Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
acd5c36
•
1 Parent(s): ea424d0

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. test.tsv +0 -0
  5. training.log +238 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:964d2d85b92f0ffacfd062af384dcd6a92921848715b8d9a8faeb34c091db904
3
+ size 443323527
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 13:20:36 0.0000 0.3725 0.1153 0.2705 0.2254 0.2459 0.1407
3
+ 2 13:25:12 0.0000 0.1931 0.1754 0.3014 0.4640 0.3654 0.2246
4
+ 3 13:29:49 0.0000 0.1431 0.2379 0.2153 0.5436 0.3084 0.1842
5
+ 4 13:34:09 0.0000 0.1053 0.2393 0.2190 0.4716 0.2991 0.1767
6
+ 5 13:38:33 0.0000 0.0800 0.3472 0.2095 0.5758 0.3072 0.1825
7
+ 6 13:42:56 0.0000 0.0566 0.3655 0.2575 0.5530 0.3514 0.2141
8
+ 7 13:47:17 0.0000 0.0440 0.3459 0.2369 0.5492 0.3311 0.1996
9
+ 8 13:51:42 0.0000 0.0300 0.4303 0.2463 0.6004 0.3493 0.2128
10
+ 9 13:56:05 0.0000 0.0205 0.5123 0.2384 0.5814 0.3381 0.2045
11
+ 10 14:00:24 0.0000 0.0138 0.4650 0.2532 0.5663 0.3499 0.2133
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,238 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 13:16:10,272 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 13:16:10,273 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
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): 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
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 13:16:10,273 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
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 13:16:10,273 Train: 20847 sentences
55
+ 2023-10-15 13:16:10,273 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 13:16:10,273 Training Params:
58
+ 2023-10-15 13:16:10,273 - learning_rate: "5e-05"
59
+ 2023-10-15 13:16:10,273 - mini_batch_size: "4"
60
+ 2023-10-15 13:16:10,273 - max_epochs: "10"
61
+ 2023-10-15 13:16:10,273 - shuffle: "True"
62
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 13:16:10,273 Plugins:
64
+ 2023-10-15 13:16:10,273 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 13:16:10,273 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 13:16:10,273 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 13:16:10,273 Computation:
70
+ 2023-10-15 13:16:10,273 - compute on device: cuda:0
71
+ 2023-10-15 13:16:10,273 - embedding storage: none
72
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 13:16:10,273 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
74
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 13:16:10,273 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 13:16:35,448 epoch 1 - iter 521/5212 - loss 1.40362758 - time (sec): 25.17 - samples/sec: 1455.51 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-15 13:17:00,458 epoch 1 - iter 1042/5212 - loss 0.89883300 - time (sec): 50.18 - samples/sec: 1460.27 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-15 13:17:25,878 epoch 1 - iter 1563/5212 - loss 0.67555628 - time (sec): 75.60 - samples/sec: 1488.02 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-15 13:17:50,812 epoch 1 - iter 2084/5212 - loss 0.57238927 - time (sec): 100.54 - samples/sec: 1477.04 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-15 13:18:16,704 epoch 1 - iter 2605/5212 - loss 0.49502228 - time (sec): 126.43 - samples/sec: 1492.43 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-15 13:18:42,337 epoch 1 - iter 3126/5212 - loss 0.45192361 - time (sec): 152.06 - samples/sec: 1483.23 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-15 13:19:08,050 epoch 1 - iter 3647/5212 - loss 0.42525465 - time (sec): 177.78 - samples/sec: 1458.09 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-15 