Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmteams-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
c895f5e
·
1 Parent(s): d39219f

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.1697564302.3ae7c61396a7.1160.8 +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 sha256:95e8ba7197a5f4133640e87ded77f989c4689515817f5538fcb0752914941876
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 17:43:03 0.0000 0.4589 0.1235 0.2144 0.5076 0.3015 0.1777
3
+ 2 17:47:48 0.0000 0.1533 0.1629 0.2322 0.6420 0.3410 0.2062
4
+ 3 17:52:31 0.0000 0.1085 0.2198 0.2408 0.6231 0.3474 0.2116
5
+ 4 17:57:21 0.0000 0.0813 0.2744 0.3187 0.5492 0.4033 0.2539
6
+ 5 18:02:12 0.0000 0.0546 0.2997 0.2917 0.5795 0.3881 0.2425
7
+ 6 18:07:01 0.0000 0.0389 0.3478 0.3020 0.6080 0.4035 0.2542
8
+ 7 18:11:52 0.0000 0.0285 0.4449 0.2785 0.5966 0.3797 0.2353
9
+ 8 18:16:40 0.0000 0.0211 0.5216 0.2629 0.6288 0.3707 0.2285
10
+ 9 18:21:25 0.0000 0.0135 0.5246 0.2742 0.6117 0.3787 0.2344
11
+ 10 18:26:14 0.0000 0.0107 0.5315 0.2765 0.6080 0.3801 0.2355
runs/events.out.tfevents.1697564302.3ae7c61396a7.1160.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:004ec79be9de67dbeeac171cb76c287a2406d754dde024c3e161fae3dadb9047
3
+ size 1464420
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 17:38:22,631 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 17:38:22,633 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 17:38:22,633 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 17:38:22,633 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 17:38:22,633 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 17:38:22,633 Train: 20847 sentences
51
+ 2023-10-17 17:38:22,634 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 17:38:22,634 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 17:38:22,634 Training Params:
54
+ 2023-10-17 17:38:22,634 - learning_rate: "3e-05"
55
+ 2023-10-17 17:38:22,634 - mini_batch_size: "8"
56
+ 2023-10-17 17:38:22,634 - max_epochs: "10"
57
+ 2023-10-17 17:38:22,634 - shuffle: "True"
58
+ 2023-10-17 17:38:22,634 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 17:38:22,634 Plugins:
60
+ 2023-10-17 17:38:22,634 - TensorboardLogger
61
+ 2023-10-17 17:38:22,634 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 17:38:22,634 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 17:38:22,634 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 17:38:22,634 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 17:38:22,634 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 17:38:22,635 Computation:
67
+ 2023-10-17 17:38:22,635 - compute on device: cuda:0
68
+ 2023-10-17 17:38:22,635 - embedding storage: none
69
+ 2023-10-17 17:38:22,635 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 17:38:22,635 Model training base path: "hmbench-newseye/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
71
+ 2023-10-17 17:38:22,635 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 17:38:22,635 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 17:38:22,635 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 17:38:49,992 epoch 1 - iter 260/2606 - loss 2.16108750 - time (sec): 27.36 - samples/sec: 1370.93 - lr: 0.000003 - momentum: 0.000000
75
+ 2023-10-17 17:39:16,294 epoch 1 - iter 520/2606 - loss 1.30692459 - time (sec): 53.66 - samples/sec: 1390.98 - lr: 0.000006 - momentum: 0.000000
76
+ 2023-10-17 17:39:43,218 epoch 1 - iter 780/2606 - loss 0.99754433 - time (sec): 80.58 - samples/sec: 1367.60 - lr: 0.000009 - momentum: 0.000000
77
+ 2023-10-17 17:40:10,453 epoch 1 - iter 1040/2606 - loss 0.82453976 - time (sec): 107.82 - samples/sec: 1349.01 - lr: 0.000012 - momentum: 0.000000
78
