Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
84a37dc
•
1 Parent(s): 90d4b8e

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 sha256:5e47c4fd9658c9d6bff99e5208ced1fc44de1cb380713f39c3ea99fd29d80d6f
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 14:38:32 0.0000 0.3961 0.1704 0.1758 0.6042 0.2723 0.1582
3
+ 2 14:41:50 0.0000 0.1559 0.1519 0.2879 0.5795 0.3847 0.2396
4
+ 3 14:45:08 0.0000 0.1090 0.2589 0.2275 0.6591 0.3382 0.2046
5
+ 4 14:48:26 0.0000 0.0782 0.2621 0.2737 0.5360 0.3624 0.2220
6
+ 5 14:51:43 0.0000 0.0589 0.3399 0.2375 0.5758 0.3363 0.2042
7
+ 6 14:55:03 0.0000 0.0432 0.3315 0.2758 0.5720 0.3722 0.2298
8
+ 7 14:58:20 0.0000 0.0309 0.4455 0.2594 0.5985 0.3620 0.2222
9
+ 8 15:01:36 0.0000 0.0233 0.5099 0.2349 0.6269 0.3418 0.2070
10
+ 9 15:04:56 0.0000 0.0153 0.4624 0.2572 0.6098 0.3618 0.2222
11
+ 10 15:08:14 0.0000 0.0102 0.4726 0.2704 0.6080 0.3743 0.2314
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 14:35:18,166 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 14:35:18,167 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 14:35:18,167 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 14:35:18,167 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 14:35:18,167 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 14:35:18,167 Train: 20847 sentences
55
+ 2023-10-15 14:35:18,167 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 14:35:18,167 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 14:35:18,167 Training Params:
58
+ 2023-10-15 14:35:18,167 - learning_rate: "5e-05"
59
+ 2023-10-15 14:35:18,167 - mini_batch_size: "8"
60
+ 2023-10-15 14:35:18,167 - max_epochs: "10"
61
+ 2023-10-15 14:35:18,167 - shuffle: "True"
62
+ 2023-10-15 14:35:18,167 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 14:35:18,167 Plugins:
64
+ 2023-10-15 14:35:18,167 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 14:35:18,167 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 14:35:18,167 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 14:35:18,167 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 14:35:18,167 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 14:35:18,167 Computation:
70
+ 2023-10-15 14:35:18,168 - compute on device: cuda:0
71
+ 2023-10-15 14:35:18,168 - embedding storage: none
72
+ 2023-10-15 14:35:18,168 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 14:35:18,168 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-15 14:35:18,168 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 14:35:18,168 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 14:35:37,481 epoch 1 - iter 260/2606 - loss 1.70398800 - time (sec): 19.31 - samples/sec: 1954.37 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-15 14:35:56,703 epoch 1 - iter 520/2606 - loss 1.04417068 - time (sec): 38.53 - samples/sec: 1936.56 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-15 14:36:16,384 epoch 1 - iter 780/2606 - loss 0.78595655 - time (sec): 58.22 - samples/sec: 1966.31 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-15 14:36:35,005 epoch 1 - iter 1040/2606 - loss 0.66742617 - time (sec): 76.84 - samples/sec: 1944.65 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-15 14:36:53,448 epoch 1 - iter 1300/2606 - loss 0.58275769 - time (sec): 95.28 - samples/sec: 1954.32 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-15 14:37:12,505 epoch 1 - iter 1560/2606 - loss 0.52348496 - time (sec): 114.34 - samples/sec: 1950.56 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-15 14:37:31,309 epoch 1 - iter 1820/2606 - loss 0.47828591 - time (sec): 133.14 - samples/sec: 1943.65 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-15 