Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert_64k-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
91d1057
•
1 Parent(s): fac0c50

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +241 -0
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-25 12:04:43,420 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 12:04:43,421 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(64001, 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-25 12:04:43,421 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 12:04:43,421 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
52
+ - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
53
+ 2023-10-25 12:04:43,421 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 12:04:43,421 Train: 20847 sentences
55
+ 2023-10-25 12:04:43,421 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 12:04:43,421 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 12:04:43,421 Training Params:
58
+ 2023-10-25 12:04:43,421 - learning_rate: "5e-05"
59
+ 2023-10-25 12:04:43,421 - mini_batch_size: "8"
60
+ 2023-10-25 12:04:43,421 - max_epochs: "10"
61
+ 2023-10-25 12:04:43,421 - shuffle: "True"
62
+ 2023-10-25 12:04:43,421 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 12:04:43,421 Plugins:
64
+ 2023-10-25 12:04:43,421 - TensorboardLogger
65
+ 2023-10-25 12:04:43,421 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 12:04:43,421 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 12:04:43,422 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 12:04:43,422 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 12:04:43,422 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 12:04:43,422 Computation:
71
+ 2023-10-25 12:04:43,422 - compute on device: cuda:0
72
+ 2023-10-25 12:04:43,422 - embedding storage: none
73
+ 2023-10-25 12:04:43,422 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 12:04:43,422 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
75
+ 2023-10-25 12:04:43,422 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 12:04:43,422 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 12:04:43,422 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 12:04:57,329 epoch 1 - iter 260/2606 - loss 1.33404298 - time (sec): 13.91 - samples/sec: 2676.54 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-25 12:05:11,911 epoch 1 - iter 520/2606 - loss 0.81750725 - time (sec): 28.49 - samples/sec: 2698.06 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-25 12:05:26,285 epoch 1 - iter 780/2606 - loss 0.63734009 - time (sec): 42.86 - samples/sec: 2662.27 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-25 12:05:40,386 epoch 1 - iter 1040/2606 - loss 0.54688049 - time (sec): 56.96 - samples/sec: 2609.80 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-25 12:05:54,652 epoch 1 - iter 1300/2606 - loss 0.48271010 - time (sec): 71.23 - samples/sec: 2585.48 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-25 12:06:09,224 epoch 1 - iter 1560/2606 - loss 0.43686814 - time (sec): 85.80 - samples/sec: 2591.06 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-25 12:06:22,977 epoch 1 - iter 1820/2606 - loss 0.40324464 - time (sec): 99.55 - samples/sec: 2576.82 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-25 12:06:37,534 epoch 1 - iter 2080/2606 - loss 0.37960994 - time (sec): 114.11 - samples/sec: 2560.25 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-25 12:06:51,663 epoch 1 - iter 2340/2606 - loss 0.35867330 - time (sec): 128.24 - samples/sec: 2557.80 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-25 12:07:06,164 epoch 1 - iter 2600/2606 - loss 0.33892838 - time (sec): 142.74 - samples/sec: 2569.97 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-25 12:07:06,496 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 12:07:06,496 EPOCH 1 done: loss 0.3387 - lr: 0.000050
90
+ 2023-10-25 12:07:10,522 DEV : loss 0.1899159997701645 - f1-score (micro avg) 0.3206
91
+ 2023-10-25 12:07:10,548 saving best model
92