+ 2023-10-17 17:40:36,352 epoch 1 - iter 1300/2606 - loss 0.71369382 - time (sec): 133.72 - samples/sec: 1341.33 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-17 17:41:03,213 epoch 1 - iter 1560/2606 - loss 0.62993344 - time (sec): 160.58 - samples/sec: 1349.33 - lr: 0.000018 - momentum: 0.000000
80
+ 2023-10-17 17:41:31,274 epoch 1 - iter 1820/2606 - loss 0.57154390 - time (sec): 188.64 - samples/sec: 1337.61 - lr: 0.000021 - momentum: 0.000000
81
+ 2023-10-17 17:41:58,975 epoch 1 - iter 2080/2606 - loss 0.52820388 - time (sec): 216.34 - samples/sec: 1339.02 - lr: 0.000024 - momentum: 0.000000
82
+ 2023-10-17 17:42:25,833 epoch 1 - iter 2340/2606 - loss 0.49230602 - time (sec): 243.20 - samples/sec: 1335.50 - lr: 0.000027 - momentum: 0.000000
83
+ 2023-10-17 17:42:54,842 epoch 1 - iter 2600/2606 - loss 0.45946927 - time (sec): 272.20 - samples/sec: 1347.07 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-17 17:42:55,398 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 17:42:55,399 EPOCH 1 done: loss 0.4589 - lr: 0.000030
86
+ 2023-10-17 17:43:03,287 DEV : loss 0.12348709255456924 - f1-score (micro avg) 0.3015
87
+ 2023-10-17 17:43:03,342 saving best model
88
+ 2023-10-17 17:43:03,884 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 17:43:31,177 epoch 2 - iter 260/2606 - loss 0.17303029 - time (sec): 27.29 - samples/sec: 1347.32 - lr: 0.000030 - momentum: 0.000000
90
+ 2023-10-17 17:43:59,525 epoch 2 - iter 520/2606 - loss 0.16886574 - time (sec): 55.64 - samples/sec: 1373.29 - lr: 0.000029 - momentum: 0.000000
91
+ 2023-10-17 17:44:26,379 epoch 2 - iter 780/2606 - loss 0.16483756 - time (sec): 82.49 - samples/sec: 1359.54 - lr: 0.000029 - momentum: 0.000000
92
+ 2023-10-17 17:44:53,079 epoch 2 - iter 1040/2606 - loss 0.16595272 - time (sec): 109.19 - samples/sec: 1354.15 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-17 17:45:19,612 epoch 2 - iter 1300/2606 - loss 0.16465462 - time (sec): 135.73 - samples/sec: 1346.09 - lr: 0.000028 - momentum: 0.000000
94
+ 2023-10-17 17:45:46,866 epoch 2 - iter 1560/2606 - loss 0.16186852 - time (sec): 162.98 - samples/sec: 1348.08 - lr: 0.000028 - momentum: 0.000000
95
+ 2023-10-17 17:46:14,295 epoch 2 - iter 1820/2606 - loss 0.16153027 - time (sec): 190.41 - samples/sec: 1335.51 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-17 17:46:41,254 epoch 2 - iter 2080/2606 - loss 0.15817180 - time (sec): 217.37 - samples/sec: 1341.88 - lr: 0.000027 - momentum: 0.000000
97
+ 2023-10-17 17:47:07,690 epoch 2 - iter 2340/2606 - loss 0.15541969 - time (sec): 243.80 - samples/sec: 1349.13 - lr: 0.000027 - momentum: 0.000000
98
+ 2023-10-17 17:47:35,914 epoch 2 - iter 2600/2606 - loss 0.15324901 - time (sec): 272.03 - samples/sec: 1348.38 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-17 17:47:36,492 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 17:47:36,492 EPOCH 2 done: loss 0.1533 - lr: 0.000027
101
+ 2023-10-17 17:47:48,737 DEV : loss 0.16292980313301086 - f1-score (micro avg) 0.341
102
+ 2023-10-17 17:47:48,800 saving best model
103
+ 2023-10-17 17:47:50,156 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 17:48:16,842 epoch 3 - iter 260/2606 - loss 0.11265805 - time (sec): 26.68 - samples/sec: 1334.85 - lr: 0.000026 - momentum: 0.000000
105
+ 2023-10-17 17:48:44,542 epoch 3 - iter 520/2606 - loss 0.10845921 - time (sec): 54.38 - samples/sec: 1343.73 - lr: 0.000026 - momentum: 0.000000
106
+ 2023-10-17 17:49:11,468 epoch 3 - iter 780/2606 - loss 0.10959502 - time (sec): 81.31 - samples/sec: 1348.75 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-17 17:49:37,863 epoch 3 - iter 1040/2606 - loss 0.11240173 - time (sec): 107.70 - samples/sec: 1348.81 - lr: 0.000025 - momentum: 0.000000
108
+ 2023-10-17 17:50:04,462 epoch 3 - iter 1300/2606 - loss 0.11114369 - time (sec): 134.30 - samples/sec: 1352.04 - lr: 0.000025 - momentum: 0.000000
