Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
stefan-it commited on
Commit
82e8371
·
1 Parent(s): 962dbd3

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 +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc7e8045a23a7146873c15b29a1d5bf775431a875bc9774e9eb2b252a2b18e79
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 10:57:19 0.0000 0.3871 0.1484 0.6875 0.7333 0.7097 0.5722
3
+ 2 10:58:52 0.0000 0.1201 0.1257 0.6912 0.7918 0.7381 0.6044
4
+ 3 11:00:28 0.0000 0.0851 0.1489 0.7281 0.8014 0.7630 0.6368
5
+ 4 11:02:01 0.0000 0.0647 0.1738 0.7513 0.7932 0.7717 0.6478
6
+ 5 11:03:33 0.0000 0.0490 0.1748 0.7660 0.7796 0.7728 0.6549
7
+ 6 11:05:07 0.0000 0.0410 0.1692 0.7591 0.7973 0.7777 0.6562
8
+ 7 11:06:40 0.0000 0.0292 0.1868 0.7716 0.7905 0.7809 0.6648
9
+ 8 11:08:13 0.0000 0.0221 0.1947 0.7880 0.8041 0.7960 0.6824
10
+ 9 11:09:48 0.0000 0.0180 0.1845 0.7911 0.8190 0.8048 0.6928
11
+ 10 11:11:32 0.0000 0.0118 0.1851 0.7902 0.8150 0.8024 0.6893
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-16 10:55:48,313 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-16 10:55:48,314 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
+ 2023-10-16 10:55:48,314 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-16 10:55:48,314 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
52
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
53
+ 2023-10-16 10:55:48,314 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-16 10:55:48,314 Train: 7142 sentences
55
+ 2023-10-16 10:55:48,314 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-16 10:55:48,314 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-16 10:55:48,315 Training Params:
58
+ 2023-10-16 10:55:48,315 - learning_rate: "5e-05"
59
+ 2023-10-16 10:55:48,315 - mini_batch_size: "4"
60
+ 2023-10-16 10:55:48,315 - max_epochs: "10"
61
+ 2023-10-16 10:55:48,315 - shuffle: "True"
62
+ 2023-10-16 10:55:48,315 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-16 10:55:48,315 Plugins:
64
+ 2023-10-16 10:55:48,315 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-16 10:55:48,315 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-16 10:55:48,315 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-16 10:55:48,315 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-16 10:55:48,315 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-16 10:55:48,315 Computation:
70
+ 2023-10-16 10:55:48,315 - compute on device: cuda:0
71
+ 2023-10-16 10:55:48,315 - embedding storage: none
72
+ 2023-10-16 10:55:48,315 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-16 10:55:48,315 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-16 10:55:48,315 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-16 10:55:48,315 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-16 10:55:57,101 epoch 1 - iter 178/1786 - loss 1.86473963 - time (sec): 8.78 - samples/sec: 2798.20 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-16 10:56:05,989 epoch 1 - iter 356/1786 - loss 1.15143669 - time (sec): 17.67 - samples/sec: 2805.49 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-16 10:56:14,650 epoch 1 - iter 534/1786 - loss 0.89028895 - time (sec): 26.33 - samples/sec: 2786.50 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-16 10:56:23,319 epoch 1 - iter 712/1786 - loss 0.72907805 - time (sec): 35.00 - samples/sec: 2791.74 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-16 10:56:32,196 epoch 1 - iter 890/1786 - loss 0.61722937 - time (sec): 43.88 - samples/sec: 2795.47 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-16 10:56:40,923 epoch 1 - iter 1068/1786 - loss 0.54496368 - time (sec): 52.61 - samples/sec: 2804.43 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-16 10:56:49,517 epoch 1 - iter 1246/1786 - loss 0.49315943 - time (sec): 61.20 - samples/sec: 2810.17 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-16 10:56:58,516 epoch 1 - iter 1424/1786 - loss 0.44831232 - time (sec): 70.20 - samples/sec: 2810.84 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-16 10:57:07,385 epoch 1 - iter 1602/1786 - loss 0.41610604 - time (sec): 79.07 - samples/sec: 2819.82 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-16 10:57:16,356 epoch 1 - iter 1780/1786 - loss 0.38771319 - time (sec): 88.04 - samples/sec: 2818.72 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-16 10:57:16,622 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-16 10:57:16,623 EPOCH 1 done: loss 0.3871 - lr: 0.000050