+ 2023-10-25 12:07:11,071 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 12:07:25,878 epoch 2 - iter 260/2606 - loss 0.18263844 - time (sec): 14.81 - samples/sec: 2609.92 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-25 12:07:40,312 epoch 2 - iter 520/2606 - loss 0.17196157 - time (sec): 29.24 - samples/sec: 2598.58 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-25 12:07:54,450 epoch 2 - iter 780/2606 - loss 0.16810378 - time (sec): 43.38 - samples/sec: 2626.37 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-25 12:08:08,143 epoch 2 - iter 1040/2606 - loss 0.16310966 - time (sec): 57.07 - samples/sec: 2587.48 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-25 12:08:22,199 epoch 2 - iter 1300/2606 - loss 0.16748312 - time (sec): 71.13 - samples/sec: 2565.37 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-25 12:08:36,287 epoch 2 - iter 1560/2606 - loss 0.17242041 - time (sec): 85.21 - samples/sec: 2564.40 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-25 12:08:51,010 epoch 2 - iter 1820/2606 - loss 0.18294843 - time (sec): 99.94 - samples/sec: 2559.73 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-25 12:09:05,058 epoch 2 - iter 2080/2606 - loss 0.18014163 - time (sec): 113.99 - samples/sec: 2562.12 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-25 12:09:19,987 epoch 2 - iter 2340/2606 - loss 0.17697195 - time (sec): 128.91 - samples/sec: 2554.90 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-25 12:09:34,480 epoch 2 - iter 2600/2606 - loss 0.17543410 - time (sec): 143.41 - samples/sec: 2557.33 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-25 12:09:34,754 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 12:09:34,754 EPOCH 2 done: loss 0.1757 - lr: 0.000044
105
+ 2023-10-25 12:09:41,662 DEV : loss 0.12365195900201797 - f1-score (micro avg) 0.1991
106
+ 2023-10-25 12:09:41,687 ----------------------------------------------------------------------------------------------------
107
+ 2023-10-25 12:09:55,576 epoch 3 - iter 260/2606 - loss 0.13097015 - time (sec): 13.89 - samples/sec: 2422.06 - lr: 0.000044 - momentum: 0.000000
108
+ 2023-10-25 12:10:09,421 epoch 3 - iter 520/2606 - loss 0.13232082 - time (sec): 27.73 - samples/sec: 2442.10 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-25 12:10:24,291 epoch 3 - iter 780/2606 - loss 0.11661536 - time (sec): 42.60 - samples/sec: 2529.83 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-25 12:10:38,435 epoch 3 - iter 1040/2606 - loss 0.12084986 - time (sec): 56.75 - samples/sec: 2521.57 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-25 12:10:52,440 epoch 3 - iter 1300/2606 - loss 0.11793891 - time (sec): 70.75 - samples/sec: 2546.11 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-25 12:11:06,700 epoch 3 - iter 1560/2606 - loss 0.11620440 - time (sec): 85.01 - samples/sec: 2552.64 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-25 12:11:21,546 epoch 3 - iter 1820/2606 - loss 0.11625199 - time (sec): 99.86 - samples/sec: 2574.35 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-25 12:11:36,104 epoch 3 - iter 2080/2606 - loss 0.11570847 - time (sec): 114.42 - samples/sec: 2571.76 - lr: 0.000040 - momentum: 0.000000
115
+ 2023-10-25 12:11:50,815 epoch 3 - iter 2340/2606 - loss 0.11725099 - time (sec): 129.13 - samples/sec: 2573.15 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-25 12:12:04,853 epoch 3 - iter 2600/2606 - loss 0.11640683 - time (sec): 143.16 - samples/sec: 2561.04 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-25 12:12:05,150 ----------------------------------------------------------------------------------------------------
118
+ 2023-10-25 12:12:05,150 EPOCH 3 done: loss 0.1164 - lr: 0.000039
119
+ 2023-10-25 12:12:12,345 DEV : loss 0.18110370635986328 - f1-score (micro avg) 0.3487
120
+ 2023-10-25 12:12:12,371 saving best model
121
+ 2023-10-25 12:12:13,034 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-25 12:12:27,120 epoch 4 - iter 260/2606 - loss 0.07314164 - time (sec): 14.08 - samples/sec: 2617.27 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-25 12:12:41,122 epoch 4 - iter 520/2606 - loss 0.07274429 - time (sec): 28.09 - samples/sec: 2607.65 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-25 12:12:55,178 epoch 4 - iter 780/2606 - loss 0.07820023 - time (sec): 42.14 - samples/sec: 2600.12 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-25 12:13:09,293 epoch 4 - iter 1040/2606 - loss 0.08079671 - time (sec): 56.26 - samples/sec: 2534.91 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-25 12:13:24,132 epoch 4 - iter 1300/2606 - loss 0.08272013 - time (sec): 71.10 - samples/sec: 2541.48 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-25 12:13:38,910 epoch 4 - iter 1560/2606 - loss 0.08294572 - time (sec): 85.88 - samples/sec: 2523.61 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-25 12:13:53,194 epoch 4 - iter 1820/2606 - loss 0.08005595 - time (sec): 100.16 - samples/sec: 2534.43 - lr: 0.000035 - momentum: 0.000000