14:37:48,903 epoch 1 - iter 2080/2606 - loss 0.44960989 - time (sec): 150.73 - samples/sec: 1945.45 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-15 14:38:07,725 epoch 1 - iter 2340/2606 - loss 0.41952533 - time (sec): 169.56 - samples/sec: 1948.33 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-15 14:38:26,224 epoch 1 - iter 2600/2606 - loss 0.39643307 - time (sec): 188.06 - samples/sec: 1948.36 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-15 14:38:26,731 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 14:38:26,731 EPOCH 1 done: loss 0.3961 - lr: 0.000050
88
+ 2023-10-15 14:38:32,575 DEV : loss 0.1703825443983078 - f1-score (micro avg) 0.2723
89
+ 2023-10-15 14:38:32,602 saving best model
90
+ 2023-10-15 14:38:32,932 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 14:38:51,535 epoch 2 - iter 260/2606 - loss 0.15333555 - time (sec): 18.60 - samples/sec: 1881.05 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-15 14:39:10,089 epoch 2 - iter 520/2606 - loss 0.15721909 - time (sec): 37.16 - samples/sec: 1918.36 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-15 14:39:29,112 epoch 2 - iter 780/2606 - loss 0.16754978 - time (sec): 56.18 - samples/sec: 1938.67 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-15 14:39:47,893 epoch 2 - iter 1040/2606 - loss 0.16366363 - time (sec): 74.96 - samples/sec: 1932.23 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-15 14:40:06,293 epoch 2 - iter 1300/2606 - loss 0.15907525 - time (sec): 93.36 - samples/sec: 1946.72 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-15 14:40:25,920 epoch 2 - iter 1560/2606 - loss 0.15929231 - time (sec): 112.99 - samples/sec: 1946.11 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-15 14:40:44,848 epoch 2 - iter 1820/2606 - loss 0.16228991 - time (sec): 131.91 - samples/sec: 1940.93 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-15 14:41:03,759 epoch 2 - iter 2080/2606 - loss 0.15903749 - time (sec): 150.83 - samples/sec: 1951.85 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-15 14:41:22,416 epoch 2 - iter 2340/2606 - loss 0.15780880 - time (sec): 169.48 - samples/sec: 1948.20 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-15 14:41:40,791 epoch 2 - iter 2600/2606 - loss 0.15594055 - time (sec): 187.86 - samples/sec: 1950.59 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-15 14:41:41,236 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 14:41:41,236 EPOCH 2 done: loss 0.1559 - lr: 0.000044
103
+ 2023-10-15 14:41:50,342 DEV : loss 0.15193568170070648 - f1-score (micro avg) 0.3847
104
+ 2023-10-15 14:41:50,371 saving best model
105
+ 2023-10-15 14:41:50,831 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 14:42:10,499 epoch 3 - iter 260/2606 - loss 0.10386961 - time (sec): 19.66 - samples/sec: 1969.96 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-15 14:42:29,883 epoch 3 - iter 520/2606 - loss 0.10697405 - time (sec): 39.05 - samples/sec: 1984.47 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-15 14:42:48,549 epoch 3 - iter 780/2606 - loss 0.11066127 - time (sec): 57.71 - samples/sec: 1984.71 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-15 14:43:06,906 epoch 3 - iter 1040/2606 - loss 0.11236558 - time (sec): 76.07 - samples/sec: 1974.96 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-15 14:43:26,140 epoch 3 - iter 1300/2606 - loss 0.10971979 - time (sec): 95.31 - samples/sec: 1975.49 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-15 14:43:44,182 epoch 3 - iter 1560/2606 - loss 0.10942417 - time (sec): 113.35 - samples/sec: 1964.00 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-15 