Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmteams-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
9ab18e5
·
1 Parent(s): 990aa1b

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697599161.3ae7c61396a7.1160.18 +3 -0
  5. test.tsv +0 -0
  6. training.log +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d87d5c5c5a99cf9c5b4a31ad8ae2980b2575597e3b23061f308313d3c0e542cf
3
+ size 440954373
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 03:26:20 0.0000 0.4224 0.1225 0.2121 0.1061 0.1414 0.0761
3
+ 2 03:33:22 0.0000 0.1790 0.1425 0.2629 0.3769 0.3097 0.1844
4
+ 3 03:40:28 0.0000 0.1327 0.1821 0.2953 0.4811 0.3660 0.2258
5
+ 4 03:47:23 0.0000 0.0973 0.2133 0.2659 0.5303 0.3542 0.2167
6
+ 5 03:54:16 0.0000 0.0700 0.3877 0.2526 0.6061 0.3565 0.2181
7
+ 6 04:01:22 0.0000 0.0482 0.4169 0.2772 0.6080 0.3808 0.2366
8
+ 7 04:08:21 0.0000 0.0338 0.4146 0.2673 0.5777 0.3655 0.2253
9
+ 8 04:15:19 0.0000 0.0237 0.5072 0.2510 0.6231 0.3578 0.2189
10
+ 9 04:22:16 0.0000 0.0156 0.4875 0.2749 0.6174 0.3804 0.2364
11
+ 10 04:29:15 0.0000 0.0100 0.4915 0.2850 0.6098 0.3884 0.2425
runs/events.out.tfevents.1697599161.3ae7c61396a7.1160.18 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56de3e6e91fa30c4aae4f2846b677c54f781a8cfd76974b58dd58d606526ef37
3
+ size 2923780
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 03:19:21,924 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 03:19:21,925 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): ElectraModel(
5
+ (embeddings): ElectraEmbeddings(
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): ElectraEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x ElectraLayer(
15
+ (attention): ElectraAttention(
16
+ (self): ElectraSelfAttention(
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): ElectraSelfOutput(
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): ElectraIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): ElectraOutput(
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
+ )
41
+ )
42
+ (locked_dropout): LockedDropout(p=0.5)
43
+ (linear): Linear(in_features=768, out_features=17, bias=True)
44
+ (loss_function): CrossEntropyLoss()
45
+ )"
46
+ 2023-10-18 03:19:21,926 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-18 03:19:21,926 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
48
+ - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
49
+ 2023-10-18 03:19:21,926 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-18 03:19:21,926 Train: 20847 sentences
51
+ 2023-10-18 03:19:21,926 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-18 03:19:21,926 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-18 03:19:21,926 Training Params:
54
+ 2023-10-18 03:19:21,926 - learning_rate: "3e-05"
55
+ 2023-10-18 03:19:21,926 - mini_batch_size: "4"
56
+ 2023-10-18 03:19:21,926 - max_epochs: "10"
57
+ 2023-10-18 03:19:21,927 - shuffle: "True"
58
+ 2023-10-18 03:19:21,927 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-18 03:19:21,927 Plugins:
60
+ 2023-10-18 03:19:21,927 - TensorboardLogger
61
+ 2023-10-18 03:19:21,927 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-18 03:19:21,927 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 03:19:21,927 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-18 03:19:21,927 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-18 03:19:21,927 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-18 03:19:21,927 Computation:
