------------------------------
106
+ 2025-05-08 16:39:58,501 EPOCH 2 done: loss 0.0885 - lr: 0.000018
107
+ 2025-05-08 16:40:03,946 DEV : loss 0.06839137524366379 - f1-score (micro avg) 0.9372
108
+ 2025-05-08 16:40:03,971 saving best model
109
+ 2025-05-08 16:40:04,951 ----------------------------------------------------------------------------------------------------
110
+ 2025-05-08 16:40:17,146 epoch 3 - iter 93/937 - loss 0.04928881 - time (sec): 12.19 - samples/sec: 1647.64 - lr: 0.000018 - momentum: 0.000000
111
+ 2025-05-08 16:40:29,461 epoch 3 - iter 186/937 - loss 0.04589421 - time (sec): 24.51 - samples/sec: 1663.15 - lr: 0.000017 - momentum: 0.000000
112
+ 2025-05-08 16:40:41,705 epoch 3 - iter 279/937 - loss 0.04343733 - time (sec): 36.75 - samples/sec: 1646.12 - lr: 0.000017 - momentum: 0.000000
113
+ 2025-05-08 16:40:53,964 epoch 3 - iter 372/937 - loss 0.04578370 - time (sec): 49.01 - samples/sec: 1650.85 - lr: 0.000017 - momentum: 0.000000
114
+ 2025-05-08 16:41:06,206 epoch 3 - iter 465/937 - loss 0.04201880 - time (sec): 61.25 - samples/sec: 1646.44 - lr: 0.000017 - momentum: 0.000000
115
+ 2025-05-08 16:41:18,565 epoch 3 - iter 558/937 - loss 0.04253207 - time (sec): 73.61 - samples/sec: 1648.62 - lr: 0.000016 - momentum: 0.000000
116
+ 2025-05-08 16:41:30,845 epoch 3 - iter 651/937 - loss 0.04646610 - time (sec): 85.89 - samples/sec: 1646.48 - lr: 0.000016 - momentum: 0.000000
117
+ 2025-05-08 16:41:43,169 epoch 3 - iter 744/937 - loss 0.04955133 - time (sec): 98.22 - samples/sec: 1653.16 - lr: 0.000016 - momentum: 0.000000
118
+ 2025-05-08 16:41:55,483 epoch 3 - iter 837/937 - loss 0.04703390 - time (sec): 110.53 - samples/sec: 1656.33 - lr: 0.000016 - momentum: 0.000000
119
+ 2025-05-08 16:42:07,683 epoch 3 - iter 930/937 - loss 0.04786307 - time (sec): 122.73 - samples/sec: 1654.62 - lr: 0.000016 - momentum: 0.000000
120
+ 2025-05-08 16:42:08,556 ----------------------------------------------------------------------------------------------------
121
+ 2025-05-08 16:42:08,556 EPOCH 3 done: loss 0.0477 - lr: 0.000016
122
+ 2025-05-08 16:42:14,012 DEV : loss 0.07307213544845581 - f1-score (micro avg) 0.9533
123
+ 2025-05-08 16:42:14,038 saving best model
124
+ 2025-05-08 16:42:15,034 ----------------------------------------------------------------------------------------------------
125
+ 2025-05-08 16:42:28,642 epoch 4 - iter 93/937 - loss 0.01898962 - time (sec): 13.61 - samples/sec: 1568.87 - lr: 0.000015 - momentum: 0.000000
126
+ 2025-05-08 16:42:40,917 epoch 4 - iter 186/937 - loss 0.02010164 - time (sec): 25.88 - samples/sec: 1612.31 - lr: 0.000015 - momentum: 0.000000
127
+ 2025-05-08 16:42:53,288 epoch 4 - iter 279/937 - loss 0.02142412 - time (sec): 38.25 - samples/sec: 1606.07 - lr: 0.000015 - momentum: 0.000000
128
+ 2025-05-08 16:43:05,506 epoch 4 - iter 372/937 - loss 0.02282148 - time (sec): 50.47 - samples/sec: 1613.79 - lr: 0.000015 - momentum: 0.000000