88
+ 2023-10-16 10:57:19,749 DEV : loss 0.14844156801700592 - f1-score (micro avg) 0.7097
89
+ 2023-10-16 10:57:19,766 saving best model
90
+ 2023-10-16 10:57:20,235 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-16 10:57:29,212 epoch 2 - iter 178/1786 - loss 0.11605918 - time (sec): 8.98 - samples/sec: 2852.83 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-16 10:57:38,138 epoch 2 - iter 356/1786 - loss 0.12100219 - time (sec): 17.90 - samples/sec: 2815.83 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-16 10:57:46,829 epoch 2 - iter 534/1786 - loss 0.12074401 - time (sec): 26.59 - samples/sec: 2813.99 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-16 10:57:55,616 epoch 2 - iter 712/1786 - loss 0.12193352 - time (sec): 35.38 - samples/sec: 2817.75 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-16 10:58:04,529 epoch 2 - iter 890/1786 - loss 0.12062941 - time (sec): 44.29 - samples/sec: 2848.46 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-16 10:58:13,257 epoch 2 - iter 1068/1786 - loss 0.12155219 - time (sec): 53.02 - samples/sec: 2842.58 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-16 10:58:21,966 epoch 2 - iter 1246/1786 - loss 0.12243129 - time (sec): 61.73 - samples/sec: 2853.56 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-16 10:58:30,614 epoch 2 - iter 1424/1786 - loss 0.12158950 - time (sec): 70.38 - samples/sec: 2851.52 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-16 10:58:39,167 epoch 2 - iter 1602/1786 - loss 0.12123775 - time (sec): 78.93 - samples/sec: 2850.42 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-16 10:58:47,748 epoch 2 - iter 1780/1786 - loss 0.11991333 - time (sec): 87.51 - samples/sec: 2836.16 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-16 10:58:48,021 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-16 10:58:48,022 EPOCH 2 done: loss 0.1201 - lr: 0.000044
103
+ 2023-10-16 10:58:52,252 DEV : loss 0.12567166984081268 - f1-score (micro avg) 0.7381
104
+ 2023-10-16 10:58:52,278 saving best model
105
+ 2023-10-16 10:58:52,816 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-16 10:59:02,942 epoch 3 - iter 178/1786 - loss 0.07602399 - time (sec): 10.12 - samples/sec: 2507.17 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-16 10:59:12,927 epoch 3 - iter 356/1786 - loss 0.08707339 - time (sec): 20.11 - samples/sec: 2555.56 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-16 10:59:21,653 epoch 3 - iter 534/1786 - loss 0.08690813 - time (sec): 28.83 - samples/sec: 2648.10 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-16 10:59:30,119 epoch 3 - iter 712/1786 - loss 0.08815103 - time (sec): 37.30 - samples/sec: 2670.85 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-16 10:59:38,819 epoch 3 - iter 890/1786 - loss 0.08755368 - time (sec): 46.00 - samples/sec: 2695.07 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-16 10:59:47,465 epoch 3 - iter 1068/1786 - loss 0.08864883 - time (sec): 54.65 - samples/sec: 2710.43 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-16 10:59:56,372 epoch 3 - iter 1246/1786 - loss 0.08622224 - time (sec): 63.55 - samples/sec: 2735.57 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-16 11:00:05,220 epoch 3 - iter 1424/1786 - loss 0.08740866 - time (sec): 72.40 - samples/sec: 2750.50 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-16 11:00:13,804 epoch 3 - iter 1602/1786 - loss 0.08628163 - time (sec): 80.98 - samples/sec: 2760.57 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-16 11:00:22,397 epoch 3 - iter 1780/1786 - loss 0.08517479 - time (sec): 89.58 - samples/sec: 2763.43 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-16 11:00:22,747 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-16 11:00:22,747 EPOCH 3 done: loss 0.0851 - lr: 0.000039
118
+ 2023-10-16 11:00:28,065 DEV : loss 0.14893780648708344 - f1-score (micro avg) 0.763
119
+ 2023-10-16 11:00:28,095 saving best model
120