14:44:02,842 epoch 3 - iter 1820/2606 - loss 0.11037765 - time (sec): 132.01 - samples/sec: 1952.43 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-15 14:44:21,244 epoch 3 - iter 2080/2606 - loss 0.11002920 - time (sec): 150.41 - samples/sec: 1947.68 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-15 14:44:40,649 epoch 3 - iter 2340/2606 - loss 0.10906780 - time (sec): 169.81 - samples/sec: 1944.70 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-15 14:44:58,744 epoch 3 - iter 2600/2606 - loss 0.10926208 - time (sec): 187.91 - samples/sec: 1947.44 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-15 14:44:59,407 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 14:44:59,407 EPOCH 3 done: loss 0.1090 - lr: 0.000039
118
+ 2023-10-15 14:45:08,482 DEV : loss 0.25894129276275635 - f1-score (micro avg) 0.3382
119
+ 2023-10-15 14:45:08,509 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 14:45:28,158 epoch 4 - iter 260/2606 - loss 0.07515214 - time (sec): 19.65 - samples/sec: 1967.37 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-15 14:45:46,029 epoch 4 - iter 520/2606 - loss 0.07733261 - time (sec): 37.52 - samples/sec: 1927.84 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-15 14:46:04,834 epoch 4 - iter 780/2606 - loss 0.07656362 - time (sec): 56.32 - samples/sec: 1944.62 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-15 14:46:23,015 epoch 4 - iter 1040/2606 - loss 0.07700269 - time (sec): 74.50 - samples/sec: 1947.83 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-15 14:46:41,548 epoch 4 - iter 1300/2606 - loss 0.07567123 - time (sec): 93.04 - samples/sec: 1954.80 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-15 14:47:00,617 epoch 4 - iter 1560/2606 - loss 0.07597083 - time (sec): 112.11 - samples/sec: 1956.79 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-15 14:47:19,328 epoch 4 - iter 1820/2606 - loss 0.07763210 - time (sec): 130.82 - samples/sec: 1952.80 - lr: 0.000035 - momentum: 0.000000
127
+ 2023-10-15 14:47:37,968 epoch 4 - iter 2080/2606 - loss 0.07854227 - time (sec): 149.46 - samples/sec: 1945.80 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-15 14:47:57,679 epoch 4 - iter 2340/2606 - loss 0.07872444 - time (sec): 169.17 - samples/sec: 1944.39 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-15 14:48:16,986 epoch 4 - iter 2600/2606 - loss 0.07824603 - time (sec): 188.48 - samples/sec: 1945.35 - lr: 0.000033 - momentum: 0.000000
130
+ 2023-10-15 14:48:17,398 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 14:48:17,398 EPOCH 4 done: loss 0.0782 - lr: 0.000033
132
+ 2023-10-15 14:48:26,390 DEV : loss 0.2620859742164612 - f1-score (micro avg) 0.3624
133
+ 2023-10-15 14:48:26,417 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-15 14:48:44,955 epoch 5 - iter 260/2606 - loss 0.05400958 - time (sec): 18.54 - samples/sec: 1982.06 - lr: 0.000033 - momentum: 0.000000
135
+ 2023-10-15 14:49:05,767 epoch 5 - iter 520/2606 - loss 0.05211167 - time (sec): 39.35 - samples/sec: 1984.42 - lr: 0.000032 - momentum: 0.000000
136
+ 2023-10-15 14:49:24,232 epoch 5 - iter 780/2606 - loss 0.05301335 - time (sec): 57.81 - samples/sec: 1966.71 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-15 14:49:43,982 epoch 5 - iter 1040/2606 - loss 0.05569211 - time (sec): 77.56 - samples/sec: 1963.80 - lr: 0.000031 - momentum: 0.000000
138
+ 2023-10-15 14:50:02,076 epoch 5 - iter 1300/2606 - loss 0.05528294 - time (sec): 95.66 - samples/sec: 1962.93 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-15 14:50:21,420 epoch 5 - iter 1560/2606 - loss 0.05556370 - time (sec): 115.00 - samples/sec: 1962.65 - lr: 0.000030 - momentum: 0.000000
