Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs8-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
0cf179e
•
1 Parent(s): 171fcf2

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 +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e3b5b0b82f97a38d4f97acae8fd91bd1c94208c25feffc8bda31937673be822
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 12:00:19 0.0000 0.3839 0.1149 0.2432 0.5947 0.3452 0.2095
3
+ 2 12:03:36 0.0000 0.1548 0.1457 0.2774 0.5947 0.3783 0.2350
4
+ 3 12:06:58 0.0000 0.1081 0.3048 0.2210 0.6648 0.3318 0.2000
5
+ 4 12:10:21 0.0000 0.0789 0.3154 0.2307 0.6004 0.3333 0.2008
6
+ 5 12:13:48 0.0000 0.0569 0.3149 0.2766 0.4830 0.3517 0.2145
7
+ 6 12:17:08 0.0000 0.0435 0.3962 0.2611 0.6136 0.3663 0.2247
8
+ 7 12:20:26 0.0000 0.0310 0.3647 0.2696 0.5720 0.3665 0.2255
9
+ 8 12:23:45 0.0000 0.0215 0.3675 0.3055 0.5682 0.3974 0.2490
10
+ 9 12:27:06 0.0000 0.0157 0.4267 0.2845 0.5777 0.3812 0.2372
11
+ 10 12:30:26 0.0000 0.0100 0.4643 0.2690 0.6042 0.3722 0.2302
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 11:57:02,477 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 11:57:02,478 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 11:57:02,478 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 11:57:02,479 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 11:57:02,479 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 11:57:02,479 Train: 20847 sentences
55
+ 2023-10-15 11:57:02,479 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 11:57:02,479 Training Params:
58
+ 2023-10-15 11:57:02,479 - learning_rate: "5e-05"
59
+ 2023-10-15 11:57:02,479 - mini_batch_size: "8"
60
+ 2023-10-15 11:57:02,479 - max_epochs: "10"
61
+ 2023-10-15 11:57:02,479 - shuffle: "True"
62
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 11:57:02,479 Plugins:
64
+ 2023-10-15 11:57:02,479 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 11:57:02,479 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 11:57:02,479 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 11:57:02,479 Computation:
70
+ 2023-10-15 11:57:02,479 - compute on device: cuda:0
71
+ 2023-10-15 11:57:02,479 - embedding storage: none
72
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 11:57:02,479 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
74
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 11:57:02,479 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 11:57:21,436 epoch 1 - iter 260/2606 - loss 1.64915807 - time (sec): 18.96 - samples/sec: 1932.05 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-15 11:57:40,112 epoch 1 - iter 520/2606 - loss 1.03092842 - time (sec): 37.63 - samples/sec: 1945.85 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-15 11:57:59,795 epoch 1 - iter 780/2606 - loss 0.76385186 - time (sec): 57.31 - samples/sec: 1959.94 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-15 11:58:18,335 epoch 1 - iter 1040/2606 - loss 0.64168041 - time (sec): 75.85 - samples/sec: 1954.01 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-15 11:58:38,681 epoch 1 - iter 1300/2606 - loss 0.54874820 - time (sec): 96.20 - samples/sec: 1954.88 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-15 11:58:57,821 epoch 1 - iter 1560/2606 - loss 0.49391289 - time (sec): 115.34 - samples/sec: 1952.63 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-15 11:59:15,529 epoch 1 - iter 1820/2606 - loss 0.45687451 - time (sec): 133.05 - samples/sec: 1944.20 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-15 11:59:34,789 epoch 1 - iter 2080/2606 - loss 0.42534550 - time (sec): 152.31 - samples/sec: 1947.69 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-15 11:59:53,865 epoch 1 - iter 2340/2606 - loss 0.40230225 - time (sec): 171.39 - samples/sec: 1941.28 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-15 12:00:12,299 epoch 1 - iter 2600/2606 - loss 0.38436630 - time (sec): 189.82 - samples/sec: 1932.40 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-15 12:00:12,697 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 12:00:12,697 EPOCH 1 done: loss 0.3839 - lr: 0.000050