129
+ 2025-05-08 16:43:17,846 epoch 4 - iter 465/937 - loss 0.02263261 - time (sec): 62.81 - samples/sec: 1632.46 - lr: 0.000014 - momentum: 0.000000
130
+ 2025-05-08 16:43:30,220 epoch 4 - iter 558/937 - loss 0.02330124 - time (sec): 75.18 - samples/sec: 1637.30 - lr: 0.000014 - momentum: 0.000000
131
+ 2025-05-08 16:43:42,488 epoch 4 - iter 651/937 - loss 0.02443078 - time (sec): 87.45 - samples/sec: 1632.91 - lr: 0.000014 - momentum: 0.000000
132
+ 2025-05-08 16:43:54,794 epoch 4 - iter 744/937 - loss 0.02445322 - time (sec): 99.76 - samples/sec: 1637.75 - lr: 0.000014 - momentum: 0.000000
133
+ 2025-05-08 16:44:06,909 epoch 4 - iter 837/937 - loss 0.02507178 - time (sec): 111.87 - samples/sec: 1633.75 - lr: 0.000014 - momentum: 0.000000
134
+ 2025-05-08 16:44:19,212 epoch 4 - iter 930/937 - loss 0.02554490 - time (sec): 124.18 - samples/sec: 1634.63 - lr: 0.000013 - momentum: 0.000000
135
+ 2025-05-08 16:44:20,092 ----------------------------------------------------------------------------------------------------
136
+ 2025-05-08 16:44:20,092 EPOCH 4 done: loss 0.0255 - lr: 0.000013
137
+ 2025-05-08 16:44:25,557 DEV : loss 0.09235326200723648 - f1-score (micro avg) 0.9594
138
+ 2025-05-08 16:44:25,582 saving best model
139
+ 2025-05-08 16:44:26,579 ----------------------------------------------------------------------------------------------------
140
+ 2025-05-08 16:44:38,913 epoch 5 - iter 93/937 - loss 0.01416651 - time (sec): 12.33 - samples/sec: 1603.37 - lr: 0.000013 - momentum: 0.000000
141
+ 2025-05-08 16:44:51,263 epoch 5 - iter 186/937 - loss 0.01363371 - time (sec): 24.68 - samples/sec: 1650.67 - lr: 0.000013 - momentum: 0.000000
142
+ 2025-05-08 16:45:03,521 epoch 5 - iter 279/937 - loss 0.01315713 - time (sec): 36.94 - samples/sec: 1650.89 - lr: 0.000013 - momentum: 0.000000
143
+ 2025-05-08 16:45:15,869 epoch 5 - iter 372/937 - loss 0.01589097 - time (sec): 49.29 - samples/sec: 1657.14 - lr: 0.000012 - momentum: 0.000000
144
+ 2025-05-08 16:45:28,072 epoch 5 - iter 465/937 - loss 0.01531867 - time (sec): 61.49 - samples/sec: 1651.82 - lr: 0.000012 - momentum: 0.000000
145
+ 2025-05-08 16:45:40,485 epoch 5 - iter 558/937 - loss 0.01951086 - time (sec): 73.91 - samples/sec: 1646.68 - lr: 0.000012 - momentum: 0.000000
146
+ 2025-05-08 16:45:52,827 epoch 5 - iter 651/937 - loss 0.01969603 - time (sec): 86.25 - samples/sec: 1652.73 - lr: 0.000012 - momentum: 0.000000
147
+ 2025-05-08 16:46:05,052 epoch 5 - iter 744/937 - loss 0.01837435 - time (sec): 98.47 - samples/sec: 1650.35 - lr: 0.000012 - momentum: 0.000000
148
+ 2025-05-08 16:46:17,313 epoch 5 - iter 837/937 - loss 0.01937513 - time (sec): 110.73 - samples/sec: 1652.26 - lr: 0.000011 - momentum: 0.000000
149