67
+ 2023-10-18 03:19:21,927 - compute on device: cuda:0
68
+ 2023-10-18 03:19:21,927 - embedding storage: none
69
+ 2023-10-18 03:19:21,927 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 03:19:21,927 Model training base path: "hmbench-newseye/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
71
+ 2023-10-18 03:19:21,928 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-18 03:19:21,928 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-18 03:19:21,928 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-18 03:20:04,382 epoch 1 - iter 521/5212 - loss 1.83279163 - time (sec): 42.45 - samples/sec: 833.82 - lr: 0.000003 - momentum: 0.000000
75
+ 2023-10-18 03:20:45,389 epoch 1 - iter 1042/5212 - loss 1.12925636 - time (sec): 83.46 - samples/sec: 850.03 - lr: 0.000006 - momentum: 0.000000
76
+ 2023-10-18 03:21:26,473 epoch 1 - iter 1563/5212 - loss 0.85023267 - time (sec): 124.54 - samples/sec: 880.12 - lr: 0.000009 - momentum: 0.000000
77
+ 2023-10-18 03:22:08,372 epoch 1 - iter 2084/5212 - loss 0.71001482 - time (sec): 166.44 - samples/sec: 882.79 - lr: 0.000012 - momentum: 0.000000
78
+ 2023-10-18 03:22:48,833 epoch 1 - iter 2605/5212 - loss 0.61025461 - time (sec): 206.90 - samples/sec: 892.93 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-18 03:23:30,099 epoch 1 - iter 3126/5212 - loss 0.54601662 - time (sec): 248.17 - samples/sec: 900.37 - lr: 0.000018 - momentum: 0.000000
80
+ 2023-10-18 03:24:12,016 epoch 1 - iter 3647/5212 - loss 0.50727328 - time (sec): 290.09 - samples/sec: 891.41 - lr: 0.000021 - momentum: 0.000000
81
+ 2023-10-18 03:24:52,068 epoch 1 - iter 4168/5212 - loss 0.47535521 - time (sec): 330.14 - samples/sec: 889.28 - lr: 0.000024 - momentum: 0.000000
82
+ 2023-10-18 03:25:33,071 epoch 1 - iter 4689/5212 - loss 0.44504840 - time (sec): 371.14 - samples/sec: 896.04 - lr: 0.000027 - momentum: 0.000000
83
+ 2023-10-18 03:26:12,456 epoch 1 - iter 5210/5212 - loss 0.42244480 - time (sec): 410.53 - samples/sec: 894.71 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-18 03:26:12,607 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-18 03:26:12,608 EPOCH 1 done: loss 0.4224 - lr: 0.000030
86
+ 2023-10-18 03:26:20,073 DEV : loss 0.12248191982507706 - f1-score (micro avg) 0.1414
87
+ 2023-10-18 03:26:20,122 saving best model
88
+ 2023-10-18 03:26:20,659 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 03:27:00,468 epoch 2 - iter 521/5212 - loss 0.20584910 - time (sec): 39.81 - samples/sec: 934.78 - lr: 0.000030 - momentum: 0.000000
90
+ 2023-10-18 03:27:41,648 epoch 2 - iter 1042/5212 - loss 0.19499185 - time (sec): 80.99 - samples/sec: 883.26 - lr: 0.000029 - momentum: 0.000000
91
+ 2023-10-18 03:28:22,295 epoch 2 - iter 1563/5212 - loss 0.19416752 - time (sec): 121.63 - samples/sec: 884.60 - lr: 0.000029 - momentum: 0.000000
92
+ 2023-10-18 03:29:03,551 epoch 2 - iter 2084/5212 - loss 0.19055400 - time (sec): 162.89 - samples/sec: 891.11 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-18 03:29:44,683 epoch 2 - iter 2605/5212 - loss 0.18640454 - time (sec): 204.02 - samples/sec: 879.40 - lr: 0.000028 - momentum: 0.000000
94
+ 2023-10-18 03:30:25,791 epoch 2 - iter 3126/5212 - loss 0.18601493 - time (sec): 245.13 - samples/sec: 881.18 - lr: 0.000028 - momentum: 0.000000
