Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
Dutch
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
34e2f85
•
1 Parent(s): a733f74

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:fe9bad3cb7d9fb7512287efd86937244dc7a9ce9f1795ca7786b796372dbd6a5
3
+ size 443311111
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 11:59:35 0.0000 0.3500 0.1319 0.6568 0.6426 0.6496 0.4886
3
+ 2 12:00:40 0.0000 0.0986 0.0866 0.8151 0.7603 0.7867 0.6589
4
+ 3 12:01:44 0.0000 0.0627 0.0884 0.8027 0.8110 0.8068 0.6929
5
+ 4 12:02:48 0.0000 0.0418 0.0962 0.8274 0.7624 0.7935 0.6746
6
+ 5 12:03:52 0.0000 0.0315 0.1145 0.8156 0.8316 0.8235 0.7137
7
+ 6 12:04:56 0.0000 0.0234 0.1459 0.8320 0.7779 0.8041 0.6858
8
+ 7 12:06:00 0.0000 0.0172 0.1582 0.8793 0.7448 0.8065 0.6860
9
+ 8 12:07:05 0.0000 0.0129 0.1818 0.8590 0.7490 0.8002 0.6795
10
+ 9 12:08:08 0.0000 0.0094 0.1775 0.8460 0.7603 0.8009 0.6809
11
+ 10 12:09:12 0.0000 0.0065 0.1837 0.8527 0.7717 0.8102 0.6936
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 11:58:32,284 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 11:58:32,285 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=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-14 11:58:32,285 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 11:58:32,286 MultiCorpus: 5777 train + 722 dev + 723 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
53
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 11:58:32,286 Train: 5777 sentences
55
+ 2023-10-14 11:58:32,286 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 11:58:32,286 Training Params:
58
+ 2023-10-14 11:58:32,286 - learning_rate: "5e-05"
59
+ 2023-10-14 11:58:32,286 - mini_batch_size: "8"
60
+ 2023-10-14 11:58:32,286 - max_epochs: "10"
61
+ 2023-10-14 11:58:32,286 - shuffle: "True"
62
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 11:58:32,286 Plugins:
64
+ 2023-10-14 11:58:32,286 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 11:58:32,286 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 11:58:32,286 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 11:58:32,286 Computation:
70
+ 2023-10-14 11:58:32,286 - compute on device: cuda:0
71
+ 2023-10-14 11:58:32,286 - embedding storage: none
72
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 11:58:32,286 Model training base path: "hmbench-icdar/nl-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 11:58:32,286 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 11:58:38,229 epoch 1 - iter 72/723 - loss 1.83466572 - time (sec): 5.94 - samples/sec: 3121.04 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-14 11:58:44,358 epoch 1 - iter 144/723 - loss 1.07312952 - time (sec): 12.07 - samples/sec: 2975.76 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-14 11:58:50,195 epoch 1 - iter 216/723 - loss 0.79762119 - time (sec): 17.91 - samples/sec: 2978.27 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-14 11:58:56,375 epoch 1 - iter 288/723 - loss 0.64875789 - time (sec): 24.09 - samples/sec: 2962.96 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-14 11:59:02,475 epoch 1 - iter 360/723 - loss 0.55533977 - time (sec): 30.19 - samples/sec: 2937.76 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-14 11:59:08,437 epoch 1 - iter 432/723 - loss 0.48997863 - time (sec): 36.15 - samples/sec: 2939.05 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-14 11:59:14,448 epoch 1 - iter 504/723 - loss 0.44372217 - time (sec): 42.16 - samples/sec: 2940.61 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-14 11:59:20,095 epoch 1 - iter 576/723 - loss 0.40786476 - time (sec): 47.81 - samples/sec: 2933.54 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-14 11:59:25,851 epoch 1 - iter 648/723 - loss 0.37553601 - time (sec): 53.56 - samples/sec: 2947.47 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-14 11:59:31,836 epoch 1 - iter 720/723 - loss 0.35055959 - time (sec): 59.55 - samples/sec: 2949.53 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-14 11:59:32,042 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 11:59:32,042 EPOCH 1 done: loss 0.3500 - lr: 0.000050
88
+ 2023-10-14 11:59:35,204 DEV : loss 0.13189315795898438 - f1-score (micro avg) 0.6496
89
+ 2023-10-14 11:59:35,220 saving best model
90