+ 2025-05-08 16:46:29,551 epoch 5 - iter 930/937 - loss 0.01967155 - time (sec): 122.97 - samples/sec: 1651.51 - lr: 0.000011 - momentum: 0.000000
150
+ 2025-05-08 16:46:30,427 ----------------------------------------------------------------------------------------------------
151
+ 2025-05-08 16:46:30,427 EPOCH 5 done: loss 0.0196 - lr: 0.000011
152
+ 2025-05-08 16:46:35,889 DEV : loss 0.08506825566291809 - f1-score (micro avg) 0.9651
153
+ 2025-05-08 16:46:35,914 saving best model
154
+ 2025-05-08 16:46:36,899 ----------------------------------------------------------------------------------------------------
155
+ 2025-05-08 16:46:49,267 epoch 6 - iter 93/937 - loss 0.00097797 - time (sec): 12.37 - samples/sec: 1688.57 - lr: 0.000011 - momentum: 0.000000
156
+ 2025-05-08 16:47:01,644 epoch 6 - iter 186/937 - loss 0.00406677 - time (sec): 24.74 - samples/sec: 1655.65 - lr: 0.000011 - momentum: 0.000000
157
+ 2025-05-08 16:47:15,115 epoch 6 - iter 279/937 - loss 0.00554206 - time (sec): 38.22 - samples/sec: 1619.79 - lr: 0.000010 - momentum: 0.000000
158
+ 2025-05-08 16:47:27,338 epoch 6 - iter 372/937 - loss 0.00611831 - time (sec): 50.44 - samples/sec: 1622.06 - lr: 0.000010 - momentum: 0.000000
159
+ 2025-05-08 16:47:39,647 epoch 6 - iter 465/937 - loss 0.00675677 - time (sec): 62.75 - samples/sec: 1629.17 - lr: 0.000010 - momentum: 0.000000
160
+ 2025-05-08 16:47:51,891 epoch 6 - iter 558/937 - loss 0.00685979 - time (sec): 74.99 - samples/sec: 1632.98 - lr: 0.000010 - momentum: 0.000000
161
+ 2025-05-08 16:48:04,124 epoch 6 - iter 651/937 - loss 0.00811317 - time (sec): 87.22 - samples/sec: 1624.14 - lr: 0.000010 - momentum: 0.000000
162
+ 2025-05-08 16:48:16,545 epoch 6 - iter 744/937 - loss 0.00856449 - time (sec): 99.64 - samples/sec: 1631.43 - lr: 0.000009 - momentum: 0.000000
163
+ 2025-05-08 16:48:28,801 epoch 6 - iter 837/937 - loss 0.00890904 - time (sec): 111.90 - samples/sec: 1637.56 - lr: 0.000009 - momentum: 0.000000
164
+ 2025-05-08 16:48:41,002 epoch 6 - iter 930/937 - loss 0.00869569 - time (sec): 124.10 - samples/sec: 1637.91 - lr: 0.000009 - momentum: 0.000000
165
+ 2025-05-08 16:48:41,887 ----------------------------------------------------------------------------------------------------
166
+ 2025-05-08 16:48:41,887 EPOCH 6 done: loss 0.0087 - lr: 0.000009
167
+ 2025-05-08 16:48:47,351 DEV : loss 0.1142197698354721 - f1-score (micro avg) 0.9642
168
+ 2025-05-08 16:48:47,377 ----------------------------------------------------------------------------------------------------
169
+ 2025-05-08 16:48:59,721 epoch 7 - iter 93/937 - loss 0.00487326 - time (sec): 12.34 - samples/sec: 1630.63 - lr: 0.000009 - momentum: 0.000000
170
+ 2025-05-08 16:49:11,966 epoch 7 - iter 186/937 - loss 0.00628660 - time (sec): 24.59 - samples/sec: 1654.42 - lr: 0.000008 - momentum: 0.000000
171
+ 2025-05-08 16:49:24,089 epoch 7 - iter 279/937 - loss 0.00580646 - time (sec): 36.71 - samples/sec: 1636.76 - lr: 0.000008 - momentum: 0.000000