95
+ 2023-10-18 03:31:07,128 epoch 2 - iter 3647/5212 - loss 0.18563916 - time (sec): 286.47 - samples/sec: 888.02 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-18 03:31:47,505 epoch 2 - iter 4168/5212 - loss 0.18313815 - time (sec): 326.84 - samples/sec: 892.84 - lr: 0.000027 - momentum: 0.000000
97
+ 2023-10-18 03:32:28,316 epoch 2 - iter 4689/5212 - loss 0.18165928 - time (sec): 367.66 - samples/sec: 890.76 - lr: 0.000027 - momentum: 0.000000
98
+ 2023-10-18 03:33:10,125 epoch 2 - iter 5210/5212 - loss 0.17883797 - time (sec): 409.46 - samples/sec: 896.84 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-18 03:33:10,277 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-18 03:33:10,278 EPOCH 2 done: loss 0.1790 - lr: 0.000027
101
+ 2023-10-18 03:33:22,810 DEV : loss 0.14249852299690247 - f1-score (micro avg) 0.3097
102
+ 2023-10-18 03:33:22,866 saving best model
103
+ 2023-10-18 03:33:24,283 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 03:34:05,071 epoch 3 - iter 521/5212 - loss 0.11933402 - time (sec): 40.78 - samples/sec: 915.69 - lr: 0.000026 - momentum: 0.000000
105
+ 2023-10-18 03:34:45,583 epoch 3 - iter 1042/5212 - loss 0.12194809 - time (sec): 81.30 - samples/sec: 939.18 - lr: 0.000026 - momentum: 0.000000
106
+ 2023-10-18 03:35:27,689 epoch 3 - iter 1563/5212 - loss 0.12990759 - time (sec): 123.40 - samples/sec: 923.92 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-18 03:36:09,840 epoch 3 - iter 2084/5212 - loss 0.13056535 - time (sec): 165.55 - samples/sec: 921.71 - lr: 0.000025 - momentum: 0.000000
108
+ 2023-10-18 03:36:50,240 epoch 3 - iter 2605/5212 - loss 0.12863125 - time (sec): 205.95 - samples/sec: 908.62 - lr: 0.000025 - momentum: 0.000000
109
+ 2023-10-18 03:37:31,554 epoch 3 - iter 3126/5212 - loss 0.13055287 - time (sec): 247.27 - samples/sec: 908.77 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-18 03:38:12,182 epoch 3 - iter 3647/5212 - loss 0.12654940 - time (sec): 287.89 - samples/sec: 903.70 - lr: 0.000024 - momentum: 0.000000
111
+ 2023-10-18 03:38:54,124 epoch 3 - iter 4168/5212 - loss 0.12600072 - time (sec): 329.84 - samples/sec: 900.17 - lr: 0.000024 - momentum: 0.000000
112
+ 2023-10-18 03:39:36,727 epoch 3 - iter 4689/5212 - loss 0.13025416 - time (sec): 372.44 - samples/sec: 890.66 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-18 03:40:17,584 epoch 3 - iter 5210/5212 - loss 0.13278228 - time (sec): 413.30 - samples/sec: 888.59 - lr: 0.000023 - momentum: 0.000000
114
+ 2023-10-18 03:40:17,738 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-18 03:40:17,739 EPOCH 3 done: loss 0.1327 - lr: 0.000023
116
+ 2023-10-18 03:40:28,207 DEV : loss 0.1821216344833374 - f1-score (micro avg) 0.366
117
+ 2023-10-18 03:40:28,256 saving best model
118
+ 2023-10-18 03:40:30,522 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 03:41:10,220 epoch 4 - iter 521/5212 - loss 0.08970416 - time (sec): 39.69 - samples/sec: 956.10 - lr: 0.000023 - momentum: 0.000000
120
+ 2023-10-18 03:41:49,917 epoch 4 - iter 1042/5212 - loss 0.09964785 - time (sec): 79.39 - samples/sec: 939.40 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-18 03:42:29,451 epoch 4 - iter 1563/5212 - loss 0.09672256 - time (sec): 118.92 - samples/sec: 932.14 - lr: 0.000022 - momentum: 0.000000
122