+ 2023-10-14 11:59:35,629 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 11:59:41,987 epoch 2 - iter 72/723 - loss 0.11751025 - time (sec): 6.36 - samples/sec: 2691.71 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-14 11:59:47,893 epoch 2 - iter 144/723 - loss 0.11290402 - time (sec): 12.26 - samples/sec: 2784.35 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-14 11:59:54,056 epoch 2 - iter 216/723 - loss 0.10761347 - time (sec): 18.43 - samples/sec: 2844.82 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-14 12:00:00,524 epoch 2 - iter 288/723 - loss 0.10620566 - time (sec): 24.89 - samples/sec: 2848.53 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-14 12:00:06,097 epoch 2 - iter 360/723 - loss 0.10503397 - time (sec): 30.47 - samples/sec: 2887.49 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-14 12:00:11,541 epoch 2 - iter 432/723 - loss 0.10197538 - time (sec): 35.91 - samples/sec: 2916.16 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-14 12:00:17,722 epoch 2 - iter 504/723 - loss 0.10172999 - time (sec): 42.09 - samples/sec: 2906.40 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-14 12:00:24,040 epoch 2 - iter 576/723 - loss 0.09966793 - time (sec): 48.41 - samples/sec: 2896.00 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-14 12:00:30,082 epoch 2 - iter 648/723 - loss 0.09976307 - time (sec): 54.45 - samples/sec: 2905.29 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-14 12:00:35,771 epoch 2 - iter 720/723 - loss 0.09860537 - time (sec): 60.14 - samples/sec: 2922.83 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-14 12:00:35,927 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 12:00:35,928 EPOCH 2 done: loss 0.0986 - lr: 0.000044
103
+ 2023-10-14 12:00:40,278 DEV : loss 0.08656182885169983 - f1-score (micro avg) 0.7867
104
+ 2023-10-14 12:00:40,298 saving best model
105
+ 2023-10-14 12:00:40,971 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 12:00:46,886 epoch 3 - iter 72/723 - loss 0.06819931 - time (sec): 5.91 - samples/sec: 2870.31 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-14 12:00:52,807 epoch 3 - iter 144/723 - loss 0.06728335 - time (sec): 11.83 - samples/sec: 2880.12 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-14 12:00:58,921 epoch 3 - iter 216/723 - loss 0.06427625 - time (sec): 17.95 - samples/sec: 2851.72 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-14 12:01:04,551 epoch 3 - iter 288/723 - loss 0.06626178 - time (sec): 23.58 - samples/sec: 2881.51 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-14 12:01:10,181 epoch 3 - iter 360/723 - loss 0.06476197 - time (sec): 29.21 - samples/sec: 2886.60 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-14 12:01:16,383 epoch 3 - iter 432/723 - loss 0.06317834 - time (sec): 35.41 - samples/sec: 2915.82 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-14 12:01:22,271 epoch 3 - iter 504/723 - loss 0.06316978 - time (sec): 41.30 - samples/sec: 2915.94 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-14 12:01:28,315 epoch 3 - iter 576/723 - loss 0.06518816 - time (sec): 47.34 - samples/sec: 2938.91 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-14 12:01:34,586 epoch 3 - iter 648/723 - loss 0.06373533 - time (sec): 53.61 - samples/sec: 2926.35 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-14 12:01:40,894 epoch 3 - iter 720/723 - loss 0.06257771 - time (sec): 59.92 - samples/sec: 2933.69 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-14 12:01:41,060 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 12:01:41,061 EPOCH 3 done: loss 0.0627 - lr: 0.000039
118
+ 2023-10-14 12:01:44,703 DEV : loss 0.08839963376522064 - f1-score (micro avg) 0.8068
119
+ 2023-10-14 12:01:44,730 saving best model
120