13:19:37,108 epoch 1 - iter 4168/5212 - loss 0.40135861 - time (sec): 206.83 - samples/sec: 1438.06 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-15 13:20:03,562 epoch 1 - iter 4689/5212 - loss 0.38395729 - time (sec): 233.29 - samples/sec: 1429.03 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-15 13:20:30,449 epoch 1 - iter 5210/5212 - loss 0.37262070 - time (sec): 260.17 - samples/sec: 1411.90 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-15 13:20:30,541 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 13:20:30,542 EPOCH 1 done: loss 0.3725 - lr: 0.000050
88
+ 2023-10-15 13:20:36,434 DEV : loss 0.11529939621686935 - f1-score (micro avg) 0.2459
89
+ 2023-10-15 13:20:36,460 saving best model
90
+ 2023-10-15 13:20:36,927 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 13:21:02,702 epoch 2 - iter 521/5212 - loss 0.19455856 - time (sec): 25.77 - samples/sec: 1478.33 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-15 13:21:28,495 epoch 2 - iter 1042/5212 - loss 0.20837326 - time (sec): 51.57 - samples/sec: 1394.34 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-15 13:21:56,301 epoch 2 - iter 1563/5212 - loss 0.19711389 - time (sec): 79.37 - samples/sec: 1372.97 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-15 13:22:22,909 epoch 2 - iter 2084/5212 - loss 0.19827196 - time (sec): 105.98 - samples/sec: 1387.68 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-15 13:22:50,780 epoch 2 - iter 2605/5212 - loss 0.19554159 - time (sec): 133.85 - samples/sec: 1376.38 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-15 13:23:16,034 epoch 2 - iter 3126/5212 - loss 0.19262159 - time (sec): 159.10 - samples/sec: 1367.76 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-15 13:23:41,047 epoch 2 - iter 3647/5212 - loss 0.19419779 - time (sec): 184.12 - samples/sec: 1389.83 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-15 13:24:06,335 epoch 2 - iter 4168/5212 - loss 0.19161465 - time (sec): 209.41 - samples/sec: 1401.90 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-15 13:24:34,437 epoch 2 - iter 4689/5212 - loss 0.19121284 - time (sec): 237.51 - samples/sec: 1404.82 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-15 13:25:02,829 epoch 2 - iter 5210/5212 - loss 0.19317263 - time (sec): 265.90 - samples/sec: 1381.36 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-15 13:25:02,932 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 13:25:02,932 EPOCH 2 done: loss 0.1931 - lr: 0.000044
103
+ 2023-10-15 13:25:12,816 DEV : loss 0.17544493079185486 - f1-score (micro avg) 0.3654
104
+ 2023-10-15 13:25:12,852 saving best model
105
+ 2023-10-15 13:25:13,376 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 13:25:39,265 epoch 3 - iter 521/5212 - loss 0.15145173 - time (sec): 25.88 - samples/sec: 1458.63 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-15 13:26:04,335 epoch 3 - iter 1042/5212 - loss 0.15678187 - time (sec): 50.95 - samples/sec: 1438.61 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-15 13:26:30,991 epoch 3 - iter 1563/5212 - loss 0.15448299 - time (sec): 77.61 - samples/sec: 1389.38 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-15 13:26:56,639 epoch 3 - iter 2084/5212 - loss 0.15527194 - time (sec): 103.26 - samples/sec: 1392.24 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-15 13:27:25,339 epoch 3 - iter 2605/5212 - loss 0.15240117 - time (sec): 131.96 - samples/sec: 1368.28 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-15 13:27:54,043 epoch 3 - iter 3126/5212 - loss 0.14898541 - time (sec): 160.66 - samples/sec: 1356.81 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-15 13:28:19,990 