140
+ 2023-10-15 14:50:40,077 epoch 5 - iter 1820/2606 - loss 0.05564560 - time (sec): 133.66 - samples/sec: 1948.82 - lr: 0.000029 - momentum: 0.000000
141
+ 2023-10-15 14:50:58,218 epoch 5 - iter 2080/2606 - loss 0.05629388 - time (sec): 151.80 - samples/sec: 1948.56 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-15 14:51:17,603 epoch 5 - iter 2340/2606 - loss 0.05874671 - time (sec): 171.18 - samples/sec: 1947.78 - lr: 0.000028 - momentum: 0.000000
143
+ 2023-10-15 14:51:35,349 epoch 5 - iter 2600/2606 - loss 0.05895855 - time (sec): 188.93 - samples/sec: 1941.44 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-15 14:51:35,736 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-15 14:51:35,737 EPOCH 5 done: loss 0.0589 - lr: 0.000028
146
+ 2023-10-15 14:51:43,968 DEV : loss 0.3398889899253845 - f1-score (micro avg) 0.3363
147
+ 2023-10-15 14:51:43,998 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-15 14:52:04,548 epoch 6 - iter 260/2606 - loss 0.03389900 - time (sec): 20.55 - samples/sec: 1875.95 - lr: 0.000027 - momentum: 0.000000
149
+ 2023-10-15 14:52:22,918 epoch 6 - iter 520/2606 - loss 0.04212432 - time (sec): 38.92 - samples/sec: 1912.60 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-15 14:52:41,566 epoch 6 - iter 780/2606 - loss 0.04368331 - time (sec): 57.57 - samples/sec: 1937.32 - lr: 0.000026 - momentum: 0.000000
151
+ 2023-10-15 14:52:59,861 epoch 6 - iter 1040/2606 - loss 0.04514083 - time (sec): 75.86 - samples/sec: 1920.95 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-15 14:53:19,505 epoch 6 - iter 1300/2606 - loss 0.04442191 - time (sec): 95.51 - samples/sec: 1925.06 - lr: 0.000025 - momentum: 0.000000
153
+ 2023-10-15 14:53:38,680 epoch 6 - iter 1560/2606 - loss 0.04517608 - time (sec): 114.68 - samples/sec: 1930.25 - lr: 0.000024 - momentum: 0.000000
154
+ 2023-10-15 14:53:59,139 epoch 6 - iter 1820/2606 - loss 0.04366513 - time (sec): 135.14 - samples/sec: 1934.69 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-15 14:54:17,058 epoch 6 - iter 2080/2606 - loss 0.04407085 - time (sec): 153.06 - samples/sec: 1930.85 - lr: 0.000023 - momentum: 0.000000
156
+ 2023-10-15 14:54:35,685 epoch 6 - iter 2340/2606 - loss 0.04345890 - time (sec): 171.69 - samples/sec: 1930.29 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-15 14:54:54,040 epoch 6 - iter 2600/2606 - loss 0.04335818 - time (sec): 190.04 - samples/sec: 1928.19 - lr: 0.000022 - momentum: 0.000000
158
+ 2023-10-15 14:54:54,516 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-15 14:54:54,517 EPOCH 6 done: loss 0.0432 - lr: 0.000022
160
+ 2023-10-15 14:55:03,040 DEV : loss 0.3315297067165375 - f1-score (micro avg) 0.3722
161
+ 2023-10-15 14:55:03,072 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-15 14:55:21,053 epoch 7 - iter 260/2606 - loss 0.02363314 - time (sec): 17.98 - samples/sec: 1876.09 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-15 14:55:39,507 epoch 7 - iter 520/2606 - loss 0.02903783 - time (sec): 36.43 - samples/sec: 1901.90 - lr: 0.000021 - momentum: 0.000000
164
+ 2023-10-15 14:55:59,059 epoch 7 - iter 780/2606 - loss 0.03146854 - time (sec): 55.99 - samples/sec: 1930.42 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-15 14:56:17,964 epoch 7 - iter 1040/2606 - loss 0.03278017 - time (sec): 74.89 - samples/sec: 1903.68 - lr: 0.000020 - momentum: 0.000000
166
+ 2023-10-15 14:56:35,796 epoch 7 - iter 1300/2606 - loss 0.03388944 - time (sec): 92.72 - samples/sec: 1906.56 - lr: 0.000019 - momentum: 0.000000
167
