Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbert_tiny-bs4-wsFalse-e10-lr5e-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
888acf8
·
1 Parent(s): e7aa5b3

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.1697652226.46dc0c540dd0.2878.9 +3 -0
  5. test.tsv +0 -0
  6. training.log +248 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92e8b2bbfcfed38b6cc959575bca2aed0fcfdd6947bf3428d74cba3cca7f03dc
3
+ size 19050210
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 18:04:02 0.0000 1.2137 0.4263 0.0000 0.0000 0.0000 0.0000
3
+ 2 18:04:21 0.0000 0.4482 0.3477 0.4135 0.1720 0.2430 0.1420
4
+ 3 18:04:41 0.0000 0.3755 0.3211 0.3842 0.2658 0.3142 0.1934
5
+ 4 18:05:01 0.0000 0.3311 0.3120 0.3658 0.2909 0.3240 0.2018
6
+ 5 18:05:19 0.0000 0.3034 0.3089 0.3601 0.3120 0.3343 0.2098
7
+ 6 18:05:39 0.0000 0.2806 0.3111 0.3686 0.3190 0.3420 0.2160
8
+ 7 18:05:58 0.0000 0.2655 0.3076 0.3592 0.3432 0.3511 0.2230
9
+ 8 18:06:18 0.0000 0.2508 0.3026 0.3631 0.3597 0.3614 0.2321
10
+ 9 18:06:37 0.0000 0.2480 0.3070 0.3801 0.3518 0.3654 0.2350
11
+ 10 18:06:56 0.0000 0.2399 0.3075 0.3768 0.3550 0.3655 0.2354
runs/events.out.tfevents.1697652226.46dc0c540dd0.2878.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da2c6b1965f8ace6e95239b67b7b8232c7df437980e9ffdae4f2a4717b6903b7
3
+ size 502124
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,248 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 18:03:46,090 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 18:03:46,091 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), 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=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=21, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-18 18:03:46,091 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 18:03:46,091 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
52
+ - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
53
+ 2023-10-18 18:03:46,091 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 18:03:46,091 Train: 3575 sentences
55
+ 2023-10-18 18:03:46,091 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 18:03:46,091 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 18:03:46,091 Training Params:
58
+ 2023-10-18 18:03:46,091 - learning_rate: "5e-05"
59
+ 2023-10-18 18:03:46,091 - mini_batch_size: "4"
60
+ 2023-10-18 18:03:46,091 - max_epochs: "10"
61
+ 2023-10-18 18:03:46,091 - shuffle: "True"
62
+ 2023-10-18 18:03:46,091 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 18:03:46,091 Plugins:
64
+ 2023-10-18 18:03:46,091 - TensorboardLogger
65
+ 2023-10-18 18:03:46,091 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 18:03:46,091 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 18:03:46,091 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 18:03:46,091 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 18:03:46,091 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 18:03:46,092 Computation:
71
+ 2023-10-18 18:03:46,092 - compute on device: cuda:0
72
+ 2023-10-18 18:03:46,092 - embedding storage: none
73
+ 2023-10-18 18:03:46,092 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 18:03:46,092 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
75
+ 2023-10-18 18:03:46,092 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 18:03:46,092 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 18:03:46,092 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 18:03:47,533 epoch 1 - iter 89/894 - loss 3.43498959 - time (sec): 1.44 - samples/sec: 5610.49 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-18 18:03:48,952 epoch 1 - iter 178/894 - loss 3.08753148 - time (sec): 2.86 - samples/sec: 5738.36 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-18 18:03:50,382 epoch 1 - iter 267/894 - loss 2.60861970 - time (sec): 4.29 - samples/sec: 5938.06 