+ 2023-10-14 12:01:45,274 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 12:01:51,166 epoch 4 - iter 72/723 - loss 0.03595879 - time (sec): 5.89 - samples/sec: 2878.82 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-14 12:01:57,619 epoch 4 - iter 144/723 - loss 0.03883693 - time (sec): 12.34 - samples/sec: 2904.68 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-14 12:02:03,480 epoch 4 - iter 216/723 - loss 0.03762509 - time (sec): 18.20 - samples/sec: 2912.16 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-14 12:02:09,721 epoch 4 - iter 288/723 - loss 0.04213370 - time (sec): 24.44 - samples/sec: 2891.27 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-14 12:02:15,341 epoch 4 - iter 360/723 - loss 0.04229835 - time (sec): 30.06 - samples/sec: 2907.80 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-14 12:02:21,491 epoch 4 - iter 432/723 - loss 0.04293415 - time (sec): 36.21 - samples/sec: 2900.78 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-14 12:02:27,520 epoch 4 - iter 504/723 - loss 0.04229960 - time (sec): 42.24 - samples/sec: 2913.46 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-14 12:02:33,219 epoch 4 - iter 576/723 - loss 0.04154991 - time (sec): 47.94 - samples/sec: 2914.53 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-14 12:02:39,129 epoch 4 - iter 648/723 - loss 0.04153434 - time (sec): 53.85 - samples/sec: 2930.31 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-14 12:02:45,170 epoch 4 - iter 720/723 - loss 0.04184976 - time (sec): 59.89 - samples/sec: 2935.18 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-14 12:02:45,325 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 12:02:45,325 EPOCH 4 done: loss 0.0418 - lr: 0.000033
133
+ 2023-10-14 12:02:48,939 DEV : loss 0.09616296738386154 - f1-score (micro avg) 0.7935
134
+ 2023-10-14 12:02:48,964 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-14 12:02:54,782 epoch 5 - iter 72/723 - loss 0.02118830 - time (sec): 5.82 - samples/sec: 2858.88 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-14 12:03:00,634 epoch 5 - iter 144/723 - loss 0.02600065 - time (sec): 11.67 - samples/sec: 2863.50 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-14 12:03:06,570 epoch 5 - iter 216/723 - loss 0.02863738 - time (sec): 17.60 - samples/sec: 2872.70 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-14 12:03:12,682 epoch 5 - iter 288/723 - loss 0.02992902 - time (sec): 23.72 - samples/sec: 2916.41 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-14 12:03:18,848 epoch 5 - iter 360/723 - loss 0.03303827 - time (sec): 29.88 - samples/sec: 2917.46 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-14 12:03:24,875 epoch 5 - iter 432/723 - loss 0.03302250 - time (sec): 35.91 - samples/sec: 2933.74 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-14 12:03:31,252 epoch 5 - iter 504/723 - loss 0.03221675 - time (sec): 42.29 - samples/sec: 2935.50 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-14 12:03:36,940 epoch 5 - iter 576/723 - loss 0.03142146 - time (sec): 47.97 - samples/sec: 2936.47 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-14 12:03:42,493 epoch 5 - iter 648/723 - loss 0.03036131 - time (sec): 53.53 - samples/sec: 2952.83 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-14 12:03:48,304 epoch 5 - iter 720/723 - loss 0.03141318 - time (sec): 59.34 - samples/sec: 2955.78 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-14 12:03:48,556 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-14 12:03:48,556 EPOCH 5 done: loss 0.0315 - lr: 0.000028
147
+ 2023-10-14 12:03:52,530 DEV : loss 0.11451639235019684 - f1-score (micro avg) 0.8235
148
+ 2023-10-14 12:03:52,550 saving best model
149
+ 2023-10-14 12:03:53,116 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-14 12:03:59,048 epoch 6 - iter 72/723 - loss 0.02304111 - time (sec): 5.93 - samples/sec: 2851.22 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-14 12:04:04,697 epoch 6 - iter 144/723 - loss 0.02294005 - time (sec): 11.58 - samples/sec: 2956.84 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-14 12:04:10,873 epoch 6 - iter 216/723 - loss 0.02326946 - time (sec): 17.75 - samples/sec: 2934.08 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-14 12:04:17,008 epoch 6 - iter 288/723 - loss 0.02667079 - time (sec): 23.89 - samples/sec: 2944.44 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-14 12:04:23,616 epoch 6 - iter 360/723 - loss 0.02748106 - time (sec): 30.50 - samples/sec: 2934.75 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-14 12:04:29,904 epoch 6 - iter 432/723 - loss 0.02624476 - time (sec): 36.79 - samples/sec: 2904.61 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-14 12:04:35,334 epoch 6 - iter 504/723 - loss 0.02505646 - time (sec): 42.22 - samples/sec: 2928.64 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-14 12:04:41,250 epoch 6 - iter 576/723 - loss 0.02403994 - time (sec): 48.13 - samples/sec: 2925.55 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-14 12:04:46,875 epoch 6 - iter 648/723 - loss 0.02367198 - time (sec): 53.76 - samples/sec: 2944.02 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-14 12:04:52,633 epoch 6 - iter 720/723 - loss 0.02343289 - time (sec): 59.51 - samples/sec: 2952.24 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-14 12:04:52,843 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-14 12:04:52,843 EPOCH 6 done: loss 0.0234 - lr: 0.000022