172
+ 2025-05-08 16:49:36,346 epoch 7 - iter 372/937 - loss 0.00641392 - time (sec): 48.97 - samples/sec: 1647.65 - lr: 0.000008 - momentum: 0.000000
173
+ 2025-05-08 16:49:48,730 epoch 7 - iter 465/937 - loss 0.00541592 - time (sec): 61.35 - samples/sec: 1649.42 - lr: 0.000008 - momentum: 0.000000
174
+ 2025-05-08 16:50:01,050 epoch 7 - iter 558/937 - loss 0.00500581 - time (sec): 73.67 - samples/sec: 1656.15 - lr: 0.000008 - momentum: 0.000000
175
+ 2025-05-08 16:50:13,357 epoch 7 - iter 651/937 - loss 0.00504434 - time (sec): 85.98 - samples/sec: 1645.32 - lr: 0.000007 - momentum: 0.000000
176
+ 2025-05-08 16:50:25,648 epoch 7 - iter 744/937 - loss 0.00508162 - time (sec): 98.27 - samples/sec: 1649.88 - lr: 0.000007 - momentum: 0.000000
177
+ 2025-05-08 16:50:37,948 epoch 7 - iter 837/937 - loss 0.00514329 - time (sec): 110.57 - samples/sec: 1653.64 - lr: 0.000007 - momentum: 0.000000
178
+ 2025-05-08 16:50:50,233 epoch 7 - iter 930/937 - loss 0.00507451 - time (sec): 122.86 - samples/sec: 1654.76 - lr: 0.000007 - momentum: 0.000000
179
+ 2025-05-08 16:50:51,087 ----------------------------------------------------------------------------------------------------
180
+ 2025-05-08 16:50:51,087 EPOCH 7 done: loss 0.0052 - lr: 0.000007
181
+ 2025-05-08 16:50:56,556 DEV : loss 0.13016767799854279 - f1-score (micro avg) 0.9601
182
+ 2025-05-08 16:50:56,581 ----------------------------------------------------------------------------------------------------
183
+ 2025-05-08 16:51:08,853 epoch 8 - iter 93/937 - loss 0.00129312 - time (sec): 12.27 - samples/sec: 1665.82 - lr: 0.000006 - momentum: 0.000000
184
+ 2025-05-08 16:51:21,108 epoch 8 - iter 186/937 - loss 0.00142083 - time (sec): 24.53 - samples/sec: 1627.81 - lr: 0.000006 - momentum: 0.000000
185
+ 2025-05-08 16:51:33,398 epoch 8 - iter 279/937 - loss 0.00195167 - time (sec): 36.82 - samples/sec: 1628.48 - lr: 0.000006 - momentum: 0.000000
186
+ 2025-05-08 16:51:47,081 epoch 8 - iter 372/937 - loss 0.00191288 - time (sec): 50.50 - samples/sec: 1599.88 - lr: 0.000006 - momentum: 0.000000
187
+ 2025-05-08 16:51:59,448 epoch 8 - iter 465/937 - loss 0.00166481 - time (sec): 62.87 - samples/sec: 1610.97 - lr: 0.000006 - momentum: 0.000000
188
+ 2025-05-08 16:52:11,719 epoch 8 - iter 558/937 - loss 0.00161414 - time (sec): 75.14 - samples/sec: 1617.95 - lr: 0.000005 - momentum: 0.000000
189
+ 2025-05-08 16:52:23,970 epoch 8 - iter 651/937 - loss 0.00157594 - time (sec): 87.39 - samples/sec: 1625.93 - lr: 0.000005 - momentum: 0.000000
190
+ 2025-05-08 16:52:36,105 epoch 8 - iter 744/937 - loss 0.00154393 - time (sec): 99.52 - samples/sec: 1628.32 - lr: 0.000005 - momentum: 0.000000
191
+ 2025-05-08 16:52:48,417 epoch 8 - iter 837/937 - loss 0.00203973 - time (sec): 111.83 - samples/sec: 1631.67 - lr: 0.000005 - momentum: 0.000000