+ 2023-10-15 14:56:54,161 epoch 7 - iter 1560/2606 - loss 0.03255298 - time (sec): 111.09 - samples/sec: 1924.34 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-15 14:57:13,152 epoch 7 - iter 1820/2606 - loss 0.03196894 - time (sec): 130.08 - samples/sec: 1927.44 - lr: 0.000018 - momentum: 0.000000
169
+ 2023-10-15 14:57:32,652 epoch 7 - iter 2080/2606 - loss 0.03065448 - time (sec): 149.58 - samples/sec: 1936.22 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-15 14:57:52,137 epoch 7 - iter 2340/2606 - loss 0.02973814 - time (sec): 169.06 - samples/sec: 1942.74 - lr: 0.000017 - momentum: 0.000000
171
+ 2023-10-15 14:58:11,366 epoch 7 - iter 2600/2606 - loss 0.03090218 - time (sec): 188.29 - samples/sec: 1943.06 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-15 14:58:12,154 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-15 14:58:12,154 EPOCH 7 done: loss 0.0309 - lr: 0.000017
174
+ 2023-10-15 14:58:20,444 DEV : loss 0.4454517364501953 - f1-score (micro avg) 0.362
175
+ 2023-10-15 14:58:20,472 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-15 14:58:39,045 epoch 8 - iter 260/2606 - loss 0.01715014 - time (sec): 18.57 - samples/sec: 2053.83 - lr: 0.000016 - momentum: 0.000000
177
+ 2023-10-15 14:58:57,659 epoch 8 - iter 520/2606 - loss 0.01871143 - time (sec): 37.19 - samples/sec: 2016.24 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-15 14:59:15,612 epoch 8 - iter 780/2606 - loss 0.02054380 - time (sec): 55.14 - samples/sec: 1967.69 - lr: 0.000015 - momentum: 0.000000
179
+ 2023-10-15 14:59:34,079 epoch 8 - iter 1040/2606 - loss 0.02016721 - time (sec): 73.61 - samples/sec: 1969.24 - lr: 0.000014 - momentum: 0.000000
180
+ 2023-10-15 14:59:51,734 epoch 8 - iter 1300/2606 - loss 0.02166010 - time (sec): 91.26 - samples/sec: 1959.92 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-15 15:00:10,857 epoch 8 - iter 1560/2606 - loss 0.02170611 - time (sec): 110.38 - samples/sec: 1952.41 - lr: 0.000013 - momentum: 0.000000
182
+ 2023-10-15 15:00:31,030 epoch 8 - iter 1820/2606 - loss 0.02156229 - time (sec): 130.56 - samples/sec: 1959.10 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-15 15:00:50,568 epoch 8 - iter 2080/2606 - loss 0.02342631 - time (sec): 150.09 - samples/sec: 1959.29 - lr: 0.000012 - momentum: 0.000000
184
+ 2023-10-15 15:01:08,797 epoch 8 - iter 2340/2606 - loss 0.02326894 - time (sec): 168.32 - samples/sec: 1958.13 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-15 15:01:27,579 epoch 8 - iter 2600/2606 - loss 0.02326670 - time (sec): 187.11 - samples/sec: 1958.63 - lr: 0.000011 - momentum: 0.000000
186
+ 2023-10-15 15:01:28,005 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-15 15:01:28,005 EPOCH 8 done: loss 0.0233 - lr: 0.000011
188
+ 2023-10-15 15:01:36,293 DEV : loss 0.5098573565483093 - f1-score (micro avg) 0.3418
189
+ 2023-10-15 15:01:36,321 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-15 15:01:53,995 epoch 9 - iter 260/2606 - loss 0.01668464 - time (sec): 17.67 - samples/sec: 1958.94 - lr: 0.000011 - momentum: 0.000000
191
+ 2023-10-15 15:02:11,655 epoch 9 - iter 520/2606 - loss 0.01529131 - time (sec): 35.33 - samples/sec: 1937.99 - lr: 0.000010 - momentum: 0.000000
192
+ 2023-10-15 15:02:31,960 epoch 9 - iter 780/2606 - loss 0.01417848 - time (sec): 55.64 - samples/sec: 1944.47 - lr: 0.000009 - momentum: 0.000000
193
+ 2023-10-15 15:02:51,522 epoch 9 - iter 1040/2606 - loss 0.01420376 - time (sec): 75.20 - samples/sec: 1930.86 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-15 15:03:11,085 epoch 9 - iter 1300/2606 - loss 0.01442784 - time (sec): 94.76 - samples/sec: 1931.52 - lr: 0.000008 - momentum: 0.000000
195