129
+ 2023-10-25 12:14:08,011 epoch 4 - iter 2080/2606 - loss 0.08143243 - time (sec): 114.98 - samples/sec: 2551.54 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-25 12:14:22,431 epoch 4 - iter 2340/2606 - loss 0.08041802 - time (sec): 129.40 - samples/sec: 2532.25 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-25 12:14:37,281 epoch 4 - iter 2600/2606 - loss 0.08144784 - time (sec): 144.25 - samples/sec: 2539.74 - lr: 0.000033 - momentum: 0.000000
132
+ 2023-10-25 12:14:37,596 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-25 12:14:37,597 EPOCH 4 done: loss 0.0813 - lr: 0.000033
134
+ 2023-10-25 12:14:44,681 DEV : loss 0.24672392010688782 - f1-score (micro avg) 0.3258
135
+ 2023-10-25 12:14:44,706 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-25 12:14:58,782 epoch 5 - iter 260/2606 - loss 0.04975046 - time (sec): 14.07 - samples/sec: 2514.00 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-25 12:15:13,022 epoch 5 - iter 520/2606 - loss 0.05842153 - time (sec): 28.31 - samples/sec: 2512.62 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-25 12:15:27,708 epoch 5 - iter 780/2606 - loss 0.05600138 - time (sec): 43.00 - samples/sec: 2555.65 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-25 12:15:42,374 epoch 5 - iter 1040/2606 - loss 0.05571986 - time (sec): 57.67 - samples/sec: 2559.74 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-25 12:15:56,321 epoch 5 - iter 1300/2606 - loss 0.05410019 - time (sec): 71.61 - samples/sec: 2592.02 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-25 12:16:11,032 epoch 5 - iter 1560/2606 - loss 0.05554265 - time (sec): 86.32 - samples/sec: 2600.43 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-25 12:16:25,229 epoch 5 - iter 1820/2606 - loss 0.05623815 - time (sec): 100.52 - samples/sec: 2591.86 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-25 12:16:39,142 epoch 5 - iter 2080/2606 - loss 0.05640579 - time (sec): 114.43 - samples/sec: 2577.66 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-25 12:16:52,969 epoch 5 - iter 2340/2606 - loss 0.05583642 - time (sec): 128.26 - samples/sec: 2566.52 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-25 12:17:07,359 epoch 5 - iter 2600/2606 - loss 0.05546860 - time (sec): 142.65 - samples/sec: 2568.74 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-25 12:17:07,671 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-25 12:17:07,671 EPOCH 5 done: loss 0.0554 - lr: 0.000028
148
+ 2023-10-25 12:17:14,198 DEV : loss 0.3652787506580353 - f1-score (micro avg) 0.3473
149
+ 2023-10-25 12:17:14,227 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-25 12:17:28,718 epoch 6 - iter 260/2606 - loss 0.04669705 - time (sec): 14.49 - samples/sec: 2618.65 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-25 12:17:44,097 epoch 6 - iter 520/2606 - loss 0.06386000 - time (sec): 29.87 - samples/sec: 2579.82 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-25 12:17:59,389 epoch 6 - iter 780/2606 - loss 0.05599133 - time (sec): 45.16 - samples/sec: 2534.51 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-25 12:18:13,860 epoch 6 - iter 1040/2606 - loss 0.05300823 - time (sec): 59.63 - samples/sec: 2554.23 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-25 12:18:27,592 epoch 6 - iter 1300/2606 - loss 0.05110818 - time (sec): 73.36 - samples/sec: 2537.33 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-25 12:18:41,721 epoch 6 - iter 1560/2606 - loss 0.04968908 - time (sec): 87.49 - samples/sec: 2540.03 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-25 12:18:56,223 epoch 6 - iter 1820/2606 - loss 0.04814738 - time (sec): 101.99 - samples/sec: 2516.29 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-25 12:19:10,794 epoch 6 - iter 2080/2606 - loss 0.04656793 - time (sec): 116.57 - samples/sec: 2511.11 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-25 12:19:24,885 epoch 6 - iter 2340/2606 - loss 0.04599298 - time (sec): 130.66 - samples/sec: 2519.24 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-25 12:19:39,483 epoch 6 - iter 2600/2606 - loss 0.04534244 - time (sec): 145.25 - samples/sec: 2523.96 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-25 12:19:39,796 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-25 12:19:39,796 EPOCH 6 done: loss 0.0453 - lr: 0.000022