epoch 3 - iter 3647/5212 - loss 0.14601801 - time (sec): 186.61 - samples/sec: 1379.17 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-15 13:28:47,027 epoch 3 - iter 4168/5212 - loss 0.14444689 - time (sec): 213.65 - samples/sec: 1371.63 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-15 13:29:14,124 epoch 3 - iter 4689/5212 - loss 0.14484556 - time (sec): 240.74 - samples/sec: 1374.81 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-15 13:29:40,292 epoch 3 - iter 5210/5212 - loss 0.14315344 - time (sec): 266.91 - samples/sec: 1376.40 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-15 13:29:40,389 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 13:29:40,389 EPOCH 3 done: loss 0.1431 - lr: 0.000039
118
+ 2023-10-15 13:29:49,797 DEV : loss 0.23793098330497742 - f1-score (micro avg) 0.3084
119
+ 2023-10-15 13:29:49,824 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 13:30:14,394 epoch 4 - iter 521/5212 - loss 0.11205687 - time (sec): 24.57 - samples/sec: 1434.64 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-15 13:30:39,652 epoch 4 - iter 1042/5212 - loss 0.10513806 - time (sec): 49.83 - samples/sec: 1419.82 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-15 13:31:05,128 epoch 4 - iter 1563/5212 - loss 0.11011535 - time (sec): 75.30 - samples/sec: 1448.52 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-15 13:31:30,130 epoch 4 - iter 2084/5212 - loss 0.11135520 - time (sec): 100.31 - samples/sec: 1455.61 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-15 13:31:54,730 epoch 4 - iter 2605/5212 - loss 0.11031903 - time (sec): 124.91 - samples/sec: 1449.83 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-15 13:32:19,870 epoch 4 - iter 3126/5212 - loss 0.10935951 - time (sec): 150.05 - samples/sec: 1461.11 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-15 13:32:44,907 epoch 4 - iter 3647/5212 - loss 0.10865845 - time (sec): 175.08 - samples/sec: 1459.09 - lr: 0.000035 - momentum: 0.000000
127
+ 2023-10-15 13:33:10,118 epoch 4 - iter 4168/5212 - loss 0.10703262 - time (sec): 200.29 - samples/sec: 1465.07 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-15 13:33:35,323 epoch 4 - iter 4689/5212 - loss 0.10572271 - time (sec): 225.50 - samples/sec: 1457.14 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-15 13:34:01,141 epoch 4 - iter 5210/5212 - loss 0.10526184 - time (sec): 251.32 - samples/sec: 1461.60 - lr: 0.000033 - momentum: 0.000000
130
+ 2023-10-15 13:34:01,232 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 13:34:01,233 EPOCH 4 done: loss 0.1053 - lr: 0.000033
132
+ 2023-10-15 13:34:09,661 DEV : loss 0.23926250636577606 - f1-score (micro avg) 0.2991
133
+ 2023-10-15 13:34:09,704 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-15 13:34:36,513 epoch 5 - iter 521/5212 - loss 0.08552624 - time (sec): 26.81 - samples/sec: 1340.65 - lr: 0.000033 - momentum: 0.000000
135
+ 2023-10-15 13:35:01,593 epoch 5 - iter 1042/5212 - loss 0.07856810 - time (sec): 51.89 - samples/sec: 1363.29 - lr: 0.000032 - momentum: 0.000000
136
+ 2023-10-15 13:35:26,794 epoch 5 - iter 1563/5212 - loss 0.08133709 - time (sec): 77.09 - samples/sec: 1389.39 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-15 13:35:52,056 epoch 5 - iter 2084/5212 - loss 0.08333790 - time (sec): 102.35 - samples/sec: 1406.19 - lr: 0.000031 - momentum: 0.000000
138
+ 2023-10-15 13:36:17,357 epoch 5 - iter 2605/5212 - loss 0.08277969 - time (sec): 127.65 - samples/sec: 1409.26 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-15 13:36:43,034 epoch 5 - iter 3126/5212 - loss 0.08273593 - time (sec): 153.33 - samples/sec: 1420.12 - lr: 0.000030 - momentum: 0.000000