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-18 18:03:51,817 epoch 1 - iter 356/894 - loss 2.13729872 - time (sec): 5.72 - samples/sec: 6025.76 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-18 18:03:53,195 epoch 1 - iter 445/894 - loss 1.85618622 - time (sec): 7.10 - samples/sec: 6050.11 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-18 18:03:54,589 epoch 1 - iter 534/894 - loss 1.65953709 - time (sec): 8.50 - samples/sec: 6001.19 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-18 18:03:55,981 epoch 1 - iter 623/894 - loss 1.50846636 - time (sec): 9.89 - samples/sec: 5992.96 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-18 18:03:57,429 epoch 1 - iter 712/894 - loss 1.38521735 - time (sec): 11.34 - samples/sec: 6015.28 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-18 18:03:58,878 epoch 1 - iter 801/894 - loss 1.28793964 - time (sec): 12.79 - samples/sec: 6076.01 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-18 18:04:00,261 epoch 1 - iter 890/894 - loss 1.21448172 - time (sec): 14.17 - samples/sec: 6079.72 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-18 18:04:00,318 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 18:04:00,318 EPOCH 1 done: loss 1.2137 - lr: 0.000050
90
+ 2023-10-18 18:04:02,580 DEV : loss 0.4263085424900055 - f1-score (micro avg) 0.0
91
+ 2023-10-18 18:04:02,608 ----------------------------------------------------------------------------------------------------
92
+ 2023-10-18 18:04:03,972 epoch 2 - iter 89/894 - loss 0.49713747 - time (sec): 1.36 - samples/sec: 6507.15 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-18 18:04:05,345 epoch 2 - iter 178/894 - loss 0.49185097 - time (sec): 2.74 - samples/sec: 6351.21 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-18 18:04:06,690 epoch 2 - iter 267/894 - loss 0.48800476 - time (sec): 4.08 - samples/sec: 6272.11 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-18 18:04:08,046 epoch 2 - iter 356/894 - loss 0.48445397 - time (sec): 5.44 - samples/sec: 6185.52 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-18 18:04:09,394 epoch 2 - iter 445/894 - loss 0.48238418 - time (sec): 6.79 - samples/sec: 6194.15 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-18 18:04:10,800 epoch 2 - iter 534/894 - loss 0.46700383 - time (sec): 8.19 - samples/sec: 6204.76 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-18 18:04:12,210 epoch 2 - iter 623/894 - loss 0.46662532 - time (sec): 9.60 - samples/sec: 6141.59 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-18 18:04:13,680 epoch 2 - iter 712/894 - loss 0.45539541 - time (sec): 11.07 - samples/sec: 6208.94 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-18 18:04:15,109 epoch 2 - iter 801/894 - loss 0.45446600 - time (sec): 12.50 - samples/sec: 6221.45 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-18 18:04:16,525 epoch 2 - iter 890/894 - loss 0.44765225 - time (sec): 13.92 - samples/sec: 6199.65 - lr: 0.000044 - momentum: 0.000000
102
+ 2023-10-18 18:04:16,582 ----------------------------------------------------------------------------------------------------
103
+ 2023-10-18 18:04:16,582 EPOCH 2 done: loss 0.4482 - lr: 0.000044
104
+ 2023-10-18 18:04:21,878 DEV : loss 0.34765392541885376 - f1-score (micro avg) 0.243
105
+ 2023-10-18 18:04:21,905 saving best model
106
+ 2023-10-18 18:04:21,941 ----------------------------------------------------------------------------------------------------
107
+ 2023-10-18 18:04:23,330 epoch 3 - iter 89/894 - loss 0.41805770 - time (sec): 1.39 - samples/sec: 5959.90 - lr: 0.000044 - momentum: 0.000000
108