88
+ 2023-10-15 12:00:19,023 DEV : loss 0.11487094312906265 - f1-score (micro avg) 0.3452
89
+ 2023-10-15 12:00:19,047 saving best model
90
+ 2023-10-15 12:00:19,408 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 12:00:38,662 epoch 2 - iter 260/2606 - loss 0.15682186 - time (sec): 19.25 - samples/sec: 1974.97 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-15 12:00:57,114 epoch 2 - iter 520/2606 - loss 0.16679577 - time (sec): 37.70 - samples/sec: 1893.71 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-15 12:01:16,199 epoch 2 - iter 780/2606 - loss 0.16159520 - time (sec): 56.79 - samples/sec: 1915.43 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-15 12:01:35,382 epoch 2 - iter 1040/2606 - loss 0.16406407 - time (sec): 75.97 - samples/sec: 1932.84 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-15 12:01:54,053 epoch 2 - iter 1300/2606 - loss 0.16300929 - time (sec): 94.64 - samples/sec: 1942.04 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-15 12:02:11,883 epoch 2 - iter 1560/2606 - loss 0.15872115 - time (sec): 112.47 - samples/sec: 1931.89 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-15 12:02:31,185 epoch 2 - iter 1820/2606 - loss 0.15896498 - time (sec): 131.78 - samples/sec: 1934.46 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-15 12:02:50,291 epoch 2 - iter 2080/2606 - loss 0.15531835 - time (sec): 150.88 - samples/sec: 1942.94 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-15 12:03:10,327 epoch 2 - iter 2340/2606 - loss 0.15458671 - time (sec): 170.92 - samples/sec: 1947.88 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-15 12:03:28,176 epoch 2 - iter 2600/2606 - loss 0.15487459 - time (sec): 188.77 - samples/sec: 1942.09 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-15 12:03:28,565 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 12:03:28,565 EPOCH 2 done: loss 0.1548 - lr: 0.000044
103
+ 2023-10-15 12:03:36,764 DEV : loss 0.1457301676273346 - f1-score (micro avg) 0.3783
104
+ 2023-10-15 12:03:36,788 saving best model
105
+ 2023-10-15 12:03:38,006 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 12:03:57,780 epoch 3 - iter 260/2606 - loss 0.10903042 - time (sec): 19.77 - samples/sec: 1908.20 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-15 12:04:16,420 epoch 3 - iter 520/2606 - loss 0.11437700 - time (sec): 38.41 - samples/sec: 1904.63 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-15 12:04:35,156 epoch 3 - iter 780/2606 - loss 0.11275118 - time (sec): 57.15 - samples/sec: 1885.24 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-15 12:04:53,844 epoch 3 - iter 1040/2606 - loss 0.11675125 - time (sec): 75.84 - samples/sec: 1892.01 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-15 12:05:12,544 epoch 3 - iter 1300/2606 - loss 0.11512835 - time (sec): 94.54 - samples/sec: 1906.30 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-15 12:05:31,526 epoch 3 - iter 1560/2606 - loss 0.11310394 - time (sec): 113.52 - samples/sec: 1918.20 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-15 12:05:50,869 epoch 3 - iter 1820/2606 - loss 0.11220806 - time (sec): 132.86 - samples/sec: 1932.70 