162
+ 2023-10-25 12:19:46,429 DEV : loss 0.36484959721565247 - f1-score (micro avg) 0.3757
163
+ 2023-10-25 12:19:46,470 saving best model
164
+ 2023-10-25 12:19:47,101 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-25 12:20:02,047 epoch 7 - iter 260/2606 - loss 0.02579681 - time (sec): 14.94 - samples/sec: 2481.46 - lr: 0.000022 - momentum: 0.000000
166
+ 2023-10-25 12:20:17,070 epoch 7 - iter 520/2606 - loss 0.02723529 - time (sec): 29.97 - samples/sec: 2514.98 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-25 12:20:31,473 epoch 7 - iter 780/2606 - loss 0.03443679 - time (sec): 44.37 - samples/sec: 2487.49 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-25 12:20:46,443 epoch 7 - iter 1040/2606 - loss 0.05293747 - time (sec): 59.34 - samples/sec: 2490.37 - lr: 0.000020 - momentum: 0.000000
169
+ 2023-10-25 12:21:01,062 epoch 7 - iter 1300/2606 - loss 0.05896085 - time (sec): 73.96 - samples/sec: 2498.42 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-25 12:21:15,728 epoch 7 - iter 1560/2606 - loss 0.05488902 - time (sec): 88.63 - samples/sec: 2525.43 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-25 12:21:30,405 epoch 7 - iter 1820/2606 - loss 0.05514632 - time (sec): 103.30 - samples/sec: 2519.10 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-25 12:21:44,288 epoch 7 - iter 2080/2606 - loss 0.05673720 - time (sec): 117.19 - samples/sec: 2518.16 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-25 12:21:58,440 epoch 7 - iter 2340/2606 - loss 0.06075104 - time (sec): 131.34 - samples/sec: 2518.27 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-25 12:22:12,005 epoch 7 - iter 2600/2606 - loss 0.07080521 - time (sec): 144.90 - samples/sec: 2527.51 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-25 12:22:12,429 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-25 12:22:12,430 EPOCH 7 done: loss 0.0709 - lr: 0.000017
177
+ 2023-10-25 12:22:18,685 DEV : loss 0.28045564889907837 - f1-score (micro avg) 0.1776
178
+ 2023-10-25 12:22:18,711 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-25 12:22:32,882 epoch 8 - iter 260/2606 - loss 0.10479596 - time (sec): 14.17 - samples/sec: 2620.27 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-25 12:22:47,009 epoch 8 - iter 520/2606 - loss 0.07922867 - time (sec): 28.30 - samples/sec: 2666.93 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-25 12:23:00,754 epoch 8 - iter 780/2606 - loss 0.08213498 - time (sec): 42.04 - samples/sec: 2670.80 - lr: 0.000015 - momentum: 0.000000
182
+ 2023-10-25 12:23:14,774 epoch 8 - iter 1040/2606 - loss 0.08392793 - time (sec): 56.06 - samples/sec: 2676.38 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-25 12:23:29,316 epoch 8 - iter 1300/2606 - loss 0.09298697 - time (sec): 70.60 - samples/sec: 2692.58 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-25 12:23:43,269 epoch 8 - iter 1560/2606 - loss 0.09808489 - time (sec): 84.56 - samples/sec: 2660.91 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-25 12:23:57,806 epoch 8 - iter 1820/2606 - loss 0.09890906 - time (sec): 99.09 - samples/sec: 2606.94 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-25 12:24:12,049 epoch 8 - iter 2080/2606 - loss 0.10345234 - time (sec): 113.34 - samples/sec: 2615.34 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-25 12:24:26,187 epoch 8 - iter 2340/2606 - loss 0.10411809 - time (sec): 127.48 - samples/sec: 2608.57 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-25 12:24:40,125 epoch 8 - iter 2600/2606 - loss 0.10325893 - time (sec): 141.41 - samples/sec: 2593.05 - lr: 0.000011 - momentum: 0.000000
189
+ 2023-10-25 12:24:40,426 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-25 12:24:40,426 EPOCH 8 done: loss 0.1032 - lr: 0.000011
191
+ 2023-10-25 12:24:46,844 DEV : loss 0.3105942904949188 - f1-score (micro avg) 0.1121
192
+ 2023-10-25 12:24:46,869 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-25 12:25:01,469 epoch 9 - iter 260/2606 - loss 0.12293943 - time (sec): 14.60 - samples/sec: 2468.64 - lr: 0.000011 - momentum: 0.000000
194