162
+ 2023-10-14 12:04:56,349 DEV : loss 0.145940899848938 - f1-score (micro avg) 0.8041
163
+ 2023-10-14 12:04:56,367 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-14 12:05:02,368 epoch 7 - iter 72/723 - loss 0.01742657 - time (sec): 6.00 - samples/sec: 2837.14 - lr: 0.000022 - momentum: 0.000000
165
+ 2023-10-14 12:05:08,372 epoch 7 - iter 144/723 - loss 0.01539851 - time (sec): 12.00 - samples/sec: 2851.72 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-14 12:05:15,025 epoch 7 - iter 216/723 - loss 0.01540275 - time (sec): 18.66 - samples/sec: 2821.55 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-14 12:05:21,000 epoch 7 - iter 288/723 - loss 0.01735549 - time (sec): 24.63 - samples/sec: 2842.93 - lr: 0.000020 - momentum: 0.000000
168
+ 2023-10-14 12:05:26,955 epoch 7 - iter 360/723 - loss 0.01650779 - time (sec): 30.59 - samples/sec: 2880.15 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-14 12:05:32,689 epoch 7 - iter 432/723 - loss 0.01640627 - time (sec): 36.32 - samples/sec: 2901.85 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-14 12:05:38,703 epoch 7 - iter 504/723 - loss 0.01624031 - time (sec): 42.33 - samples/sec: 2904.94 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-14 12:05:44,741 epoch 7 - iter 576/723 - loss 0.01639039 - time (sec): 48.37 - samples/sec: 2907.00 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-14 12:05:50,379 epoch 7 - iter 648/723 - loss 0.01690292 - time (sec): 54.01 - samples/sec: 2909.39 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-14 12:05:56,792 epoch 7 - iter 720/723 - loss 0.01707398 - time (sec): 60.42 - samples/sec: 2907.57 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-14 12:05:57,013 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-14 12:05:57,013 EPOCH 7 done: loss 0.0172 - lr: 0.000017
176
+ 2023-10-14 12:06:00,508 DEV : loss 0.15821607410907745 - f1-score (micro avg) 0.8065
177
+ 2023-10-14 12:06:00,524 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-14 12:06:06,459 epoch 8 - iter 72/723 - loss 0.01327058 - time (sec): 5.93 - samples/sec: 3001.68 - lr: 0.000016 - momentum: 0.000000
179
+ 2023-10-14 12:06:12,306 epoch 8 - iter 144/723 - loss 0.01093102 - time (sec): 11.78 - samples/sec: 2984.11 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-14 12:06:18,692 epoch 8 - iter 216/723 - loss 0.01220562 - time (sec): 18.17 - samples/sec: 2931.79 - lr: 0.000015 - momentum: 0.000000
181
+ 2023-10-14 12:06:24,532 epoch 8 - iter 288/723 - loss 0.01255941 - time (sec): 24.01 - samples/sec: 2940.87 - lr: 0.000014 - momentum: 0.000000
182
+ 2023-10-14 12:06:30,631 epoch 8 - iter 360/723 - loss 0.01253527 - time (sec): 30.11 - samples/sec: 2941.73 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-14 12:06:36,280 epoch 8 - iter 432/723 - loss 0.01294269 - time (sec): 35.75 - samples/sec: 2963.97 - lr: 0.000013 - momentum: 0.000000
184
+ 2023-10-14 12:06:41,907 epoch 8 - iter 504/723 - loss 0.01262547 - time (sec): 41.38 - samples/sec: 2956.91 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-14 12:06:48,186 epoch 8 - iter 576/723 - loss 0.01217781 - time (sec): 47.66 - samples/sec: 2944.97 - lr: 0.000012 - momentum: 0.000000
186
+ 2023-10-14 12:06:54,429 epoch 8 - iter 648/723 - loss 0.01292360 - time (sec): 53.90 - samples/sec: 2940.90 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-14 12:07:00,216 epoch 8 - iter 720/723 - loss 0.01290911 - time (sec): 59.69 - samples/sec: 2939.40 - lr: 0.000011 - momentum: 0.000000
188
+ 2023-10-14 12:07:00,519 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-14 12:07:00,519 EPOCH 8 done: loss 0.0129 - lr: 0.000011
190
+ 2023-10-14 12:07:05,301 DEV : loss 0.1818445473909378 - f1-score (micro avg) 0.8002
191
+ 2023-10-14 12:07:05,325 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-14 12:07:11,461 epoch 9 - iter 72/723 - loss 0.00671491 - time (sec): 6.14 - samples/sec: 2841.22 - lr: 0.000011 - momentum: 0.000000