+ 2023-10-18 03:43:09,777 epoch 4 - iter 2084/5212 - loss 0.09502537 - time (sec): 159.25 - samples/sec: 907.60 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-18 03:43:49,792 epoch 4 - iter 2605/5212 - loss 0.09393171 - time (sec): 199.27 - samples/sec: 906.53 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-18 03:44:30,616 epoch 4 - iter 3126/5212 - loss 0.09122606 - time (sec): 240.09 - samples/sec: 915.38 - lr: 0.000021 - momentum: 0.000000
125
+ 2023-10-18 03:45:10,328 epoch 4 - iter 3647/5212 - loss 0.09624500 - time (sec): 279.80 - samples/sec: 908.35 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-18 03:45:50,998 epoch 4 - iter 4168/5212 - loss 0.09654796 - time (sec): 320.47 - samples/sec: 917.98 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-18 03:46:31,034 epoch 4 - iter 4689/5212 - loss 0.09538458 - time (sec): 360.51 - samples/sec: 914.47 - lr: 0.000020 - momentum: 0.000000
128
+ 2023-10-18 03:47:12,757 epoch 4 - iter 5210/5212 - loss 0.09731958 - time (sec): 402.23 - samples/sec: 913.10 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-18 03:47:12,913 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-18 03:47:12,914 EPOCH 4 done: loss 0.0973 - lr: 0.000020
131
+ 2023-10-18 03:47:23,934 DEV : loss 0.21330803632736206 - f1-score (micro avg) 0.3542
132
+ 2023-10-18 03:47:23,987 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-18 03:48:05,589 epoch 5 - iter 521/5212 - loss 0.06634032 - time (sec): 41.60 - samples/sec: 906.45 - lr: 0.000020 - momentum: 0.000000
134
+ 2023-10-18 03:48:47,702 epoch 5 - iter 1042/5212 - loss 0.06492006 - time (sec): 83.71 - samples/sec: 899.24 - lr: 0.000019 - momentum: 0.000000
135
+ 2023-10-18 03:49:28,821 epoch 5 - iter 1563/5212 - loss 0.06508209 - time (sec): 124.83 - samples/sec: 925.06 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-18 03:50:09,901 epoch 5 - iter 2084/5212 - loss 0.06506006 - time (sec): 165.91 - samples/sec: 924.91 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-18 03:50:48,532 epoch 5 - iter 2605/5212 - loss 0.06689591 - time (sec): 204.54 - samples/sec: 931.22 - lr: 0.000018 - momentum: 0.000000
138
+ 2023-10-18 03:51:24,605 epoch 5 - iter 3126/5212 - loss 0.06985311 - time (sec): 240.62 - samples/sec: 929.27 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-18 03:52:05,734 epoch 5 - iter 3647/5212 - loss 0.06917979 - time (sec): 281.74 - samples/sec: 912.20 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-18 03:52:46,634 epoch 5 - iter 4168/5212 - loss 0.07243061 - time (sec): 322.64 - samples/sec: 913.48 - lr: 0.000017 - momentum: 0.000000
141
+ 2023-10-18 03:53:26,449 epoch 5 - iter 4689/5212 - loss 0.07211728 - time (sec): 362.46 - samples/sec: 913.38 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-18 03:54:05,931 epoch 5 - iter 5210/5212 - loss 0.07004349 - time (sec): 401.94 - samples/sec: 913.89 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-18 03:54:06,088 ----------------------------------------------------------------------------------------------------
144
+ 2023-10-18 03:54:06,088 EPOCH 5 done: loss 0.0700 - lr: 0.000017
145
+ 2023-10-18 03:54:16,814 DEV : loss 0.38766807317733765 - f1-score (micro avg) 0.3565
146
+ 2023-10-18 03:54:16,863 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-18 03:54:55,980 epoch 6 - iter 521/5212 - loss 0.04183136 - time (sec): 39.11 - samples/sec: 867.75 - lr: 0.000016 - momentum: 0.000000