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-15 12:06:09,318 epoch 3 - iter 2080/2606 - loss 0.11087298 - time (sec): 151.31 - samples/sec: 1934.37 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-15 12:06:29,929 epoch 3 - iter 2340/2606 - loss 0.11008659 - time (sec): 171.92 - samples/sec: 1918.75 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-15 12:06:50,002 epoch 3 - iter 2600/2606 - loss 0.10813718 - time (sec): 191.99 - samples/sec: 1910.97 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-15 12:06:50,397 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 12:06:50,397 EPOCH 3 done: loss 0.1081 - lr: 0.000039
118
+ 2023-10-15 12:06:58,735 DEV : loss 0.30479252338409424 - f1-score (micro avg) 0.3318
119
+ 2023-10-15 12:06:58,777 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 12:07:18,000 epoch 4 - iter 260/2606 - loss 0.08238358 - time (sec): 19.22 - samples/sec: 1831.30 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-15 12:07:38,192 epoch 4 - iter 520/2606 - loss 0.08072289 - time (sec): 39.41 - samples/sec: 1790.47 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-15 12:07:59,891 epoch 4 - iter 780/2606 - loss 0.08208316 - time (sec): 61.11 - samples/sec: 1780.28 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-15 12:08:18,418 epoch 4 - iter 1040/2606 - loss 0.08409013 - time (sec): 79.64 - samples/sec: 1829.08 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-15 12:08:36,789 epoch 4 - iter 1300/2606 - loss 0.08241002 - time (sec): 98.01 - samples/sec: 1845.09 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-15 12:08:56,156 epoch 4 - iter 1560/2606 - loss 0.08187116 - time (sec): 117.38 - samples/sec: 1860.51 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-15 12:09:15,526 epoch 4 - iter 1820/2606 - loss 0.08319336 - time (sec): 136.75 - samples/sec: 1865.81 - lr: 0.000035 - momentum: 0.000000
127
+ 2023-10-15 12:09:34,317 epoch 4 - iter 2080/2606 - loss 0.08131155 - time (sec): 155.54 - samples/sec: 1884.19 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-15 12:09:52,846 epoch 4 - iter 2340/2606 - loss 0.08077706 - time (sec): 174.07 - samples/sec: 1883.36 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-15 12:10:12,717 epoch 4 - iter 2600/2606 - loss 0.07876999 - time (sec): 193.94 - samples/sec: 1888.93 - lr: 0.000033 - momentum: 0.000000
130
+ 2023-10-15 12:10:13,105 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 12:10:13,105 EPOCH 4 done: loss 0.0789 - lr: 0.000033
132
+ 2023-10-15 12:10:21,438 DEV : loss 0.3154134452342987 - f1-score (micro avg) 0.3333
133
+ 2023-10-15 12:10:21,475 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-15 12:10:40,522 epoch 5 - iter 260/2606 - loss 0.05258224 - time (sec): 19.04 - samples/sec: 1881.49 - lr: 0.000033 - momentum: 0.000000
135
+ 2023-10-15 12:10:58,880 epoch 5 - iter 520/2606 - loss 0.04989730 - time (sec): 37.40 - samples/sec: 1885.34 - lr: 0.000032 - momentum: 0.000000
136
+ 2023-10-15 12:11:18,620 epoch 5 - iter 780/2606 - loss 0.05552002 - time (sec): 57.14 - samples/sec: 1871.31 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-15 12:11:38,270 epoch 5 - iter 1040/2606 - loss 0.05584031 - time (sec): 76.79 - samples/sec: 1871.27 - lr: 0.000031 - momentum: 0.000000
138
+ 2023-10-15 12:11:59,453 epoch 5 - iter 1300/2606 - loss 0.05720019 - time (sec): 97.98 - samples/sec: 1834.07 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-15 12:12:19,751 epoch 5 - iter 1560/2606 - loss 0.05743220 - time (sec): 118.27 - samples/sec: 1838.76 - lr: 0.000030 - momentum: 0.000000
140
