Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmteams-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
eac4e2b
·
1 Parent(s): de173b7

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.1697549736.3ae7c61396a7.1160.4 +3 -0
  5. test.tsv +0 -0
  6. training.log +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4621b1bfa94de9a7dad307201c8a4a737d50542f00d3a2fc23d61e9bb3f6608c
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 13:40:17 0.0000 0.4546 0.1286 0.2007 0.4356 0.2748 0.1597
3
+ 2 13:45:02 0.0000 0.1573 0.1841 0.2108 0.4811 0.2931 0.1727
4
+ 3 13:49:47 0.0000 0.1099 0.1961 0.2697 0.5114 0.3532 0.2155
5
+ 4 13:54:33 0.0000 0.0771 0.3065 0.2432 0.6913 0.3598 0.2200
6
+ 5 13:59:22 0.0000 0.0556 0.3809 0.2539 0.6742 0.3689 0.2270
7
+ 6 14:04:10 0.0000 0.0397 0.3598 0.2651 0.6136 0.3703 0.2283
8
+ 7 14:08:58 0.0000 0.0277 0.4652 0.2635 0.6117 0.3683 0.2265
9
+ 8 14:13:41 0.0000 0.0217 0.4531 0.2802 0.5890 0.3797 0.2354
10
+ 9 14:18:29 0.0000 0.0142 0.5336 0.2652 0.6098 0.3697 0.2274
11
+ 10 14:23:22 0.0000 0.0110 0.5365 0.2604 0.6023 0.3636 0.2230
runs/events.out.tfevents.1697549736.3ae7c61396a7.1160.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37ac0475e5889f6f04118fad5bbcde4b6f6dcc5738ddbb0d90c473d41166252d
3
+ size 1464420
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 13:35:36,187 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 13:35:36,189 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
+ 2023-10-17 13:35:36,189 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 13:35:36,189 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
+ 2023-10-17 13:35:36,189 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 13:35:36,189 Train: 20847 sentences
51
+ 2023-10-17 13:35:36,189 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 13:35:36,189 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 13:35:36,189 Training Params:
54
+ 2023-10-17 13:35:36,189 - learning_rate: "3e-05"
55
+ 2023-10-17 13:35:36,189 - mini_batch_size: "8"
56
+ 2023-10-17 13:35:36,190 - max_epochs: "10"
57
+ 2023-10-17 13:35:36,190 - shuffle: "True"
58
+ 2023-10-17 13:35:36,190 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 13:35:36,190 Plugins:
60
+ 2023-10-17 13:35:36,190 - TensorboardLogger
61
+ 2023-10-17 13:35:36,190 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 13:35:36,190 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 13:35:36,190 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 13:35:36,190 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 13:35:36,190 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 13:35:36,190 Computation:
67
+ 2023-10-17 13:35:36,190 - compute on device: cuda:0
68
+ 2023-10-17 13:35:36,190 - embedding storage: none
69
+ 2023-10-17 13:35:36,190 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 13:35:36,190 Model training base path: "hmbench-newseye/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
71
+ 2023-10-17 13:35:36,190 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 13:35:36,191 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 13:35:36,191 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 13:36:03,247 epoch 1 - iter 260/2606 - loss 2.23417308 - time (sec): 27.05 - samples/sec: 1321.16 - lr: 0.000003 - momentum: 0.000000
75
+ 2023-10-17 13:36:32,899 epoch 1 - iter 520/2606 - loss 1.31459848 - time (sec): 56.71 - samples/sec: 1299.74 - lr: 0.000006 - momentum: 0.000000
76
+ 2023-10-17 13:37:00,740 epoch 1 - iter 780/2606 - loss 0.98767282 - time (sec): 84.55 - samples/sec: 1322.73 - lr: 0.000009 - momentum: 0.000000
77
+ 2023-10-17 13:37:28,917 epoch 1 - iter 1040/2606 - loss 0.80515632 - time (sec): 112.72 - samples/sec: 1326.87 - lr: 0.000012 - momentum: 0.000000
78