140
+ 2023-10-15 13:37:08,780 epoch 5 - iter 3647/5212 - loss 0.08084529 - time (sec): 179.07 - samples/sec: 1425.88 - lr: 0.000029 - momentum: 0.000000
141
+ 2023-10-15 13:37:34,843 epoch 5 - iter 4168/5212 - loss 0.08016018 - time (sec): 205.14 - samples/sec: 1436.62 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-15 13:38:00,799 epoch 5 - iter 4689/5212 - loss 0.07996177 - time (sec): 231.09 - samples/sec: 1442.23 - lr: 0.000028 - momentum: 0.000000
143
+ 2023-10-15 13:38:25,393 epoch 5 - iter 5210/5212 - loss 0.08003152 - time (sec): 255.69 - samples/sec: 1436.83 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-15 13:38:25,484 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-15 13:38:25,484 EPOCH 5 done: loss 0.0800 - lr: 0.000028
146
+ 2023-10-15 13:38:33,739 DEV : loss 0.3471781313419342 - f1-score (micro avg) 0.3072
147
+ 2023-10-15 13:38:33,768 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-15 13:38:59,130 epoch 6 - iter 521/5212 - loss 0.04893610 - time (sec): 25.36 - samples/sec: 1550.29 - lr: 0.000027 - momentum: 0.000000
149
+ 2023-10-15 13:39:24,147 epoch 6 - iter 1042/5212 - loss 0.05092956 - time (sec): 50.38 - samples/sec: 1516.20 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-15 13:39:50,524 epoch 6 - iter 1563/5212 - loss 0.05455800 - time (sec): 76.76 - samples/sec: 1509.47 - lr: 0.000026 - momentum: 0.000000
151
+ 2023-10-15 13:40:16,811 epoch 6 - iter 2084/5212 - loss 0.05559083 - time (sec): 103.04 - samples/sec: 1460.28 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-15 13:40:41,525 epoch 6 - iter 2605/5212 - loss 0.05732132 - time (sec): 127.76 - samples/sec: 1434.33 - lr: 0.000025 - momentum: 0.000000
153
+ 2023-10-15 13:41:07,304 epoch 6 - iter 3126/5212 - loss 0.05727400 - time (sec): 153.54 - samples/sec: 1440.88 - lr: 0.000024 - momentum: 0.000000
154
+ 2023-10-15 13:41:32,019 epoch 6 - iter 3647/5212 - loss 0.05654496 - time (sec): 178.25 - samples/sec: 1428.82 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-15 13:41:57,953 epoch 6 - iter 4168/5212 - loss 0.05631661 - time (sec): 204.18 - samples/sec: 1442.93 - lr: 0.000023 - momentum: 0.000000
156
+ 2023-10-15 13:42:23,594 epoch 6 - iter 4689/5212 - loss 0.05593561 - time (sec): 229.82 - samples/sec: 1445.76 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-15 13:42:48,376 epoch 6 - iter 5210/5212 - loss 0.05663169 - time (sec): 254.61 - samples/sec: 1442.84 - lr: 0.000022 - momentum: 0.000000
158
+ 2023-10-15 13:42:48,471 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-15 13:42:48,471 EPOCH 6 done: loss 0.0566 - lr: 0.000022
160
+ 2023-10-15 13:42:56,767 DEV : loss 0.3654634654521942 - f1-score (micro avg) 0.3514
161
+ 2023-10-15 13:42:56,796 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-15 13:43:21,792 epoch 7 - iter 521/5212 - loss 0.03312481 - time (sec): 25.00 - samples/sec: 1410.93 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-15 13:43:47,410 epoch 7 - iter 1042/5212 - loss 0.03774246 - time (sec): 50.61 - samples/sec: 1425.50 - lr: 0.000021 - momentum: 0.000000
164
+ 2023-10-15 13:44:12,732 epoch 7 - iter 1563/5212 - loss 0.04380652 - time (sec): 75.94 - samples/sec: 1443.44 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-15 13:44:37,716 epoch 7 - iter 2084/5212 - loss 0.04357735 - time (sec): 100.92 - samples/sec: 1453.08 - lr: 0.000020 - momentum: 0.000000
166
+ 2023-10-15 13:45:02,712 epoch 7 - iter 2605/5212 - loss 0.04260736 - time (sec): 125.92 - samples/sec: 1453.59 - lr: 0.000019 - momentum: 0.000000
167