+ 2023-10-15 12:12:40,318 epoch 5 - iter 1820/2606 - loss 0.05650180 - time (sec): 138.84 - samples/sec: 1833.34 - lr: 0.000029 - momentum: 0.000000
141
+ 2023-10-15 12:13:01,319 epoch 5 - iter 2080/2606 - loss 0.05647238 - time (sec): 159.84 - samples/sec: 1839.62 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-15 12:13:21,100 epoch 5 - iter 2340/2606 - loss 0.05690768 - time (sec): 179.62 - samples/sec: 1852.23 - lr: 0.000028 - momentum: 0.000000
143
+ 2023-10-15 12:13:40,116 epoch 5 - iter 2600/2606 - loss 0.05691287 - time (sec): 198.64 - samples/sec: 1845.60 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-15 12:13:40,605 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-15 12:13:40,605 EPOCH 5 done: loss 0.0569 - lr: 0.000028
146
+ 2023-10-15 12:13:48,784 DEV : loss 0.31486544013023376 - f1-score (micro avg) 0.3517
147
+ 2023-10-15 12:13:48,815 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-15 12:14:08,648 epoch 6 - iter 260/2606 - loss 0.03313725 - time (sec): 19.83 - samples/sec: 1979.16 - lr: 0.000027 - momentum: 0.000000
149
+ 2023-10-15 12:14:27,561 epoch 6 - iter 520/2606 - loss 0.03690228 - time (sec): 38.74 - samples/sec: 1969.87 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-15 12:14:47,369 epoch 6 - iter 780/2606 - loss 0.04112557 - time (sec): 58.55 - samples/sec: 1974.98 - lr: 0.000026 - momentum: 0.000000
151
+ 2023-10-15 12:15:05,825 epoch 6 - iter 1040/2606 - loss 0.04204714 - time (sec): 77.01 - samples/sec: 1950.19 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-15 12:15:23,370 epoch 6 - iter 1300/2606 - loss 0.04230053 - time (sec): 94.55 - samples/sec: 1935.24 - lr: 0.000025 - momentum: 0.000000
153
+ 2023-10-15 12:15:42,734 epoch 6 - iter 1560/2606 - loss 0.04315007 - time (sec): 113.92 - samples/sec: 1939.01 - lr: 0.000024 - momentum: 0.000000
154
+ 2023-10-15 12:16:00,920 epoch 6 - iter 1820/2606 - loss 0.04303887 - time (sec): 132.10 - samples/sec: 1925.16 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-15 12:16:21,944 epoch 6 - iter 2080/2606 - loss 0.04257364 - time (sec): 153.13 - samples/sec: 1917.79 - lr: 0.000023 - momentum: 0.000000
156
+ 2023-10-15 12:16:41,637 epoch 6 - iter 2340/2606 - loss 0.04413961 - time (sec): 172.82 - samples/sec: 1917.84 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-15 12:16:59,561 epoch 6 - iter 2600/2606 - loss 0.04348856 - time (sec): 190.74 - samples/sec: 1922.13 - lr: 0.000022 - momentum: 0.000000
158
+ 2023-10-15 12:16:59,994 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-15 12:16:59,994 EPOCH 6 done: loss 0.0435 - lr: 0.000022
160
+ 2023-10-15 12:17:08,237 DEV : loss 0.39616382122039795 - f1-score (micro avg) 0.3663
161
+ 2023-10-15 12:17:08,263 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-15 12:17:26,344 epoch 7 - iter 260/2606 - loss 0.02748825 - time (sec): 18.08 - samples/sec: 1948.18 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-15 12:17:45,453 epoch 7 - iter 520/2606 - loss 0.02809898 - time (sec): 37.19 - samples/sec: 1935.59 - lr: 0.000021 - momentum: 0.000000
164
+ 2023-10-15 12:18:04,516 epoch 7 - iter 780/2606 - loss 0.03121037 - time (sec): 56.25 - samples/sec: 1946.76 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-15 12:18:23,273 epoch 7 - iter 1040/2606 - loss 0.03049926 - time (sec): 75.01 - samples/sec: 1948.46 - lr: 0.000020 - momentum: 0.000000
166
+ 2023-10-15 12:18:42,018 epoch 7 - iter 1300/2606 - loss 0.02965768 - time (sec): 93.75 - samples/sec: 1946.91 - lr: 0.000019 - momentum: 0.000000
167
+ 2023-10-15 12:19:01,527 epoch 7 - iter 1560/2606 - loss 0.03081344 - time (sec): 113.26 - samples/sec: 1938.31 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-15 12:19:21,614 epoch 7 - iter 1820/2606 - loss 0.03116566 - time (sec): 133.35 - samples/sec: 1935.29 - lr: 0.000018 - momentum: 0.000000