+ 2023-10-17 13:37:57,443 epoch 1 - iter 1300/2606 - loss 0.69065513 - time (sec): 141.25 - samples/sec: 1325.36 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-17 13:38:24,254 epoch 1 - iter 1560/2606 - loss 0.62104172 - time (sec): 168.06 - samples/sec: 1325.26 - lr: 0.000018 - momentum: 0.000000
80
+ 2023-10-17 13:38:49,550 epoch 1 - iter 1820/2606 - loss 0.56057291 - time (sec): 193.36 - samples/sec: 1349.16 - lr: 0.000021 - momentum: 0.000000
81
+ 2023-10-17 13:39:15,981 epoch 1 - iter 2080/2606 - loss 0.51882473 - time (sec): 219.79 - samples/sec: 1343.75 - lr: 0.000024 - momentum: 0.000000
82
+ 2023-10-17 13:39:42,923 epoch 1 - iter 2340/2606 - loss 0.48259607 - time (sec): 246.73 - samples/sec: 1345.69 - lr: 0.000027 - momentum: 0.000000
83
+ 2023-10-17 13:40:08,974 epoch 1 - iter 2600/2606 - loss 0.45516666 - time (sec): 272.78 - samples/sec: 1344.72 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-17 13:40:09,555 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 13:40:09,555 EPOCH 1 done: loss 0.4546 - lr: 0.000030
86
+ 2023-10-17 13:40:17,382 DEV : loss 0.12860068678855896 - f1-score (micro avg) 0.2748
87
+ 2023-10-17 13:40:17,440 saving best model
88
+ 2023-10-17 13:40:17,985 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 13:40:45,255 epoch 2 - iter 260/2606 - loss 0.17952761 - time (sec): 27.27 - samples/sec: 1355.53 - lr: 0.000030 - momentum: 0.000000
90
+ 2023-10-17 13:41:12,418 epoch 2 - iter 520/2606 - loss 0.17895549 - time (sec): 54.43 - samples/sec: 1353.01 - lr: 0.000029 - momentum: 0.000000
91
+ 2023-10-17 13:41:39,526 epoch 2 - iter 780/2606 - loss 0.17220507 - time (sec): 81.54 - samples/sec: 1357.16 - lr: 0.000029 - momentum: 0.000000
92
+ 2023-10-17 13:42:06,127 epoch 2 - iter 1040/2606 - loss 0.16936553 - time (sec): 108.14 - samples/sec: 1359.56 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-17 13:42:34,312 epoch 2 - iter 1300/2606 - loss 0.16526069 - time (sec): 136.32 - samples/sec: 1362.58 - lr: 0.000028 - momentum: 0.000000
94
+ 2023-10-17 13:43:01,989 epoch 2 - iter 1560/2606 - loss 0.16534493 - time (sec): 164.00 - samples/sec: 1355.56 - lr: 0.000028 - momentum: 0.000000
95
+ 2023-10-17 13:43:28,081 epoch 2 - iter 1820/2606 - loss 0.16526124 - time (sec): 190.09 - samples/sec: 1355.72 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-17 13:43:55,714 epoch 2 - iter 2080/2606 - loss 0.15960605 - time (sec): 217.73 - samples/sec: 1351.07 - lr: 0.000027 - momentum: 0.000000
97
+ 2023-10-17 13:44:23,104 epoch 2 - iter 2340/2606 - loss 0.15681530 - time (sec): 245.12 - samples/sec: 1351.33 - lr: 0.000027 - momentum: 0.000000
98
+ 2023-10-17 13:44:50,177 epoch 2 - iter 2600/2606 - loss 0.15743046 - time (sec): 272.19 - samples/sec: 1347.84 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-17 13:44:50,695 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 13:44:50,696 EPOCH 2 done: loss 0.1573 - lr: 0.000027
101
+ 2023-10-17 13:45:02,783 DEV : loss 0.18409471213817596 - f1-score (micro avg) 0.2931
102
+ 2023-10-17 13:45:02,846 saving best model
103
+ 2023-10-17 13:45:04,273 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 13:45:31,452 epoch 3 - iter 260/2606 - loss 0.12993628 - time (sec): 27.17 - samples/sec: 1373.20 - lr: 0.000026 - momentum: 0.000000
105
+ 2023-10-17 13:45:56,995 epoch 3 - iter 520/2606 - loss 0.12197809 - time (sec): 52.72 - samples/sec: 1378.15 - lr: 0.000026 - momentum: 0.000000
106
+ 2023-10-17 13:46:23,269 epoch 3 - iter 780/2606 - loss 0.11338115 - time (sec): 78.99 - samples/sec: 1349.83 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-17 13:46:51,571 epoch 3 - iter 1040/2606 - loss 0.11312511 - time (sec): 107.29 - samples/sec: 1356.63 - lr: 0.000025 - momentum: 0.000000
108
+ 2023-10-17 13:47:19,624 epoch 3 - iter 1300/2606 - loss 0.11104534 - time (sec): 135.35 - samples/sec: 1356.87 - lr: 0.000025 - momentum: 0.000000