109
+ 2023-10-17 17:50:32,138 epoch 3 - iter 1560/2606 - loss 0.10929810 - time (sec): 161.98 - samples/sec: 1364.81 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-17 17:50:57,682 epoch 3 - iter 1820/2606 - loss 0.11045895 - time (sec): 187.52 - samples/sec: 1371.97 - lr: 0.000024 - momentum: 0.000000
111
+ 2023-10-17 17:51:24,187 epoch 3 - iter 2080/2606 - loss 0.11033176 - time (sec): 214.03 - samples/sec: 1378.25 - lr: 0.000024 - momentum: 0.000000
112
+ 2023-10-17 17:51:49,493 epoch 3 - iter 2340/2606 - loss 0.10971803 - time (sec): 239.33 - samples/sec: 1370.69 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-17 17:52:17,879 epoch 3 - iter 2600/2606 - loss 0.10861513 - time (sec): 267.72 - samples/sec: 1370.17 - lr: 0.000023 - momentum: 0.000000
114
+ 2023-10-17 17:52:18,429 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 17:52:18,429 EPOCH 3 done: loss 0.1085 - lr: 0.000023
116
+ 2023-10-17 17:52:31,403 DEV : loss 0.21983903646469116 - f1-score (micro avg) 0.3474
117
+ 2023-10-17 17:52:31,463 saving best model
118
+ 2023-10-17 17:52:32,872 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-17 17:53:00,390 epoch 4 - iter 260/2606 - loss 0.07713242 - time (sec): 27.51 - samples/sec: 1346.41 - lr: 0.000023 - momentum: 0.000000
120
+ 2023-10-17 17:53:28,664 epoch 4 - iter 520/2606 - loss 0.08561030 - time (sec): 55.79 - samples/sec: 1351.93 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-17 17:53:56,211 epoch 4 - iter 780/2606 - loss 0.08498077 - time (sec): 83.33 - samples/sec: 1326.24 - lr: 0.000022 - momentum: 0.000000
122
+ 2023-10-17 17:54:22,655 epoch 4 - iter 1040/2606 - loss 0.08337658 - time (sec): 109.78 - samples/sec: 1336.06 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-17 17:54:49,664 epoch 4 - iter 1300/2606 - loss 0.08143883 - time (sec): 136.79 - samples/sec: 1339.71 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-17 17:55:16,666 epoch 4 - iter 1560/2606 - loss 0.08262716 - time (sec): 163.79 - samples/sec: 1336.65 - lr: 0.000021 - momentum: 0.000000
125
+ 2023-10-17 17:55:44,033 epoch 4 - iter 1820/2606 - loss 0.08312524 - time (sec): 191.16 - samples/sec: 1336.14 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-17 17:56:13,210 epoch 4 - iter 2080/2606 - loss 0.08305110 - time (sec): 220.33 - samples/sec: 1332.00 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-17 17:56:40,320 epoch 4 - iter 2340/2606 - loss 0.08202034 - time (sec): 247.44 - samples/sec: 1329.58 - lr: 0.000020 - momentum: 0.000000
128
+ 2023-10-17 17:57:08,440 epoch 4 - iter 2600/2606 - loss 0.08136397 - time (sec): 275.56 - samples/sec: 1330.33 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-17 17:57:09,019 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-17 17:57:09,019 EPOCH 4 done: loss 0.0813 - lr: 0.000020
131
+ 2023-10-17 17:57:21,370 DEV : loss 0.27437207102775574 - f1-score (micro avg) 0.4033
132
+ 2023-10-17 17:57:21,425 saving best model
133
+ 2023-10-17 17:57:22,832 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-17 17:57:50,404 epoch 5 - iter 260/2606 - loss 0.05833828 - time (sec): 27.57 - samples/sec: 1316.84 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-17 17:58:17,100 epoch 5 - iter 520/2606 - loss 0.05838581 - time (sec): 54.26 - samples/sec: 1338.09 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-17 17:58:45,873 epoch 5 - iter 780/2606 - loss 0.05325511 - time (sec): 83.04 - samples/sec: 1335.64 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-17 17:59:14,285 epoch 5 - iter 1040/2606 - loss 0.05210183 - time (sec): 111.45 - samples/sec: 1331.46 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-17 17:59:42,645 epoch 5 - iter 1300/2606 - loss 0.05559593 - time (sec): 139.81 - samples/sec: 1323.08 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-17 18:00:10,380 epoch 5 - iter 1560/2606 - loss 0.05487736 - time (sec): 167.54 - samples/sec: 1329.02 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-17 18:00:37,660 epoch 5 - iter 1820/2606 - loss 0.05506955 - time (sec): 194.82 - samples/sec: 1328.17 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-17 18:01:04,469 epoch 5 - iter 2080/2606 - loss 0.05519986 - time (sec): 221.63 - samples/sec: 1324.06 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-17 18:01:32,780 epoch 5 - iter 2340/2606 - loss 0.05451422 - time (sec): 249.94 - samples/sec: 1326.95 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-17 18:01:59,316 epoch 5 - iter 2600/2606 - loss 0.05460912 - time (sec): 276.48 - samples/sec: 1326.47 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-17 18:01:59,877 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-17 18:01:59,877 EPOCH 5 done: loss 0.0546 - lr: 0.000017
146
+ 2023-10-17 18:02:12,437 DEV : loss 0.2996568977832794 - f1-score (micro avg) 0.3881
147
+ 2023-10-17 18:02:12,493 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-17 18:02:40,068 epoch 6 - iter 260/2606 - loss 0.03423666 - time (sec): 27.57 - samples/sec: 1364.83 - lr: 0.000016 - momentum: 0.000000
149
+ 2023-10-17 18:03:06,656 epoch 6 - iter 520/2606 - loss 0.03348953 - time (sec): 54.16 - samples/sec: 1321.35 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-17 18:03:33,792 epoch 6 - iter 780/2606 - loss 0.03525955 - time (sec): 81.30 - samples/sec: 1315.45 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-17 18:04:00,415 epoch 6 - iter 1040/2606 - loss 0.03884942 - time (sec): 107.92 - samples/sec: 1313.70 - lr: 0.000015 - momentum: 0.000000
152
+ 2023-10-17 18:04:27,780 epoch 6 - iter 1300/2606 - loss 0.03894248 - time (sec): 135.28 - samples/sec: 1310.85 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-17 18:04:55,792 epoch 6 - iter 1560/2606 - loss 0.03858817 - time (sec): 163.30 - samples/sec: 1314.79 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-17 18:05:23,682 epoch 6 - iter 1820/2606 - loss 0.03919379 - time (sec): 191.19 - samples/sec: 1315.67 - lr: 0.000014 - momentum: 0.000000
155
+ 2023-10-17 18:05:53,070 epoch 6 - iter 2080/2606 - loss 0.03898069 - time (sec): 220.57 - samples/sec: 1320.80 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-17 18:06:21,530 epoch 6 - iter 2340/2606 - loss 0.03869181 - time (sec): 249.03 - samples/sec: 1332.24 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-17 18:06:48,297 epoch 6 - iter 2600/2606 - loss 0.03893522 - time (sec): 275.80 - samples/sec: 1328.89 - lr: 0.000013 - momentum: 0.000000
158
+ 2023-10-17 18:06:49,003 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-17 18:06:49,003 EPOCH 6 done: loss 0.0389 - lr: 0.000013
160
+ 2023-10-17 18:07:01,825 DEV : loss 0.34783273935317993 - f1-score (micro avg) 0.4035
161
+ 2023-10-17 18:07:01,879 saving best model
162
+ 2023-10-17 18:07:03,283 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-17 18:07:31,404 epoch 7 - iter 260/2606 - loss 0.02894101 - time (sec): 28.12 - samples/sec: 1363.81 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-17 18:07:58,656 epoch 7 - iter 520/2606 - loss 0.02724580 - time (sec): 55.37 - samples/sec: 1352.93 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-17 18:08:27,592 epoch 7 - iter 780/2606 - loss 0.03067550 - time (sec): 84.30 - samples/sec: 1333.06 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-17 18:08:54,620 epoch 7 - iter 1040/2606 - loss 0.02932069 - time (sec): 111.33 - samples/sec: 1319.87 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-17 18:09:21,942 epoch 7 - iter 1300/2606 - loss 0.02883956 - time (sec): 138.65 - samples/sec: 1322.62 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-17 18:09:49,932 epoch 7 - iter 1560/2606 - loss 0.02806752 - time (sec): 166.65 - samples/sec: 1331.28 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-17 18:10:16,796 epoch 7 - iter 1820/2606 - loss 0.02809168 - time (sec): 193.51 - samples/sec: 1329.05 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-17 18:10:44,355 epoch 7 - iter 2080/2606 - loss 0.02817759 - time (sec): 221.07 - samples/sec: 1321.32 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-17 18:11:12,175 epoch 7 - iter 2340/2606 - loss 0.02835946 - time (sec): 248.89 - samples/sec: 1318.43 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-17 18:11:40,537 epoch 7 - iter 2600/2606 - loss 0.02853150 - time (sec): 277.25 - samples/sec: 1320.98 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-17 18:11:41,332 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-17 18:11:41,332 EPOCH 7 done: loss 0.0285 - lr: 0.000010
175
+ 2023-10-17 18:11:52,778 DEV : loss 0.4449138045310974 - f1-score (micro avg) 0.3797
176
+ 2023-10-17 18:11:52,843 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-17 18:12:20,333 epoch 8 - iter 260/2606 - loss 0.02355353 - time (sec): 27.49 - samples/sec: 1308.83 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-17 18:12:48,478 epoch 8 - iter 520/2606 - loss 0.02242287 - time (sec): 55.63 - samples/sec: 1301.33 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-17 18:13:17,210 epoch 8 - iter 780/2606 - loss 0.02142082 - time (sec): 84.36 - samples/sec: 1302.22 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-17 18:13:44,116 epoch 8 - iter 1040/2606 - loss 0.02198135 - time (sec): 111.27 - samples/sec: 1318.24 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-17 18:14:11,882 epoch 8 - iter 1300/2606 - loss 0.02245041 - time (sec): 139.04 - samples/sec: 1320.33 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-17 18:14:38,817 epoch 8 - iter 1560/2606 - loss 0.02235451 - time (sec): 165.97 - samples/sec: 1320.02 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-17 18:15:06,477 epoch 8 - iter 1820/2606 - loss 0.02124624 - time (sec): 193.63 - samples/sec: 1321.69 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-17 18:15:33,535 epoch 8 - iter 2080/2606 - loss 0.02100098 - time (sec): 220.69 - samples/sec: 1321.00 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-17 18:16:00,553 epoch 8 - iter 2340/2606 - loss 0.02124546 - time (sec): 247.71 - samples/sec: 1323.91 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-17 18:16:29,052 epoch 8 - iter 2600/2606 - loss 0.02098403 - time (sec): 276.21 - samples/sec: 1327.93 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-17 18:16:29,596 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-17 18:16:29,597 EPOCH 8 done: loss 0.0211 - lr: 0.000007
189
+ 2023-10-17 18:16:40,893 DEV : loss 0.5216322541236877 - f1-score (micro avg) 0.3707
190
+ 2023-10-17 18:16:40,956 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-17 18:17:09,504 epoch 9 - iter 260/2606 - loss 0.01271966 - time (sec): 28.55 - samples/sec: 1427.43 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-17 18:17:36,981 epoch 9 - iter 520/2606 - loss 0.01328728 - time (sec): 56.02 - samples/sec: 1396.29 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-17 18:18:04,506 epoch 9 - iter 780/2606 - loss 0.01441554 - time (sec): 83.55 - samples/sec: 1357.80 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-17 18:18:32,793 epoch 9 - iter 1040/2606 - loss 0.01396545 - time (sec): 111.83 - samples/sec: 1344.18 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-17 18:18:59,609 epoch 9 - iter 1300/2606 - loss 0.01392302 - time (sec): 