169
+ 2023-10-15 12:19:40,080 epoch 7 - iter 2080/2606 - loss 0.03093073 - time (sec): 151.82 - samples/sec: 1936.42 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-15 12:19:58,599 epoch 7 - iter 2340/2606 - loss 0.03093953 - time (sec): 170.33 - samples/sec: 1937.08 - lr: 0.000017 - momentum: 0.000000
171
+ 2023-10-15 12:20:17,845 epoch 7 - iter 2600/2606 - loss 0.03105048 - time (sec): 189.58 - samples/sec: 1931.82 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-15 12:20:18,356 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-15 12:20:18,356 EPOCH 7 done: loss 0.0310 - lr: 0.000017
174
+ 2023-10-15 12:20:26,693 DEV : loss 0.36471426486968994 - f1-score (micro avg) 0.3665
175
+ 2023-10-15 12:20:26,727 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-15 12:20:46,325 epoch 8 - iter 260/2606 - loss 0.01927091 - time (sec): 19.60 - samples/sec: 1907.34 - lr: 0.000016 - momentum: 0.000000
177
+ 2023-10-15 12:21:04,809 epoch 8 - iter 520/2606 - loss 0.02292594 - time (sec): 38.08 - samples/sec: 1901.26 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-15 12:21:23,743 epoch 8 - iter 780/2606 - loss 0.02009535 - time (sec): 57.01 - samples/sec: 1910.47 - lr: 0.000015 - momentum: 0.000000
179
+ 2023-10-15 12:21:42,591 epoch 8 - iter 1040/2606 - loss 0.01979498 - time (sec): 75.86 - samples/sec: 1906.19 - lr: 0.000014 - momentum: 0.000000
180
+ 2023-10-15 12:22:01,421 epoch 8 - iter 1300/2606 - loss 0.02040310 - time (sec): 94.69 - samples/sec: 1906.76 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-15 12:22:21,402 epoch 8 - iter 1560/2606 - loss 0.02110463 - time (sec): 114.67 - samples/sec: 1922.44 - lr: 0.000013 - momentum: 0.000000
182
+ 2023-10-15 12:22:40,169 epoch 8 - iter 1820/2606 - loss 0.02252478 - time (sec): 133.44 - samples/sec: 1931.25 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-15 12:22:58,486 epoch 8 - iter 2080/2606 - loss 0.02230129 - time (sec): 151.76 - samples/sec: 1924.93 - lr: 0.000012 - momentum: 0.000000
184
+ 2023-10-15 12:23:17,451 epoch 8 - iter 2340/2606 - loss 0.02176476 - time (sec): 170.72 - samples/sec: 1930.49 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-15 12:23:36,387 epoch 8 - iter 2600/2606 - loss 0.02155911 - time (sec): 189.66 - samples/sec: 1931.48 - lr: 0.000011 - momentum: 0.000000
186
+ 2023-10-15 12:23:36,903 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-15 12:23:36,903 EPOCH 8 done: loss 0.0215 - lr: 0.000011
188
+ 2023-10-15 12:23:45,945 DEV : loss 0.36754974722862244 - f1-score (micro avg) 0.3974
189
+ 2023-10-15 12:23:45,976 saving best model
190
+ 2023-10-15 12:23:46,501 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-15 12:24:07,094 epoch 9 - iter 260/2606 - loss 0.01881708 - time (sec): 20.59 - samples/sec: 1915.94 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-15 12:24:26,071 epoch 9 - iter 520/2606 - loss 0.01565953 - time (sec): 39.57 - samples/sec: 1945.70 - lr: 0.000010 - momentum: 0.000000
193