+ 2023-10-18 18:04:24,759 epoch 3 - iter 178/894 - loss 0.43276993 - time (sec): 2.82 - samples/sec: 6072.98 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-18 18:04:26,230 epoch 3 - iter 267/894 - loss 0.41738376 - time (sec): 4.29 - samples/sec: 6161.21 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-18 18:04:27,684 epoch 3 - iter 356/894 - loss 0.41407899 - time (sec): 5.74 - samples/sec: 6145.33 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-18 18:04:29,116 epoch 3 - iter 445/894 - loss 0.40637262 - time (sec): 7.17 - samples/sec: 6093.57 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-18 18:04:30,550 epoch 3 - iter 534/894 - loss 0.38982328 - time (sec): 8.61 - samples/sec: 6103.09 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-18 18:04:32,011 epoch 3 - iter 623/894 - loss 0.38362119 - time (sec): 10.07 - samples/sec: 6074.46 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-18 18:04:33,369 epoch 3 - iter 712/894 - loss 0.37912964 - time (sec): 11.43 - samples/sec: 6056.07 - lr: 0.000040 - momentum: 0.000000
115
+ 2023-10-18 18:04:34,763 epoch 3 - iter 801/894 - loss 0.37858686 - time (sec): 12.82 - samples/sec: 6078.61 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-18 18:04:36,131 epoch 3 - iter 890/894 - loss 0.37562114 - time (sec): 14.19 - samples/sec: 6068.41 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-18 18:04:36,190 ----------------------------------------------------------------------------------------------------
118
+ 2023-10-18 18:04:36,190 EPOCH 3 done: loss 0.3755 - lr: 0.000039
119
+ 2023-10-18 18:04:41,515 DEV : loss 0.3211207091808319 - f1-score (micro avg) 0.3142
120
+ 2023-10-18 18:04:41,542 saving best model
121
+ 2023-10-18 18:04:41,581 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-18 18:04:42,998 epoch 4 - iter 89/894 - loss 0.32635474 - time (sec): 1.42 - samples/sec: 6060.62 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-18 18:04:44,429 epoch 4 - iter 178/894 - loss 0.35437975 - time (sec): 2.85 - samples/sec: 6088.53 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-18 18:04:45,831 epoch 4 - iter 267/894 - loss 0.35089723 - time (sec): 4.25 - samples/sec: 6157.04 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-18 18:04:47,232 epoch 4 - iter 356/894 - loss 0.33648000 - time (sec): 5.65 - samples/sec: 6280.87 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-18 18:04:48,680 epoch 4 - iter 445/894 - loss 0.33816098 - time (sec): 7.10 - samples/sec: 6355.06 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-18 18:04:50,092 epoch 4 - iter 534/894 - loss 0.34098676 - time (sec): 8.51 - samples/sec: 6224.61 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-18 18:04:51,451 epoch 4 - iter 623/894 - loss 0.33157366 - time (sec): 9.87 - samples/sec: 6234.07 - lr: 0.000035 - momentum: 0.000000
129
+ 2023-10-18 18:04:52,819 epoch 4 - iter 712/894 - loss 0.33569841 - time (sec): 11.24 - samples/sec: 6211.82 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-18 18:04:54,194 epoch 4 - iter 801/894 - loss 0.33458161 - time (sec): 12.61 - samples/sec: 6141.90 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-18 18:04:55,599 epoch 4 - iter 890/894 - loss 0.33097427 - time (sec): 14.02 - samples/sec: 6154.52 - lr: 0.000033 - momentum: 0.000000
132
+ 2023-10-18 18:04:55,654 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-18 18:04:55,654 EPOCH 4 done: loss 0.3311 - lr: 0.000033
134
+ 2023-10-18 18:05:00,985 DEV : loss 0.3120403289794922 - f1-score (micro avg) 0.324
135
+ 2023-10-18 18:05:01,015 saving best model
136
+ 2023-10-18 18:05:01,058 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-18 18:05:02,483 epoch 5 - iter 89/894 - loss 0.30524145 - time (sec): 1.43 - samples/sec: 5936.06 - lr: 0.000033 - momentum: 0.000000
138