148
+ 2023-10-18 03:55:37,034 epoch 6 - iter 1042/5212 - loss 0.04501640 - time (sec): 80.17 - samples/sec: 859.08 - lr: 0.000016 - momentum: 0.000000
149
+ 2023-10-18 03:56:19,218 epoch 6 - iter 1563/5212 - loss 0.04580421 - time (sec): 122.35 - samples/sec: 866.16 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-18 03:57:01,956 epoch 6 - iter 2084/5212 - loss 0.04530397 - time (sec): 165.09 - samples/sec: 886.96 - lr: 0.000015 - momentum: 0.000000
151
+ 2023-10-18 03:57:44,263 epoch 6 - iter 2605/5212 - loss 0.04814648 - time (sec): 207.40 - samples/sec: 878.33 - lr: 0.000015 - momentum: 0.000000
152
+ 2023-10-18 03:58:26,365 epoch 6 - iter 3126/5212 - loss 0.04711126 - time (sec): 249.50 - samples/sec: 870.16 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-18 03:59:06,658 epoch 6 - iter 3647/5212 - loss 0.04888926 - time (sec): 289.79 - samples/sec: 876.91 - lr: 0.000014 - momentum: 0.000000
154
+ 2023-10-18 03:59:49,040 epoch 6 - iter 4168/5212 - loss 0.04790461 - time (sec): 332.17 - samples/sec: 878.60 - lr: 0.000014 - momentum: 0.000000
155
+ 2023-10-18 04:00:30,324 epoch 6 - iter 4689/5212 - loss 0.04810702 - time (sec): 373.46 - samples/sec: 886.03 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-18 04:01:11,444 epoch 6 - iter 5210/5212 - loss 0.04828416 - time (sec): 414.58 - samples/sec: 885.43 - lr: 0.000013 - momentum: 0.000000
157
+ 2023-10-18 04:01:11,621 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-18 04:01:11,621 EPOCH 6 done: loss 0.0482 - lr: 0.000013
159
+ 2023-10-18 04:01:22,184 DEV : loss 0.4169054627418518 - f1-score (micro avg) 0.3808
160
+ 2023-10-18 04:01:22,236 saving best model
161
+ 2023-10-18 04:01:23,661 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-18 04:02:03,534 epoch 7 - iter 521/5212 - loss 0.03038810 - time (sec): 39.87 - samples/sec: 879.71 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-18 04:02:44,330 epoch 7 - iter 1042/5212 - loss 0.02730908 - time (sec): 80.67 - samples/sec: 915.38 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-18 04:03:23,839 epoch 7 - iter 1563/5212 - loss 0.02928101 - time (sec): 120.17 - samples/sec: 912.64 - lr: 0.000012 - momentum: 0.000000
165
+ 2023-10-18 04:04:03,990 epoch 7 - iter 2084/5212 - loss 0.03005702 - time (sec): 160.33 - samples/sec: 921.96 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-18 04:04:44,794 epoch 7 - iter 2605/5212 - loss 0.03238450 - time (sec): 201.13 - samples/sec: 924.26 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-18 04:05:26,677 epoch 7 - iter 3126/5212 - loss 0.03278459 - time (sec): 243.01 - samples/sec: 914.60 - lr: 0.000011 - momentum: 0.000000
168
+ 2023-10-18 04:06:08,135 epoch 7 - iter 3647/5212 - loss 0.03473285 - time (sec): 284.47 - samples/sec: 910.94 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-18 04:06:47,890 epoch 7 - iter 4168/5212 - loss 0.03455628 - time (sec): 324.23 - samples/sec: 902.33 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-18 04:07:27,879 epoch 7 - iter 4689/5212 - loss 0.03445089 - time (sec): 364.21 - samples/sec: 906.21 - lr: 0.000010 - momentum: 0.000000
171
+ 2023-10-18 04:08:09,216 epoch 7 - iter 5210/5212 - loss 0.03377586 - time (sec): 405.55 - samples/sec: 905.67 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-18 04:08:09,373 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-18 04:08:09,373 EPOCH 7 done: loss 0.0338 - lr: 0.000010