+ 2023-10-15 13:45:27,992 epoch 7 - iter 3126/5212 - loss 0.04472481 - time (sec): 151.20 - samples/sec: 1454.29 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-15 13:45:54,498 epoch 7 - iter 3647/5212 - loss 0.04429013 - time (sec): 177.70 - samples/sec: 1454.84 - lr: 0.000018 - momentum: 0.000000
169
+ 2023-10-15 13:46:19,450 epoch 7 - iter 4168/5212 - loss 0.04403234 - time (sec): 202.65 - samples/sec: 1452.76 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-15 13:46:44,374 epoch 7 - iter 4689/5212 - loss 0.04448230 - time (sec): 227.58 - samples/sec: 1451.83 - lr: 0.000017 - momentum: 0.000000
171
+ 2023-10-15 13:47:09,346 epoch 7 - iter 5210/5212 - loss 0.04403781 - time (sec): 252.55 - samples/sec: 1453.77 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-15 13:47:09,455 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-15 13:47:09,455 EPOCH 7 done: loss 0.0440 - lr: 0.000017
174
+ 2023-10-15 13:47:17,889 DEV : loss 0.34585681557655334 - f1-score (micro avg) 0.3311
175
+ 2023-10-15 13:47:17,919 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-15 13:47:43,267 epoch 8 - iter 521/5212 - loss 0.02689241 - time (sec): 25.35 - samples/sec: 1476.10 - lr: 0.000016 - momentum: 0.000000
177
+ 2023-10-15 13:48:08,186 epoch 8 - iter 1042/5212 - loss 0.02653419 - time (sec): 50.27 - samples/sec: 1445.14 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-15 13:48:33,822 epoch 8 - iter 1563/5212 - loss 0.02637713 - time (sec): 75.90 - samples/sec: 1441.02 - lr: 0.000015 - momentum: 0.000000
179
+ 2023-10-15 13:48:59,594 epoch 8 - iter 2084/5212 - loss 0.02733328 - time (sec): 101.67 - samples/sec: 1424.32 - lr: 0.000014 - momentum: 0.000000
180
+ 2023-10-15 13:49:24,714 epoch 8 - iter 2605/5212 - loss 0.02880069 - time (sec): 126.79 - samples/sec: 1426.83 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-15 13:49:51,571 epoch 8 - iter 3126/5212 - loss 0.02954909 - time (sec): 153.65 - samples/sec: 1435.69 - lr: 0.000013 - momentum: 0.000000
182
+ 2023-10-15 13:50:16,561 epoch 8 - iter 3647/5212 - loss 0.03040527 - time (sec): 178.64 - samples/sec: 1444.08 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-15 13:50:41,614 epoch 8 - iter 4168/5212 - loss 0.03036981 - time (sec): 203.69 - samples/sec: 1438.53 - lr: 0.000012 - momentum: 0.000000
184
+ 2023-10-15 13:51:06,615 epoch 8 - iter 4689/5212 - loss 0.02995803 - time (sec): 228.69 - samples/sec: 1442.97 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-15 13:51:33,708 epoch 8 - iter 5210/5212 - loss 0.03001161 - time (sec): 255.79 - samples/sec: 1435.88 - lr: 0.000011 - momentum: 0.000000
186
+ 2023-10-15 13:51:33,815 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-15 13:51:33,816 EPOCH 8 done: loss 0.0300 - lr: 0.000011
188
+ 2023-10-15 13:51:42,835 DEV : loss 0.43034762144088745 - f1-score (micro avg) 0.3493
189
+ 2023-10-15 13:51:42,865 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-15 13:52:08,590 epoch 9 - iter 521/5212 - loss 0.02763096 - time (sec): 25.72 - samples/sec: 1534.61 - lr: 0.000011 - momentum: 0.000000
191
+ 2023-10-15 13:52:33,552 epoch 9 - iter 1042/5212 - loss 0.02207242 - time (sec): 50.69 - samples/sec: 1519.59 - lr: 0.000010 - momentum: 0.000000
192
+ 2023-10-15 13:52:58,597 epoch 9 - iter 1563/5212 - loss 0.02236283 - time (sec): 75.73 - samples/sec: 1460.68 - lr: 0.000009 - momentum: 0.000000
193
+ 2023-10-15 13:53:23,759 epoch 9 - iter 2084/5212 - loss 0.02177159 - time (sec): 100.89 - samples/sec: 1469.42 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-15 13:53:49,224 epoch 9 - iter 2605/5212 - loss 0.02114199 - time (sec): 126.36 - samples/sec: 1462.95 - lr: 0.000008 - momentum: 0.000000
195