+ 2023-10-18 18:05:03,938 epoch 5 - iter 178/894 - loss 0.30512104 - time (sec): 2.88 - samples/sec: 6353.14 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-18 18:05:05,364 epoch 5 - iter 267/894 - loss 0.31020994 - time (sec): 4.31 - samples/sec: 6345.63 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-18 18:05:06,758 epoch 5 - iter 356/894 - loss 0.30779353 - time (sec): 5.70 - samples/sec: 6257.62 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-18 18:05:08,055 epoch 5 - iter 445/894 - loss 0.31278385 - time (sec): 7.00 - samples/sec: 6241.21 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-18 18:05:09,427 epoch 5 - iter 534/894 - loss 0.31087426 - time (sec): 8.37 - samples/sec: 6201.39 - lr: 0.000030 - momentum: 0.000000
143
+ 2023-10-18 18:05:10,775 epoch 5 - iter 623/894 - loss 0.30915571 - time (sec): 9.72 - samples/sec: 6240.06 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-18 18:05:12,195 epoch 5 - iter 712/894 - loss 0.30863944 - time (sec): 11.14 - samples/sec: 6265.60 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-18 18:05:13,587 epoch 5 - iter 801/894 - loss 0.30557407 - time (sec): 12.53 - samples/sec: 6240.28 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-18 18:05:14,822 epoch 5 - iter 890/894 - loss 0.30351739 - time (sec): 13.76 - samples/sec: 6264.59 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-18 18:05:14,878 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-18 18:05:14,878 EPOCH 5 done: loss 0.3034 - lr: 0.000028
149
+ 2023-10-18 18:05:19,904 DEV : loss 0.30890053510665894 - f1-score (micro avg) 0.3343
150
+ 2023-10-18 18:05:19,931 saving best model
151
+ 2023-10-18 18:05:19,967 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-18 18:05:21,381 epoch 6 - iter 89/894 - loss 0.24665018 - time (sec): 1.41 - samples/sec: 6563.05 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-18 18:05:22,749 epoch 6 - iter 178/894 - loss 0.27338459 - time (sec): 2.78 - samples/sec: 6371.40 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-18 18:05:24,465 epoch 6 - iter 267/894 - loss 0.29310495 - time (sec): 4.50 - samples/sec: 5780.54 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-18 18:05:25,850 epoch 6 - iter 356/894 - loss 0.29432340 - time (sec): 5.88 - samples/sec: 5880.07 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-18 18:05:27,252 epoch 6 - iter 445/894 - loss 0.29662643 - time (sec): 7.28 - samples/sec: 5994.72 - lr: 0.000025 - momentum: 0.000000
157
+ 2023-10-18 18:05:28,571 epoch 6 - iter 534/894 - loss 0.29414115 - time (sec): 8.60 - samples/sec: 6051.60 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-18 18:05:29,952 epoch 6 - iter 623/894 - loss 0.28659247 - time (sec): 9.98 - samples/sec: 6043.06 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-18 18:05:31,338 epoch 6 - iter 712/894 - loss 0.28157192 - time (sec): 11.37 - samples/sec: 6063.04 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-18 18:05:32,733 epoch 6 - iter 801/894 - loss 0.27671928 - time (sec): 12.77 - samples/sec: 6091.72 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-18 18:05:34,087 epoch 6 - iter 890/894 - loss 0.28086399 - time (sec): 14.12 - samples/sec: 6101.38 - lr: 0.000022 - momentum: 0.000000
162
+ 2023-10-18 18:05:34,151 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-18 18:05:34,152 EPOCH 6 done: loss 0.2806 - lr: 0.000022
164
+ 2023-10-18 18:05:39,129 DEV : loss 0.31113117933273315 - f1-score (micro avg) 0.342
165
+ 2023-10-18 18:05:39,156 saving best model
166
+ 2023-10-18 18:05:39,193 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-18 18:05:40,640 epoch 7 - iter 89/894 - loss 0.23315533 - time (sec): 1.45 - samples/sec: 5639.23 - lr: 0.000022 - momentum: 0.000000
168
+ 2023-10-18 18:05:42,111 epoch 7 - iter 178/894 - loss 0.26534915 - time (sec): 2.92 - samples/sec: 6063.91 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-18 18:05:43,499 epoch 7 - iter 267/894 - loss 0.26302261 - time (sec): 4.31 - samples/sec: 6038.79 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-18 18:05:44,942 epoch 7 - iter 356/894 - loss 0.26080884 - time (sec): 5.75 - samples/sec: 6263.67 - lr: 0.000020 - momentum: 0.000000