109
+ 2023-10-17 13:47:47,029 epoch 3 - iter 1560/2606 - loss 0.10839794 - time (sec): 162.75 - samples/sec: 1358.23 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-17 13:48:13,376 epoch 3 - iter 1820/2606 - loss 0.10823766 - time (sec): 189.10 - samples/sec: 1345.95 - lr: 0.000024 - momentum: 0.000000
111
+ 2023-10-17 13:48:41,227 epoch 3 - iter 2080/2606 - loss 0.10901249 - time (sec): 216.95 - samples/sec: 1347.31 - lr: 0.000024 - momentum: 0.000000
112
+ 2023-10-17 13:49:08,936 epoch 3 - iter 2340/2606 - loss 0.10998392 - time (sec): 244.66 - samples/sec: 1343.61 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-17 13:49:35,404 epoch 3 - iter 2600/2606 - loss 0.11017179 - time (sec): 271.13 - samples/sec: 1351.05 - lr: 0.000023 - momentum: 0.000000
114
+ 2023-10-17 13:49:36,143 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 13:49:36,143 EPOCH 3 done: loss 0.1099 - lr: 0.000023
116
+ 2023-10-17 13:49:47,073 DEV : loss 0.19609655439853668 - f1-score (micro avg) 0.3532
117
+ 2023-10-17 13:49:47,126 saving best model
118
+ 2023-10-17 13:49:48,541 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-17 13:50:16,836 epoch 4 - iter 260/2606 - loss 0.06308814 - time (sec): 28.29 - samples/sec: 1315.79 - lr: 0.000023 - momentum: 0.000000
120
+ 2023-10-17 13:50:44,010 epoch 4 - iter 520/2606 - loss 0.06791172 - time (sec): 55.46 - samples/sec: 1324.28 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-17 13:51:10,499 epoch 4 - iter 780/2606 - loss 0.06827024 - time (sec): 81.95 - samples/sec: 1332.80 - lr: 0.000022 - momentum: 0.000000
122
+ 2023-10-17 13:51:37,765 epoch 4 - iter 1040/2606 - loss 0.07212856 - time (sec): 109.22 - samples/sec: 1329.72 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-17 13:52:04,808 epoch 4 - iter 1300/2606 - loss 0.07317313 - time (sec): 136.26 - samples/sec: 1346.19 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-17 13:52:31,550 epoch 4 - iter 1560/2606 - loss 0.07332059 - time (sec): 163.00 - samples/sec: 1346.05 - lr: 0.000021 - momentum: 0.000000
125
+ 2023-10-17 13:52:59,143 epoch 4 - iter 1820/2606 - loss 0.07391824 - time (sec): 190.60 - samples/sec: 1345.38 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-17 13:53:26,459 epoch 4 - iter 2080/2606 - loss 0.07288315 - time (sec): 217.91 - samples/sec: 1337.06 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-17 13:53:54,604 epoch 4 - iter 2340/2606 - loss 0.07585692 - time (sec): 246.06 - samples/sec: 1341.65 - lr: 0.000020 - momentum: 0.000000
128
+ 2023-10-17 13:54:21,565 epoch 4 - iter 2600/2606 - loss 0.07713987 - time (sec): 273.02 - samples/sec: 1342.54 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-17 13:54:22,228 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-17 13:54:22,228 EPOCH 4 done: loss 0.0771 - lr: 0.000020
131
+ 2023-10-17 13:54:33,663 DEV : loss 0.30645275115966797 - f1-score (micro avg) 0.3598
132
+ 2023-10-17 13:54:33,734 saving best model
133
+ 2023-10-17 13:54:35,171 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-17 13:55:03,942 epoch 5 - iter 260/2606 - loss 0.04419626 - time (sec): 28.77 - samples/sec: 1307.72 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-17 13:55:32,835 epoch 5 - iter 520/2606 - loss 0.05530628 - time (sec): 57.66 - samples/sec: 1297.68 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-17 13:56:01,826 epoch 5 - iter 780/2606 - loss 0.05500235 - time (sec): 86.65 - samples/sec: 1315.68 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-17 13:56:30,166 epoch 5 - iter 1040/2606 - loss 0.05623530 - time (sec): 114.99 - samples/sec: 1325.25 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-17 13:56:57,009 epoch 5 - iter 1300/2606 - loss 0.05618477 - time (sec): 