192
+ 2025-05-08 16:53:00,806 epoch 8 - iter 930/937 - loss 0.00201068 - time (sec): 124.22 - samples/sec: 1635.01 - lr: 0.000004 - momentum: 0.000000
193
+ 2025-05-08 16:53:01,700 ----------------------------------------------------------------------------------------------------
194
+ 2025-05-08 16:53:01,700 EPOCH 8 done: loss 0.0020 - lr: 0.000004
195
+ 2025-05-08 16:53:07,171 DEV : loss 0.13071465492248535 - f1-score (micro avg) 0.963
196
+ 2025-05-08 16:53:07,196 ----------------------------------------------------------------------------------------------------
197
+ 2025-05-08 16:53:19,384 epoch 9 - iter 93/937 - loss 0.00241956 - time (sec): 12.19 - samples/sec: 1618.47 - lr: 0.000004 - momentum: 0.000000
198
+ 2025-05-08 16:53:31,629 epoch 9 - iter 186/937 - loss 0.00180043 - time (sec): 24.43 - samples/sec: 1619.30 - lr: 0.000004 - momentum: 0.000000
199
+ 2025-05-08 16:53:43,993 epoch 9 - iter 279/937 - loss 0.00146007 - time (sec): 36.80 - samples/sec: 1610.43 - lr: 0.000004 - momentum: 0.000000
200
+ 2025-05-08 16:53:56,261 epoch 9 - iter 372/937 - loss 0.00125696 - time (sec): 49.06 - samples/sec: 1630.15 - lr: 0.000004 - momentum: 0.000000
201
+ 2025-05-08 16:54:08,611 epoch 9 - iter 465/937 - loss 0.00108053 - time (sec): 61.41 - samples/sec: 1652.78 - lr: 0.000003 - momentum: 0.000000
202
+ 2025-05-08 16:54:20,898 epoch 9 - iter 558/937 - loss 0.00138047 - time (sec): 73.70 - samples/sec: 1649.28 - lr: 0.000003 - momentum: 0.000000
203
+ 2025-05-08 16:54:33,205 epoch 9 - iter 651/937 - loss 0.00118774 - time (sec): 86.01 - samples/sec: 1649.73 - lr: 0.000003 - momentum: 0.000000
204
+ 2025-05-08 16:54:45,518 epoch 9 - iter 744/937 - loss 0.00108115 - time (sec): 98.32 - samples/sec: 1655.74 - lr: 0.000003 - momentum: 0.000000
205
+ 2025-05-08 16:54:57,805 epoch 9 - iter 837/937 - loss 0.00100400 - time (sec): 110.61 - samples/sec: 1653.22 - lr: 0.000002 - momentum: 0.000000
206
+ 2025-05-08 16:55:10,050 epoch 9 - iter 930/937 - loss 0.00119770 - time (sec): 122.85 - samples/sec: 1652.52 - lr: 0.000002 - momentum: 0.000000
207
+ 2025-05-08 16:55:10,937 ----------------------------------------------------------------------------------------------------
208
+ 2025-05-08 16:55:10,937 EPOCH 9 done: loss 0.0012 - lr: 0.000002
209
+ 2025-05-08 16:55:16,406 DEV : loss 0.13402138650417328 - f1-score (micro avg) 0.9658
210
+ 2025-05-08 16:55:16,432 saving best model
211
+ 2025-05-08 16:55:17,424 ----------------------------------------------------------------------------------------------------
212
+ 2025-05-08 16:55:29,704 epoch 10 - iter 93/937 - loss 0.00164854 - time (sec): 12.28 - samples/sec: 1574.90 - lr: 0.000002 - momentum: 0.000000
213
+ 2025-05-08 16:55:41,942 epoch 10 - iter 186/937 - loss 0.00084318 - time (sec): 24.52 - samples/sec: 1599.16 - lr: 0.000002 - momentum: 0.000000
214