174
+ 2023-10-18 04:08:21,313 DEV : loss 0.4146248996257782 - f1-score (micro avg) 0.3655
175
+ 2023-10-18 04:08:21,367 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-18 04:09:01,416 epoch 8 - iter 521/5212 - loss 0.02016099 - time (sec): 40.05 - samples/sec: 863.03 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-18 04:09:40,784 epoch 8 - iter 1042/5212 - loss 0.02013029 - time (sec): 79.41 - samples/sec: 899.63 - lr: 0.000009 - momentum: 0.000000
178
+ 2023-10-18 04:10:20,681 epoch 8 - iter 1563/5212 - loss 0.02218114 - time (sec): 119.31 - samples/sec: 890.34 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-18 04:11:02,074 epoch 8 - iter 2084/5212 - loss 0.02230411 - time (sec): 160.71 - samples/sec: 894.26 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-18 04:11:43,582 epoch 8 - iter 2605/5212 - loss 0.02391278 - time (sec): 202.21 - samples/sec: 889.11 - lr: 0.000008 - momentum: 0.000000
181
+ 2023-10-18 04:12:25,493 epoch 8 - iter 3126/5212 - loss 0.02395996 - time (sec): 244.12 - samples/sec: 890.34 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-18 04:13:05,968 epoch 8 - iter 3647/5212 - loss 0.02297499 - time (sec): 284.60 - samples/sec: 893.24 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-18 04:13:47,054 epoch 8 - iter 4168/5212 - loss 0.02294454 - time (sec): 325.68 - samples/sec: 904.67 - lr: 0.000007 - momentum: 0.000000
184
+ 2023-10-18 04:14:28,015 epoch 8 - iter 4689/5212 - loss 0.02340719 - time (sec): 366.65 - samples/sec: 905.03 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-18 04:15:08,109 epoch 8 - iter 5210/5212 - loss 0.02374256 - time (sec): 406.74 - samples/sec: 903.14 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-18 04:15:08,256 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-18 04:15:08,257 EPOCH 8 done: loss 0.0237 - lr: 0.000007
188
+ 2023-10-18 04:15:19,800 DEV : loss 0.5071883201599121 - f1-score (micro avg) 0.3578
189
+ 2023-10-18 04:15:19,852 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-18 04:15:59,436 epoch 9 - iter 521/5212 - loss 0.01442529 - time (sec): 39.58 - samples/sec: 947.63 - lr: 0.000006 - momentum: 0.000000
191
+ 2023-10-18 04:16:38,979 epoch 9 - iter 1042/5212 - loss 0.01283333 - time (sec): 79.13 - samples/sec: 932.88 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-18 04:17:18,528 epoch 9 - iter 1563/5212 - loss 0.01456866 - time (sec): 118.67 - samples/sec: 935.39 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-18 04:17:59,316 epoch 9 - iter 2084/5212 - loss 0.01491522 - time (sec): 159.46 - samples/sec: 941.58 - lr: 0.000005 - momentum: 0.000000
194
+ 2023-10-18 04:18:39,643 epoch 9 - iter 2605/5212 - loss 0.01523402 - time (sec): 199.79 - samples/sec: 935.30 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-18 04:19:19,300 epoch 9 - iter 3126/5212 - loss 0.01601374 - time (sec): 239.45 - samples/sec: 920.30 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-18 04:20:00,234 epoch 9 - iter 3647/5212 - loss 0.01624935 - time (sec): 280.38 - samples/sec: 921.98 - lr: 0.000004 - momentum: 0.000000
197