193
+ 2023-10-14 12:07:17,079 epoch 9 - iter 144/723 - loss 0.00809482 - time (sec): 11.75 - samples/sec: 2847.47 - lr: 0.000010 - momentum: 0.000000
194
+ 2023-10-14 12:07:23,944 epoch 9 - iter 216/723 - loss 0.00970598 - time (sec): 18.62 - samples/sec: 2880.97 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-14 12:07:29,480 epoch 9 - iter 288/723 - loss 0.00914238 - time (sec): 24.15 - samples/sec: 2911.70 - lr: 0.000009 - momentum: 0.000000
196
+ 2023-10-14 12:07:35,420 epoch 9 - iter 360/723 - loss 0.00931955 - time (sec): 30.09 - samples/sec: 2937.31 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-14 12:07:41,293 epoch 9 - iter 432/723 - loss 0.00901514 - time (sec): 35.97 - samples/sec: 2943.28 - lr: 0.000008 - momentum: 0.000000
198
+ 2023-10-14 12:07:47,142 epoch 9 - iter 504/723 - loss 0.00881443 - time (sec): 41.82 - samples/sec: 2949.50 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-14 12:07:53,123 epoch 9 - iter 576/723 - loss 0.00888603 - time (sec): 47.80 - samples/sec: 2937.46 - lr: 0.000007 - momentum: 0.000000
200
+ 2023-10-14 12:07:58,907 epoch 9 - iter 648/723 - loss 0.00902752 - time (sec): 53.58 - samples/sec: 2943.85 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-14 12:08:04,899 epoch 9 - iter 720/723 - loss 0.00947007 - time (sec): 59.57 - samples/sec: 2945.66 - lr: 0.000006 - momentum: 0.000000
202
+ 2023-10-14 12:08:05,164 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-14 12:08:05,164 EPOCH 9 done: loss 0.0094 - lr: 0.000006
204
+ 2023-10-14 12:08:08,738 DEV : loss 0.1774718016386032 - f1-score (micro avg) 0.8009
205
+ 2023-10-14 12:08:08,755 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-14 12:08:15,069 epoch 10 - iter 72/723 - loss 0.00669000 - time (sec): 6.31 - samples/sec: 2867.51 - lr: 0.000005 - momentum: 0.000000
207
+ 2023-10-14 12:08:20,844 epoch 10 - iter 144/723 - loss 0.00736923 - time (sec): 12.09 - samples/sec: 2948.87 - lr: 0.000004 - momentum: 0.000000
208
+ 2023-10-14 12:08:27,527 epoch 10 - iter 216/723 - loss 0.00741936 - time (sec): 18.77 - samples/sec: 2838.69 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-14 12:08:33,953 epoch 10 - iter 288/723 - loss 0.00692285 - time (sec): 25.20 - samples/sec: 2850.72 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-14 12:08:39,756 epoch 10 - iter 360/723 - loss 0.00587923 - time (sec): 31.00 - samples/sec: 2879.20 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-14 12:08:45,462 epoch 10 - iter 432/723 - loss 0.00582156 - time (sec): 36.71 - samples/sec: 2909.42 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 12:08:51,557 epoch 10 - iter 504/723 - loss 0.00635550 - time (sec): 42.80 - samples/sec: 2904.28 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-14 12:08:57,278 epoch 10 - iter 576/723 - loss 0.00666718 - time (sec): 48.52 - samples/sec: 2915.73 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 12:09:02,977 epoch 10 - iter 648/723 - loss 0.00681012 - time (sec): 54.22 - samples/sec: 2914.70 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-14 12:09:08,726 epoch 10 - iter 720/723 - loss 0.00652961 - time (sec): 59.97 - samples/sec: 2927.48 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-14 12:09:09,024 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-14 12:09:09,024 EPOCH 10 done: loss 0.0065 - lr: 0.000000
218
+ 2023-10-14 12:09:12,568 DEV : loss 0.18370041251182556 - f1-score (micro avg) 0.8102
219
+ 2023-10-14 12:09:12,998 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-14 12:09:12,999 Loading model from best epoch ...
221
+ 2023-10-14 12:09:14,567 SequenceTagger predicts: Dictionary with 13 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
222
+ 2023-10-14 12:09:17,806
223
+ Results:
224
+ - F-score (micro) 0.7924
225
+ - F-score (macro) 0.6936
226
+ - Accuracy 0.6775
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ PER 0.7451 0.8610 0.7988 482
232
+ LOC 0.8682 0.8057 0.8358 458
233
+ ORG 0.4754 0.4203 0.4462 69
234
+
235
+ micro avg 0.7795 0.8057 0.7924 1009
236
+ macro avg 0.6962 0.6957 0.6936 1009
237
+ weighted avg 0.7825 0.8057 0.7915 1009
238
+
239
+ 2023-10-14 12:09:17,806 ----------------------------------------------------------------------------------------------------