141.83 - samples/sec: 1328.11 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-17 13:57:24,218 epoch 5 - iter 1560/2606 - loss 0.05551545 - time (sec): 169.04 - samples/sec: 1335.73 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-17 13:57:50,677 epoch 5 - iter 1820/2606 - loss 0.05518345 - time (sec): 195.50 - samples/sec: 1342.10 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-17 13:58:16,808 epoch 5 - iter 2080/2606 - loss 0.05536442 - time (sec): 221.63 - samples/sec: 1340.42 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-17 13:58:43,597 epoch 5 - iter 2340/2606 - loss 0.05513621 - time (sec): 248.42 - samples/sec: 1338.91 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-17 13:59:10,177 epoch 5 - iter 2600/2606 - loss 0.05550176 - time (sec): 275.00 - samples/sec: 1333.24 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-17 13:59:10,813 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-17 13:59:10,813 EPOCH 5 done: loss 0.0556 - lr: 0.000017
146
+ 2023-10-17 13:59:22,685 DEV : loss 0.38089314103126526 - f1-score (micro avg) 0.3689
147
+ 2023-10-17 13:59:22,758 saving best model
148
+ 2023-10-17 13:59:24,340 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-17 13:59:51,579 epoch 6 - iter 260/2606 - loss 0.03634504 - time (sec): 27.24 - samples/sec: 1312.08 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-17 14:00:19,359 epoch 6 - iter 520/2606 - loss 0.03563055 - time (sec): 55.02 - samples/sec: 1327.10 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-17 14:00:45,525 epoch 6 - iter 780/2606 - loss 0.03541167 - time (sec): 81.18 - samples/sec: 1310.80 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-17 14:01:13,098 epoch 6 - iter 1040/2606 - loss 0.03662892 - time (sec): 108.75 - samples/sec: 1311.16 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-17 14:01:41,840 epoch 6 - iter 1300/2606 - loss 0.03651950 - time (sec): 137.50 - samples/sec: 1316.73 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-17 14:02:09,866 epoch 6 - iter 1560/2606 - loss 0.03685032 - time (sec): 165.52 - samples/sec: 1311.43 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-17 14:02:36,853 epoch 6 - iter 1820/2606 - loss 0.03810930 - time (sec): 192.51 - samples/sec: 1319.49 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-17 14:03:04,418 epoch 6 - iter 2080/2606 - loss 0.03893618 - time (sec): 220.07 - samples/sec: 1331.23 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-17 14:03:30,680 epoch 6 - iter 2340/2606 - loss 0.03981572 - time (sec): 246.34 - samples/sec: 1332.99 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-17 14:03:58,240 epoch 6 - iter 2600/2606 - loss 0.03973320 - time (sec): 273.90 - samples/sec: 1337.44 - lr: 0.000013 - momentum: 0.000000
159
+ 2023-10-17 14:03:59,011 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-17 14:03:59,011 EPOCH 6 done: loss 0.0397 - lr: 0.000013
161
+ 2023-10-17 14:04:10,365 DEV : loss 0.3598186671733856 - f1-score (micro avg) 0.3703
162
+ 2023-10-17 14:04:10,425 saving best model
163
+ 2023-10-17 14:04:11,836 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-17 14:04:38,872 epoch 7 - iter 260/2606 - loss 0.02914898 - time (sec): 27.03 - samples/sec: 1354.67 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-17 14:05:05,820 epoch 7 - iter 520/2606 - loss 0.02552344 - time (sec): 53.98 - samples/sec: 1376.27 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-17 14:05:31,950 epoch 7 - iter 780/2606 - loss 0.02775229 - time (sec): 80.11 - samples/sec: 1363.39 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-17 14:06:00,317 epoch 7 - iter 1040/2606 - loss 0.02635722 - time (sec): 108.48 - samples/sec: 1364.31 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-17 14:06:28,215 epoch 7 - iter 1300/2606 - loss 0.02810019 - time (sec): 136.37 - samples/sec: 1367.56 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-17 14:06:55,868 epoch 7 - iter 1560/2606 - loss 0.02775623 - time (sec): 164.03 - samples/sec: 1361.77 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-17 14:07:24,511 epoch 7 - iter 1820/2606 - loss 0.02725007 - time (sec): 192.67 - samples/sec: 1356.57 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-17 14:07:52,475 epoch 7 - iter 2080/2606 - loss 0.02794776 - time (sec): 220.64 - samples/sec: 1351.16 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-17 14:08:18,907 epoch 7 - iter 2340/2606 - loss 0.02788600 - time (sec): 247.07 - samples/sec: 1343.37 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-17 14:08:45,921 epoch 7 - iter 2600/2606 - loss 0.02768411 - time (sec): 274.08 - samples/sec: 1339.04 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-17 14:08:46,550 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-17 14:08:46,550 EPOCH 7 done: loss 0.0277 - lr: 0.000010
176
+ 2023-10-17 14:08:58,244 DEV : loss 0.46519169211387634 - f1-score (micro avg) 0.3683
177
+ 2023-10-17 14:08:58,320 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-17 14:09:25,709 epoch 8 - iter 260/2606 - loss 0.01354824 - time (sec): 27.39 - samples/sec: 1270.70 - lr: 0.000010 - momentum: 0.000000
179
+ 2023-10-17 14:09:52,894 epoch 8 - iter 520/2606 - loss 0.01746483 - time (sec): 54.57 - samples/sec: 1295.91 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-17 14:10:20,826 epoch 8 - iter 780/2606 - loss 0.01722493 - time (sec): 82.50 - samples/sec: 1322.69 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-17 14:10:48,150 epoch 8 - iter 1040/2606 - loss 0.01799795 - time (sec): 109.83 - samples/sec: 1318.28 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-17 14:11:14,849 epoch 8 - iter 1300/2606 - loss 0.02034286 - time (sec): 136.53 - samples/sec: 1321.37 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-17 14:11:41,434 epoch 8 - iter 1560/2606 - loss 0.02094045 - time (sec): 163.11 - samples/sec: 1329.60 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-17 14:12:08,680 epoch 8 - iter 1820/2606 - loss 0.02116218 - time (sec): 190.36 - samples/sec: 1339.00 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-17 14:12:35,381 epoch 8 - iter 2080/2606 - loss 0.02089064 - time (sec): 217.06 - samples/sec: 1342.79 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-17 14:13:02,686 epoch 8 - iter 2340/2606 - loss 0.02211357 - time (sec): 244.36 - samples/sec: 1347.86 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-17 14:13:28,922 epoch 8 - iter 2600/2606 - loss 0.02165998 - time (sec): 270.60 - samples/sec: 1354.06 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-17 14:13:29,518 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-17 14:13:29,519 EPOCH 8 done: loss 0.0217 - lr: 0.000007
190
+ 2023-10-17 14:13:41,241 DEV : loss 0.4531383812427521 - f1-score (micro avg) 0.3797
191
+ 2023-10-17 14:13:41,306 saving best model
192
+ 2023-10-17 14:13:42,719 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-17 14:14:10,088 epoch 9 - iter 260/2606 - loss 0.01374020 - time (sec): 27.36 - samples/sec: 1343.84 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-17 14:14:37,237 epoch 9 - iter 520/2606 - loss 0.01399606 - time (sec): 54.51 - samples/sec: 1373.81 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-17 14:15:04,557 epoch 9 - iter 780/2606 - loss 0.01498264 - time (sec): 81.83 - samples/sec: 1362.74 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-17 14:15:30,917 epoch 9 - iter 1040/2606 - loss 0.01450304 - time (sec): 108.19 - samples/sec: 1363.37 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-17 14:15:57,984 epoch 9 - iter 1300/2606 - loss 