+ 2023-10-15 12:24:43,744 epoch 9 - iter 780/2606 - loss 0.01545121 - time (sec): 57.24 - samples/sec: 1930.42 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-15 12:25:03,140 epoch 9 - iter 1040/2606 - loss 0.01662704 - time (sec): 76.64 - samples/sec: 1933.14 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-15 12:25:22,562 epoch 9 - iter 1300/2606 - loss 0.01629208 - time (sec): 96.06 - samples/sec: 1921.45 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-15 12:25:41,153 epoch 9 - iter 1560/2606 - loss 0.01635871 - time (sec): 114.65 - samples/sec: 1919.58 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-15 12:26:00,050 epoch 9 - iter 1820/2606 - loss 0.01600567 - time (sec): 133.55 - samples/sec: 1914.66 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-15 12:26:19,201 epoch 9 - iter 2080/2606 - loss 0.01600079 - time (sec): 152.70 - samples/sec: 1915.23 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-15 12:26:37,301 epoch 9 - iter 2340/2606 - loss 0.01585391 - time (sec): 170.80 - samples/sec: 1912.05 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-15 12:26:57,026 epoch 9 - iter 2600/2606 - loss 0.01575358 - time (sec): 190.52 - samples/sec: 1921.35 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-15 12:26:57,653 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-15 12:26:57,654 EPOCH 9 done: loss 0.0157 - lr: 0.000006
203
+ 2023-10-15 12:27:06,597 DEV : loss 0.4267415404319763 - f1-score (micro avg) 0.3812
204
+ 2023-10-15 12:27:06,622 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-15 12:27:26,051 epoch 10 - iter 260/2606 - loss 0.00922264 - time (sec): 19.43 - samples/sec: 1970.34 - lr: 0.000005 - momentum: 0.000000
206
+ 2023-10-15 12:27:45,707 epoch 10 - iter 520/2606 - loss 0.01108063 - time (sec): 39.08 - samples/sec: 1937.27 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-15 12:28:05,674 epoch 10 - iter 780/2606 - loss 0.00999059 - time (sec): 59.05 - samples/sec: 1947.99 - lr: 0.000004 - momentum: 0.000000
208
+ 2023-10-15 12:28:25,124 epoch 10 - iter 1040/2606 - loss 0.01000085 - time (sec): 78.50 - samples/sec: 1938.59 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-15 12:28:44,187 epoch 10 - iter 1300/2606 - loss 0.00987250 - time (sec): 97.56 - samples/sec: 1928.75 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-15 12:29:02,183 epoch 10 - iter 1560/2606 - loss 0.00985511 - time (sec): 115.56 - samples/sec: 1916.59 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-15 12:29:20,689 epoch 10 - iter 1820/2606 - loss 0.00987583 - time (sec): 134.07 - samples/sec: 1919.64 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-15 12:29:39,612 epoch 10 - iter 2080/2606 - loss 0.00980796 - time (sec): 152.99 - samples/sec: 1922.22 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 12:29:59,077 epoch 10 - iter 2340/2606 - loss 0.01015117 - time (sec): 172.45 - samples/sec: 1922.11 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-15 12:30:17,379 epoch 10 - iter 2600/2606 - loss 0.00999133 - time (sec): 190.76 - samples/sec: 1921.72 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-15 12:30:17,751 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-15 12:30:17,751 EPOCH 10 done: loss 0.0100 - lr: 0.000000
217
+ 2023-10-15 12:30:26,946 DEV : loss 0.4642558991909027 - f1-score (micro avg) 0.3722
218
+ 2023-10-15 12:30:27,450 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-15 12:30:27,451 Loading model from best epoch ...
220
+ 2023-10-15 12:30:29,221 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
+ 2023-10-15 12:30:45,144
222
+ Results:
223
+ - F-score (micro) 0.4529
224
+ - F-score (macro) 0.3045
225
+ - Accuracy 0.2957
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.5466 0.5461 0.5464 1214
231
+ PER 0.3978 0.3948 0.3963 808
232
+ ORG 0.2791 0.2720 0.2755 353
233
+ HumanProd 0.0000 0.0000 0.0000 15
234
+
235
+ micro avg 0.4549 0.4510 0.4529 2390
236
+ macro avg 0.3059 0.3032 0.3045 2390
237
+ weighted avg 0.4533 0.4510 0.4522 2390
238
+
239
+ 2023-10-15 12:30:45,144 ----------------------------------------------------------------------------------------------------