171
+ 2023-10-18 18:05:46,327 epoch 7 - iter 445/894 - loss 0.26482923 - time (sec): 7.13 - samples/sec: 6225.72 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-18 18:05:47,756 epoch 7 - iter 534/894 - loss 0.26287988 - time (sec): 8.56 - samples/sec: 6286.91 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-18 18:05:49,126 epoch 7 - iter 623/894 - loss 0.26066862 - time (sec): 9.93 - samples/sec: 6195.50 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-18 18:05:50,531 epoch 7 - iter 712/894 - loss 0.26624906 - time (sec): 11.34 - samples/sec: 6200.49 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-18 18:05:51,901 epoch 7 - iter 801/894 - loss 0.26615665 - time (sec): 12.71 - samples/sec: 6142.77 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-18 18:05:53,286 epoch 7 - iter 890/894 - loss 0.26606576 - time (sec): 14.09 - samples/sec: 6113.64 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-18 18:05:53,353 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-18 18:05:53,353 EPOCH 7 done: loss 0.2655 - lr: 0.000017
179
+ 2023-10-18 18:05:58,659 DEV : loss 0.30758559703826904 - f1-score (micro avg) 0.3511
180
+ 2023-10-18 18:05:58,686 saving best model
181
+ 2023-10-18 18:05:58,729 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-18 18:06:00,114 epoch 8 - iter 89/894 - loss 0.24002002 - time (sec): 1.38 - samples/sec: 5734.34 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-18 18:06:01,442 epoch 8 - iter 178/894 - loss 0.24034680 - time (sec): 2.71 - samples/sec: 5644.66 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-18 18:06:02,825 epoch 8 - iter 267/894 - loss 0.24390847 - time (sec): 4.10 - samples/sec: 5864.69 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-18 18:06:04,244 epoch 8 - iter 356/894 - loss 0.25396324 - time (sec): 5.51 - samples/sec: 5878.33 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-18 18:06:05,606 epoch 8 - iter 445/894 - loss 0.24715195 - time (sec): 6.88 - samples/sec: 5878.94 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-18 18:06:06,987 epoch 8 - iter 534/894 - loss 0.24673498 - time (sec): 8.26 - samples/sec: 5870.15 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-18 18:06:08,382 epoch 8 - iter 623/894 - loss 0.24348441 - time (sec): 9.65 - samples/sec: 6027.05 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-18 18:06:09,817 epoch 8 - iter 712/894 - loss 0.25278893 - time (sec): 11.09 - samples/sec: 6105.49 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-18 18:06:11,196 epoch 8 - iter 801/894 - loss 0.25276481 - time (sec): 12.47 - samples/sec: 6088.28 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-18 18:06:12,626 epoch 8 - iter 890/894 - loss 0.25135174 - time (sec): 13.90 - samples/sec: 6126.30 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-18 18:06:12,718 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-18 18:06:12,718 EPOCH 8 done: loss 0.2508 - lr: 0.000011
194
+ 2023-10-18 18:06:18,040 DEV : loss 0.3025902807712555 - f1-score (micro avg) 0.3614
195
+ 2023-10-18 18:06:18,068 saving best model
196
+ 2023-10-18 18:06:18,109 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-18 18:06:19,549 epoch 9 - iter 89/894 - loss 0.26937101 - time (sec): 1.44 - samples/sec: 6843.28 - lr: 0.000011 - momentum: 0.000000
198