+ 2023-10-25 12:25:15,333 epoch 9 - iter 520/2606 - loss 0.12263835 - time (sec): 28.46 - samples/sec: 2559.07 - lr: 0.000010 - momentum: 0.000000
195
+ 2023-10-25 12:25:29,557 epoch 9 - iter 780/2606 - loss 0.14482561 - time (sec): 42.69 - samples/sec: 2559.99 - lr: 0.000009 - momentum: 0.000000
196
+ 2023-10-25 12:25:43,465 epoch 9 - iter 1040/2606 - loss 0.15345955 - time (sec): 56.59 - samples/sec: 2570.82 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-25 12:25:58,521 epoch 9 - iter 1300/2606 - loss 0.15045690 - time (sec): 71.65 - samples/sec: 2560.35 - lr: 0.000008 - momentum: 0.000000
198
+ 2023-10-25 12:26:12,756 epoch 9 - iter 1560/2606 - loss 0.14920787 - time (sec): 85.89 - samples/sec: 2568.73 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-25 12:26:26,678 epoch 9 - iter 1820/2606 - loss 0.14989750 - time (sec): 99.81 - samples/sec: 2586.72 - lr: 0.000007 - momentum: 0.000000
200
+ 2023-10-25 12:26:40,643 epoch 9 - iter 2080/2606 - loss 0.15181262 - time (sec): 113.77 - samples/sec: 2578.12 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-25 12:26:55,783 epoch 9 - iter 2340/2606 - loss 0.15048265 - time (sec): 128.91 - samples/sec: 2565.28 - lr: 0.000006 - momentum: 0.000000
202
+ 2023-10-25 12:27:09,724 epoch 9 - iter 2600/2606 - loss 0.14790385 - time (sec): 142.85 - samples/sec: 2564.79 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-25 12:27:10,104 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-25 12:27:10,104 EPOCH 9 done: loss 0.1478 - lr: 0.000006
205
+ 2023-10-25 12:27:16,359 DEV : loss 0.2825768291950226 - f1-score (micro avg) 0.0264
206
+ 2023-10-25 12:27:16,386 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-25 12:27:31,009 epoch 10 - iter 260/2606 - loss 0.12349381 - time (sec): 14.62 - samples/sec: 2544.38 - lr: 0.000005 - momentum: 0.000000
208
+ 2023-10-25 12:27:44,544 epoch 10 - iter 520/2606 - loss 0.12212158 - time (sec): 28.16 - samples/sec: 2556.67 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-25 12:28:00,425 epoch 10 - iter 780/2606 - loss 0.12260170 - time (sec): 44.04 - samples/sec: 2533.39 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-25 12:28:14,921 epoch 10 - iter 1040/2606 - loss 0.12686123 - time (sec): 58.53 - samples/sec: 2549.96 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-25 12:28:28,874 epoch 10 - iter 1300/2606 - loss 0.12305199 - time (sec): 72.49 - samples/sec: 2528.69 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-25 12:28:44,113 epoch 10 - iter 1560/2606 - loss 0.12360049 - time (sec): 87.73 - samples/sec: 2501.90 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-25 12:28:58,489 epoch 10 - iter 1820/2606 - loss 0.12493764 - time (sec): 102.10 - samples/sec: 2499.93 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-25 12:29:13,302 epoch 10 - iter 2080/2606 - loss 0.12640882 - time (sec): 116.91 - samples/sec: 2502.36 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-25 12:29:27,892 epoch 10 - iter 2340/2606 - loss 0.12676932 - time (sec): 131.50 - samples/sec: 2510.93 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-25 12:29:41,917 epoch 10 - iter 2600/2606 - loss 0.12636628 - time (sec): 145.53 - samples/sec: 2521.08 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-25 12:29:42,199 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-25 12:29:42,199 EPOCH 10 done: loss 0.1263 - lr: 0.000000
219
+ 2023-10-25 12:29:49,286 DEV : loss 0.30260512232780457 - f1-score (micro avg) 0.0731
220
+ 2023-10-25 12:29:49,854 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-25 12:29:49,855 Loading model from best epoch ...
222
+ 2023-10-25 12:29:51,548 SequenceTagger predicts: Dictionary with 17 tags: 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-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
223
+ 2023-10-25 12:30:03,057
224
+ Results:
225
+ - F-score (micro) 0.451
226
+ - F-score (macro) 0.2998
227
+ - Accuracy 0.2947
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ LOC 0.4629 0.5964 0.5212 1214
233
+ PER 0.4000 0.4356 0.4171 808
234
+ ORG 0.2843 0.2408 0.2607 353
235
+ HumanProd 0.0000 0.0000 0.0000 15
236
+
237
+ micro avg 0.4210 0.4858 0.4510 2390
238
+ macro avg 0.2868 0.3182 0.2998 2390
239
+ weighted avg 0.4124 0.4858 0.4443 2390
240
+
241
+ 2023-10-25 12:30:03,057 ----------------------------------------------------------------------------------------------------