+ 2023-10-18 04:20:41,771 epoch 9 - iter 4168/5212 - loss 0.01580964 - time (sec): 321.92 - samples/sec: 922.24 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-18 04:21:22,918 epoch 9 - iter 4689/5212 - loss 0.01570222 - time (sec): 363.06 - samples/sec: 909.44 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-18 04:22:04,987 epoch 9 - iter 5210/5212 - loss 0.01558208 - time (sec): 405.13 - samples/sec: 906.76 - lr: 0.000003 - momentum: 0.000000
200
+ 2023-10-18 04:22:05,144 ----------------------------------------------------------------------------------------------------
201
+ 2023-10-18 04:22:05,144 EPOCH 9 done: loss 0.0156 - lr: 0.000003
202
+ 2023-10-18 04:22:16,655 DEV : loss 0.48745957016944885 - f1-score (micro avg) 0.3804
203
+ 2023-10-18 04:22:16,707 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-18 04:22:57,456 epoch 10 - iter 521/5212 - loss 0.01136545 - time (sec): 40.75 - samples/sec: 923.15 - lr: 0.000003 - momentum: 0.000000
205
+ 2023-10-18 04:23:38,255 epoch 10 - iter 1042/5212 - loss 0.00982628 - time (sec): 81.54 - samples/sec: 923.95 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-18 04:24:19,541 epoch 10 - iter 1563/5212 - loss 0.00965992 - time (sec): 122.83 - samples/sec: 925.48 - lr: 0.000002 - momentum: 0.000000
207
+ 2023-10-18 04:25:00,868 epoch 10 - iter 2084/5212 - loss 0.00994628 - time (sec): 164.16 - samples/sec: 915.83 - lr: 0.000002 - momentum: 0.000000
208
+ 2023-10-18 04:25:41,468 epoch 10 - iter 2605/5212 - loss 0.00974175 - time (sec): 204.76 - samples/sec: 917.17 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-18 04:26:20,930 epoch 10 - iter 3126/5212 - loss 0.00970478 - time (sec): 244.22 - samples/sec: 908.66 - lr: 0.000001 - momentum: 0.000000
210
+ 2023-10-18 04:27:01,441 epoch 10 - iter 3647/5212 - loss 0.00972024 - time (sec): 284.73 - samples/sec: 905.91 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-18 04:27:42,935 epoch 10 - iter 4168/5212 - loss 0.00949818 - time (sec): 326.22 - samples/sec: 899.83 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-18 04:28:23,657 epoch 10 - iter 4689/5212 - loss 0.01005578 - time (sec): 366.95 - samples/sec: 903.98 - lr: 0.000000 - momentum: 0.000000
213
+ 2023-10-18 04:29:04,033 epoch 10 - iter 5210/5212 - loss 0.01004520 - time (sec): 407.32 - samples/sec: 901.82 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-18 04:29:04,178 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-18 04:29:04,178 EPOCH 10 done: loss 0.0100 - lr: 0.000000
216
+ 2023-10-18 04:29:15,567 DEV : loss 0.49148064851760864 - f1-score (micro avg) 0.3884
217
+ 2023-10-18 04:29:15,625 saving best model
218
+ 2023-10-18 04:29:17,553 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-18 04:29:17,556 Loading model from best epoch ...
220
+ 2023-10-18 04:29:19,877 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
221
+ 2023-10-18 04:29:38,571
222
+ Results:
223
+ - F-score (micro) 0.4881
224
+ - F-score (macro) 0.3383
225
+ - Accuracy 0.3269
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.5096 0.5914 0.5475 1214
231
+ PER 0.4402 0.5012 0.4688 808
232
+ ORG 0.3263 0.3484 0.3370 353
233
+ HumanProd 0.0000 0.0000 0.0000 15
234
+
235
+ micro avg 0.4589 0.5213 0.4881 2390
236
+ macro avg 0.3190 0.3603 0.3383 2390
237
+ weighted avg 0.4559 0.5213 0.4863 2390
238
+
239
+ 2023-10-18 04:29:38,571 ----------------------------------------------------------------------------------------------------