+ 2023-10-18 18:06:20,950 epoch 9 - iter 178/894 - loss 0.27973157 - time (sec): 2.84 - samples/sec: 6485.42 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-18 18:06:22,254 epoch 9 - iter 267/894 - loss 0.26986158 - time (sec): 4.14 - samples/sec: 6628.31 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-18 18:06:23,627 epoch 9 - iter 356/894 - loss 0.26383497 - time (sec): 5.52 - samples/sec: 6341.42 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-18 18:06:24,991 epoch 9 - iter 445/894 - loss 0.24995919 - time (sec): 6.88 - samples/sec: 6305.16 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-18 18:06:26,386 epoch 9 - iter 534/894 - loss 0.25955274 - time (sec): 8.28 - samples/sec: 6287.82 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-18 18:06:27,769 epoch 9 - iter 623/894 - loss 0.25421326 - time (sec): 9.66 - samples/sec: 6256.93 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-18 18:06:29,187 epoch 9 - iter 712/894 - loss 0.24978285 - time (sec): 11.08 - samples/sec: 6238.94 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-18 18:06:30,545 epoch 9 - iter 801/894 - loss 0.24720986 - time (sec): 12.44 - samples/sec: 6240.93 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-18 18:06:31,938 epoch 9 - iter 890/894 - loss 0.24696720 - time (sec): 13.83 - samples/sec: 6227.11 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-18 18:06:31,997 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-18 18:06:31,997 EPOCH 9 done: loss 0.2480 - lr: 0.000006
209
+ 2023-10-18 18:06:37,345 DEV : loss 0.30696946382522583 - f1-score (micro avg) 0.3654
210
+ 2023-10-18 18:06:37,372 saving best model
211
+ 2023-10-18 18:06:37,412 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-18 18:06:38,837 epoch 10 - iter 89/894 - loss 0.27149473 - time (sec): 1.42 - samples/sec: 6264.84 - lr: 0.000005 - momentum: 0.000000
213
+ 2023-10-18 18:06:40,241 epoch 10 - iter 178/894 - loss 0.26748895 - time (sec): 2.83 - samples/sec: 6289.23 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-18 18:06:41,607 epoch 10 - iter 267/894 - loss 0.25190229 - time (sec): 4.19 - samples/sec: 6163.76 - lr: 0.000004 - momentum: 0.000000
215
+ 2023-10-18 18:06:42,959 epoch 10 - iter 356/894 - loss 0.25823080 - time (sec): 5.55 - samples/sec: 6079.28 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-18 18:06:44,375 epoch 10 - iter 445/894 - loss 0.25629453 - time (sec): 6.96 - samples/sec: 6145.82 - lr: 0.000003 - momentum: 0.000000
217
+ 2023-10-18 18:06:45,752 epoch 10 - iter 534/894 - loss 0.25065215 - time (sec): 8.34 - samples/sec: 6081.34 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-18 18:06:47,190 epoch 10 - iter 623/894 - loss 0.24408374 - time (sec): 9.78 - samples/sec: 6129.24 - lr: 0.000002 - momentum: 0.000000
219
+ 2023-10-18 18:06:48,545 epoch 10 - iter 712/894 - loss 0.24020067 - time (sec): 11.13 - samples/sec: 6179.73 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-18 18:06:49,811 epoch 10 - iter 801/894 - loss 0.23988736 - time (sec): 12.40 - samples/sec: 6246.35 - lr: 0.000001 - momentum: 0.000000
221
+ 2023-10-18 18:06:51,221 epoch 10 - iter 890/894 - loss 0.24026831 - time (sec): 13.81 - samples/sec: 6243.75 - lr: 0.000000 - momentum: 0.000000
222
+ 2023-10-18 18:06:51,281 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-18 18:06:51,281 EPOCH 10 done: loss 0.2399 - lr: 0.000000
224
+ 2023-10-18 18:06:56,279 DEV : loss 0.30754354596138 - f1-score (micro avg) 0.3655
225
+ 2023-10-18 18:06:56,305 saving best model
226
+ 2023-10-18 18:06:56,371 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-18 18:06:56,372 Loading model from best epoch ...
228
+ 2023-10-18 18:06:56,446 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
229
+ 2023-10-18 18:06:58,721
230
+ Results:
231
+ - F-score (micro) 0.357
232
+ - F-score (macro) 0.1876
233
+ - Accuracy 0.2303
234
+
235
+ By class:
236
+ precision recall f1-score support
237
+
238
+ loc 0.4760 0.5654 0.5169 596
239
+ pers 0.1696 0.2282 0.1946 333
240
+ org 0.5000 0.0076 0.0149 132
241
+ time 0.2500 0.1837 0.2118 49
242
+ prod 0.0000 0.0000 0.0000 66
243
+
244
+ micro avg 0.3543 0.3597 0.3570 1176
245
+ macro avg 0.2791 0.1970 0.1876 1176
246
+ weighted avg 0.3558 0.3597 0.3276 1176
247
+
248
+ 2023-10-18 18:06:58,722 ----------------------------------------------------------------------------------------------------