+ 2023-10-15 13:54:14,015 epoch 9 - iter 3126/5212 - loss 0.02090215 - time (sec): 151.15 - samples/sec: 1458.83 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-15 13:54:39,333 epoch 9 - iter 3647/5212 - loss 0.02100536 - time (sec): 176.47 - samples/sec: 1454.71 - lr: 0.000007 - momentum: 0.000000
197
+ 2023-10-15 13:55:04,484 epoch 9 - iter 4168/5212 - loss 0.02104071 - time (sec): 201.62 - samples/sec: 1453.99 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-15 13:55:30,243 epoch 9 - iter 4689/5212 - loss 0.02069058 - time (sec): 227.38 - samples/sec: 1439.28 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-15 13:55:56,291 epoch 9 - iter 5210/5212 - loss 0.02049929 - time (sec): 253.42 - samples/sec: 1449.41 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-15 13:55:56,382 ----------------------------------------------------------------------------------------------------
201
+ 2023-10-15 13:55:56,383 EPOCH 9 done: loss 0.0205 - lr: 0.000006
202
+ 2023-10-15 13:56:05,381 DEV : loss 0.5122669339179993 - f1-score (micro avg) 0.3381
203
+ 2023-10-15 13:56:05,407 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-15 13:56:30,432 epoch 10 - iter 521/5212 - loss 0.01178875 - time (sec): 25.02 - samples/sec: 1530.92 - lr: 0.000005 - momentum: 0.000000
205
+ 2023-10-15 13:56:55,856 epoch 10 - iter 1042/5212 - loss 0.01468480 - time (sec): 50.45 - samples/sec: 1507.44 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-15 13:57:21,319 epoch 10 - iter 1563/5212 - loss 0.01459527 - time (sec): 75.91 - samples/sec: 1516.01 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-15 13:57:46,537 epoch 10 - iter 2084/5212 - loss 0.01485247 - time (sec): 101.13 - samples/sec: 1506.02 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-15 13:58:11,468 epoch 10 - iter 2605/5212 - loss 0.01444242 - time (sec): 126.06 - samples/sec: 1494.49 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-15 13:58:36,105 epoch 10 - iter 3126/5212 - loss 0.01442687 - time (sec): 150.70 - samples/sec: 1471.55 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-15 13:59:01,128 epoch 10 - iter 3647/5212 - loss 0.01393897 - time (sec): 175.72 - samples/sec: 1466.96 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-15 13:59:26,149 epoch 10 - iter 4168/5212 - loss 0.01439542 - time (sec): 200.74 - samples/sec: 1466.86 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-15 13:59:51,374 epoch 10 - iter 4689/5212 - loss 0.01427652 - time (sec): 225.97 - samples/sec: 1470.47 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 14:00:15,042 epoch 10 - iter 5210/5212 - loss 0.01384805 - time (sec): 249.63 - samples/sec: 1471.63 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-15 14:00:15,126 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-15 14:00:15,126 EPOCH 10 done: loss 0.0138 - lr: 0.000000
216
+ 2023-10-15 14:00:24,139 DEV : loss 0.46498292684555054 - f1-score (micro avg) 0.3499
217
+ 2023-10-15 14:00:24,544 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-15 14:00:24,545 Loading model from best epoch ...
219
+ 2023-10-15 14:00:26,145 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
220
+ 2023-10-15 14:00:42,346
221
+ Results:
222
+ - F-score (micro) 0.2834
223
+ - F-score (macro) 0.1787
224
+ - Accuracy 0.166
225
+
226
+ By class:
227
+ precision recall f1-score support
228
+
229
+ LOC 0.3432 0.3534 0.3482 1214
230
+ PER 0.3094 0.2030 0.2451 808
231
+ ORG 0.1345 0.1105 0.1213 353
232
+ HumanProd 0.0000 0.0000 0.0000 15
233
+
234
+ micro avg 0.3053 0.2644 0.2834 2390
235
+ macro avg 0.1968 0.1667 0.1787 2390
236
+ weighted avg 0.2988 0.2644 0.2777 2390
237
+
238
+ 2023-10-15 14:00:42,346 ----------------------------------------------------------------------------------------------------