138.65 - samples/sec: 1349.91 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-17 18:19:26,372 epoch 9 - iter 1560/2606 - loss 0.01326058 - time (sec): 165.41 - samples/sec: 1349.25 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-17 18:19:53,465 epoch 9 - iter 1820/2606 - loss 0.01302468 - time (sec): 192.51 - samples/sec: 1344.95 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-17 18:20:21,078 epoch 9 - iter 2080/2606 - loss 0.01339090 - time (sec): 220.12 - samples/sec: 1348.97 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-17 18:20:48,720 epoch 9 - iter 2340/2606 - loss 0.01350316 - time (sec): 247.76 - samples/sec: 1349.79 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-17 18:21:13,941 epoch 9 - iter 2600/2606 - loss 0.01339440 - time (sec): 272.98 - samples/sec: 1343.38 - lr: 0.000003 - momentum: 0.000000
201
+ 2023-10-17 18:21:14,446 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-17 18:21:14,446 EPOCH 9 done: loss 0.0135 - lr: 0.000003
203
+ 2023-10-17 18:21:25,731 DEV : loss 0.5246009230613708 - f1-score (micro avg) 0.3787
204
+ 2023-10-17 18:21:25,786 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-17 18:21:53,457 epoch 10 - iter 260/2606 - loss 0.01018569 - time (sec): 27.67 - samples/sec: 1373.20 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-17 18:22:22,398 epoch 10 - iter 520/2606 - loss 0.01216228 - time (sec): 56.61 - samples/sec: 1368.55 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-17 18:22:48,583 epoch 10 - iter 780/2606 - loss 0.01210997 - time (sec): 82.79 - samples/sec: 1337.18 - lr: 0.000002 - momentum: 0.000000
208
+ 2023-10-17 18:23:16,814 epoch 10 - iter 1040/2606 - loss 0.01155669 - time (sec): 111.03 - samples/sec: 1338.09 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-17 18:23:44,121 epoch 10 - iter 1300/2606 - loss 0.01112163 - time (sec): 138.33 - samples/sec: 1322.20 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-17 18:24:11,721 epoch 10 - iter 1560/2606 - loss 0.01128677 - time (sec): 165.93 - samples/sec: 1324.56 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-17 18:24:41,182 epoch 10 - iter 1820/2606 - loss 0.01108459 - time (sec): 195.39 - samples/sec: 1327.32 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-17 18:25:08,860 epoch 10 - iter 2080/2606 - loss 0.01112871 - time (sec): 223.07 - samples/sec: 1326.85 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-17 18:25:35,011 epoch 10 - iter 2340/2606 - loss 0.01122759 - time (sec): 249.22 - samples/sec: 1322.79 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-17 18:26:02,919 epoch 10 - iter 2600/2606 - loss 0.01076733 - time (sec): 277.13 - samples/sec: 1322.70 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-17 18:26:03,521 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-17 18:26:03,521 EPOCH 10 done: loss 0.0107 - lr: 0.000000
217
+ 2023-10-17 18:26:14,910 DEV : loss 0.5314543843269348 - f1-score (micro avg) 0.3801
218
+ 2023-10-17 18:26:15,512 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-17 18:26:15,514 Loading model from best epoch ...
220
+ 2023-10-17 18:26:18,076 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
+ 2023-10-17 18:26:38,263
222
+ Results:
223
+ - F-score (micro) 0.4345
224
+ - F-score (macro) 0.3022
225
+ - Accuracy 0.282
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.4256 0.5255 0.4703 1214
231
+ PER 0.4325 0.4480 0.4401 808
232
+ ORG 0.2940 0.3031 0.2985 353
233
+ HumanProd 0.0000 0.0000 0.0000 15
234
+
235
+ micro avg 0.4091 0.4632 0.4345 2390
236
+ macro avg 0.2880 0.3192 0.3022 2390
237
+ weighted avg 0.4058 0.4632 0.4318 2390
238
+
239
+ 2023-10-17 18:26:38,263 ----------------------------------------------------------------------------------------------------