0.01479871 - time (sec): 135.26 - samples/sec: 1366.94 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-17 14:16:25,970 epoch 9 - iter 1560/2606 - loss 0.01476623 - time (sec): 163.25 - samples/sec: 1360.61 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-17 14:16:53,168 epoch 9 - iter 1820/2606 - loss 0.01440257 - time (sec): 190.44 - samples/sec: 1345.53 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-17 14:17:22,071 epoch 9 - iter 2080/2606 - loss 0.01471634 - time (sec): 219.35 - samples/sec: 1356.71 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-17 14:17:48,025 epoch 9 - iter 2340/2606 - loss 0.01462050 - time (sec): 245.30 - samples/sec: 1347.69 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-17 14:18:15,810 epoch 9 - iter 2600/2606 - loss 0.01420519 - time (sec): 273.09 - samples/sec: 1343.04 - lr: 0.000003 - momentum: 0.000000
203
+ 2023-10-17 14:18:16,365 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-17 14:18:16,366 EPOCH 9 done: loss 0.0142 - lr: 0.000003
205
+ 2023-10-17 14:18:29,180 DEV : loss 0.5335880517959595 - f1-score (micro avg) 0.3697
206
+ 2023-10-17 14:18:29,239 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-17 14:18:57,797 epoch 10 - iter 260/2606 - loss 0.00788040 - time (sec): 28.56 - samples/sec: 1317.69 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-17 14:19:25,473 epoch 10 - iter 520/2606 - loss 0.00890575 - time (sec): 56.23 - samples/sec: 1311.63 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-17 14:19:53,273 epoch 10 - iter 780/2606 - loss 0.00939444 - time (sec): 84.03 - samples/sec: 1292.56 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-17 14:20:21,086 epoch 10 - iter 1040/2606 - loss 0.01008456 - time (sec): 111.84 - samples/sec: 1280.70 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-17 14:20:50,402 epoch 10 - iter 1300/2606 - loss 0.01075747 - time (sec): 141.16 - samples/sec: 1271.11 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-17 14:21:17,606 epoch 10 - iter 1560/2606 - loss 0.01058155 - time (sec): 168.36 - samples/sec: 1273.33 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-17 14:21:45,269 epoch 10 - iter 1820/2606 - loss 0.01068539 - time (sec): 196.03 - samples/sec: 1275.73 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-17 14:22:13,609 epoch 10 - iter 2080/2606 - loss 0.01111151 - time (sec): 224.37 - samples/sec: 1285.50 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-17 14:22:42,423 epoch 10 - iter 2340/2606 - loss 0.01084573 - time (sec): 253.18 - samples/sec: 1298.87 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-17 14:23:09,546 epoch 10 - iter 2600/2606 - loss 0.01098140 - time (sec): 280.30 - samples/sec: 1308.85 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-17 14:23:10,109 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-17 14:23:10,109 EPOCH 10 done: loss 0.0110 - lr: 0.000000
219
+ 2023-10-17 14:23:22,465 DEV : loss 0.536491334438324 - f1-score (micro avg) 0.3636
220
+ 2023-10-17 14:23:23,103 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-17 14:23:23,105 Loading model from best epoch ...
222
+ 2023-10-17 14:23:25,695 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
+ 2023-10-17 14:23:45,508
224
+ Results:
225
+ - F-score (micro) 0.4513
226
+ - F-score (macro) 0.3196
227
+ - Accuracy 0.2959
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ LOC 0.4768 0.5074 0.4916 1214
233
+ PER 0.4066 0.5062 0.4509 808
234
+ ORG 0.3068 0.3711 0.3359 353
235
+ HumanProd 0.0000 0.0000 0.0000 15
236
+
237
+ micro avg 0.4230 0.4837 0.4513 2390
238
+ macro avg 0.2975 0.3462 0.3196 2390
239
+ weighted avg 0.4249 0.4837 0.4518 2390
240
+
241
+ 2023-10-17 14:23:45,509 ----------------------------------------------------------------------------------------------------