+ 2023-10-15 15:03:30,402 epoch 9 - iter 1560/2606 - loss 0.01622053 - time (sec): 114.08 - samples/sec: 1925.97 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-15 15:03:49,697 epoch 9 - iter 1820/2606 - loss 0.01569716 - time (sec): 133.37 - samples/sec: 1929.46 - lr: 0.000007 - momentum: 0.000000
197
+ 2023-10-15 15:04:08,230 epoch 9 - iter 2080/2606 - loss 0.01571005 - time (sec): 151.91 - samples/sec: 1933.51 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-15 15:04:28,322 epoch 9 - iter 2340/2606 - loss 0.01553269 - time (sec): 172.00 - samples/sec: 1921.80 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-15 15:04:47,584 epoch 9 - iter 2600/2606 - loss 0.01532827 - time (sec): 191.26 - samples/sec: 1919.20 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-15 15:04:47,880 ----------------------------------------------------------------------------------------------------
201
+ 2023-10-15 15:04:47,880 EPOCH 9 done: loss 0.0153 - lr: 0.000006
202
+ 2023-10-15 15:04:56,163 DEV : loss 0.4624152183532715 - f1-score (micro avg) 0.3618
203
+ 2023-10-15 15:04:56,192 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-15 15:05:15,338 epoch 10 - iter 260/2606 - loss 0.00882181 - time (sec): 19.14 - samples/sec: 1930.56 - lr: 0.000005 - momentum: 0.000000
205
+ 2023-10-15 15:05:34,619 epoch 10 - iter 520/2606 - loss 0.01164627 - time (sec): 38.43 - samples/sec: 1921.16 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-15 15:05:54,353 epoch 10 - iter 780/2606 - loss 0.01191395 - time (sec): 58.16 - samples/sec: 1945.73 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-15 15:06:13,183 epoch 10 - iter 1040/2606 - loss 0.01088257 - time (sec): 76.99 - samples/sec: 1947.12 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-15 15:06:30,946 epoch 10 - iter 1300/2606 - loss 0.01110488 - time (sec): 94.75 - samples/sec: 1943.64 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-15 15:06:49,375 epoch 10 - iter 1560/2606 - loss 0.01164780 - time (sec): 113.18 - samples/sec: 1944.76 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-15 15:07:09,170 epoch 10 - iter 1820/2606 - loss 0.01131708 - time (sec): 132.98 - samples/sec: 1941.93 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-15 15:07:27,445 epoch 10 - iter 2080/2606 - loss 0.01073956 - time (sec): 151.25 - samples/sec: 1940.29 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-15 15:07:46,629 epoch 10 - iter 2340/2606 - loss 0.01037212 - time (sec): 170.44 - samples/sec: 1938.05 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 15:08:05,232 epoch 10 - iter 2600/2606 - loss 0.01025554 - time (sec): 189.04 - samples/sec: 1937.16 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-15 15:08:05,747 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-15 15:08:05,747 EPOCH 10 done: loss 0.0102 - lr: 0.000000
216
+ 2023-10-15 15:08:14,814 DEV : loss 0.4725865125656128 - f1-score (micro avg) 0.3743
217
+ 2023-10-15 15:08:15,190 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-15 15:08:15,191 Loading model from best epoch ...
219
+ 2023-10-15 15:08:16,646 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 15:08:31,788
221
+ Results:
222
+ - F-score (micro) 0.4127
223
+ - F-score (macro) 0.2811
224
+ - Accuracy 0.2633
225
+
226
+ By class:
227
+ precision recall f1-score support
228
+
229
+ LOC 0.5280 0.4745 0.4998 1214
230
+ PER 0.3436 0.4295 0.3817 808
231
+ ORG 0.2143 0.2805 0.2429 353
232
+ HumanProd 0.0000 0.0000 0.0000 15
233
+
234
+ micro avg 0.3988 0.4276 0.4127 2390
235
+ macro avg 0.2715 0.2961 0.2811 2390
236
+ weighted avg 0.4160 0.4276 0.4188 2390
237
+
238
+ 2023-10-15 15:08:31,788 ----------------------------------------------------------------------------------------------------