+ 2025-05-08 16:55:54,292 epoch 10 - iter 279/937 - loss 0.00062474 - time (sec): 36.87 - samples/sec: 1628.86 - lr: 0.000002 - momentum: 0.000000
215
+ 2025-05-08 16:56:06,581 epoch 10 - iter 372/937 - loss 0.00047798 - time (sec): 49.16 - samples/sec: 1629.04 - lr: 0.000001 - momentum: 0.000000
216
+ 2025-05-08 16:56:18,756 epoch 10 - iter 465/937 - loss 0.00038804 - time (sec): 61.33 - samples/sec: 1636.96 - lr: 0.000001 - momentum: 0.000000
217
+ 2025-05-08 16:56:32,377 epoch 10 - iter 558/937 - loss 0.00039768 - time (sec): 74.95 - samples/sec: 1618.78 - lr: 0.000001 - momentum: 0.000000
218
+ 2025-05-08 16:56:44,569 epoch 10 - iter 651/937 - loss 0.00034346 - time (sec): 87.14 - samples/sec: 1629.88 - lr: 0.000001 - momentum: 0.000000
219
+ 2025-05-08 16:56:56,802 epoch 10 - iter 744/937 - loss 0.00053786 - time (sec): 99.38 - samples/sec: 1636.84 - lr: 0.000000 - momentum: 0.000000
220
+ 2025-05-08 16:57:09,165 epoch 10 - iter 837/937 - loss 0.00068597 - time (sec): 111.74 - samples/sec: 1634.46 - lr: 0.000000 - momentum: 0.000000
221
+ 2025-05-08 16:57:21,552 epoch 10 - iter 930/937 - loss 0.00067756 - time (sec): 124.13 - samples/sec: 1637.73 - lr: 0.000000 - momentum: 0.000000
222
+ 2025-05-08 16:57:22,414 ----------------------------------------------------------------------------------------------------
223
+ 2025-05-08 16:57:22,414 EPOCH 10 done: loss 0.0007 - lr: 0.000000
224
+ 2025-05-08 16:57:27,886 DEV : loss 0.13694016635417938 - f1-score (micro avg) 0.9645
225
+ 2025-05-08 16:57:29,653 ----------------------------------------------------------------------------------------------------
226
+ 2025-05-08 16:57:29,654 Loading model from best epoch ...
227
+ 2025-05-08 16:57:29,959 --------------------------------------------------
228
+ 2025-05-08 16:57:29,959 - Loading SequenceTagger
229
+ 2025-05-08 16:57:31,867 - Predicts 17 classes: ['O', 'S-LOC', 'B-LOC', 'E-LOC', 'I-LOC', 'S-PER', 'B-PER', 'E-PER', 'I-PER', 'S-ORG', 'B-ORG', 'E-ORG', 'I-ORG', 'S-MISC', 'B-MISC', 'E-MISC', 'I-MISC']
230
+ 2025-05-08 16:57:31,940 --------------------------------------------------
231
+ 2025-05-08 16:57:31,941 - Model license: No license information available
232
+ 2025-05-08 16:57:31,941 --------------------------------------------------
233
+ 2025-05-08 16:57:36,988
234
+ Results:
235
+ - F-score (micro) 0.9234
236
+ - F-score (macro) 0.9111
237
+ - Accuracy 0.8908
238
+
239
+ By class:
240
+ precision recall f1-score support
241
+
242
+ ORG 0.8935 0.9145 0.9039 1661
243
+ LOC 0.9315 0.9376 0.9346 1668
244
+ PER 0.9811 0.9635 0.9722 1617
245
+ MISC 0.8178 0.8504 0.8338 702
246
+
247
+ micro avg 0.9194 0.9274 0.9234 5648
248
+ macro avg 0.9060 0.9165 0.9111 5648
249
+ weighted avg 0.9204 0.9274 0.9238 5648
250
+
251
+ 2025-05-08 16:57:36,988 ----------------------------------------------------------------------------------------------------