+ 2023-10-16 11:00:28,742 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-16 11:00:37,510 epoch 4 - iter 178/1786 - loss 0.05930779 - time (sec): 8.77 - samples/sec: 2921.05 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-16 11:00:46,255 epoch 4 - iter 356/1786 - loss 0.06322745 - time (sec): 17.51 - samples/sec: 2865.97 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-16 11:00:54,769 epoch 4 - iter 534/1786 - loss 0.06363083 - time (sec): 26.03 - samples/sec: 2834.26 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-16 11:01:03,344 epoch 4 - iter 712/1786 - loss 0.06394879 - time (sec): 34.60 - samples/sec: 2818.00 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-16 11:01:12,317 epoch 4 - iter 890/1786 - loss 0.06327639 - time (sec): 43.57 - samples/sec: 2815.14 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-16 11:01:21,136 epoch 4 - iter 1068/1786 - loss 0.06455310 - time (sec): 52.39 - samples/sec: 2814.42 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-16 11:01:29,981 epoch 4 - iter 1246/1786 - loss 0.06478828 - time (sec): 61.24 - samples/sec: 2837.72 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-16 11:01:38,780 epoch 4 - iter 1424/1786 - loss 0.06518193 - time (sec): 70.04 - samples/sec: 2829.75 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-16 11:01:47,514 epoch 4 - iter 1602/1786 - loss 0.06559590 - time (sec): 78.77 - samples/sec: 2834.31 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-16 11:01:56,245 epoch 4 - iter 1780/1786 - loss 0.06474763 - time (sec): 87.50 - samples/sec: 2832.00 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-16 11:01:56,584 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-16 11:01:56,584 EPOCH 4 done: loss 0.0647 - lr: 0.000033
133
+ 2023-10-16 11:02:01,327 DEV : loss 0.17378392815589905 - f1-score (micro avg) 0.7717
134
+ 2023-10-16 11:02:01,343 saving best model
135
+ 2023-10-16 11:02:01,929 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-16 11:02:10,798 epoch 5 - iter 178/1786 - loss 0.04377203 - time (sec): 8.86 - samples/sec: 2853.88 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-16 11:02:19,500 epoch 5 - iter 356/1786 - loss 0.04754169 - time (sec): 17.56 - samples/sec: 2891.83 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-16 11:02:28,081 epoch 5 - iter 534/1786 - loss 0.04802223 - time (sec): 26.15 - samples/sec: 2810.91 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-16 11:02:36,836 epoch 5 - iter 712/1786 - loss 0.04911834 - time (sec): 34.90 - samples/sec: 2859.03 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-16 11:02:45,355 epoch 5 - iter 890/1786 - loss 0.04948678 - time (sec): 43.42 - samples/sec: 2844.94 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-16 11:02:54,112 epoch 5 - iter 1068/1786 - loss 0.04806990 - time (sec): 52.18 - samples/sec: 2848.29 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-16 11:03:02,929 epoch 5 - iter 1246/1786 - loss 0.04825570 - time (sec): 60.99 - samples/sec: 2841.27 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-16 11:03:11,923 epoch 5 - iter 1424/1786 - loss 0.04891461 - time (sec): 69.99 - samples/sec: 2857.31 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-16 11:03:20,483 epoch 5 - iter 1602/1786 - loss 0.04962470 - time (sec): 78.55 - samples/sec: 2857.09 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-16 11:03:29,083 epoch 5 - iter 1780/1786 - loss 0.04904968 - time (sec): 87.15 - samples/sec: 2845.88 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-16 11:03:29,351 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-16 11:03:29,351 EPOCH 5 done: loss 0.0490 - lr: 0.000028
148
+ 2023-10-16 11:03:33,517 DEV : loss 0.17484842240810394 - f1-score (micro avg) 0.7728
149
+ 2023-10-16 11:03:33,537 saving best model
150
+ 2023-10-16 11:03:34,174 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-16 11:03:43,090 epoch 6 - iter 178/1786 - loss 0.03564048 - time (sec): 8.91 - samples/sec: 2725.57 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-16 11:03:52,442 epoch 6 - iter 356/1786 - loss 0.03852911 - time (sec): 18.26 - samples/sec: 2720.29 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-16 11:04:01,277 epoch 6 - iter 534/1786 - loss 0.04136964 - time (sec): 27.10 - samples/sec: 2738.01 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-16 11:04:09,800 epoch 6 - iter 712/1786 - loss 0.04024227 - time (sec): 35.62 - samples/sec: 2769.91 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-16 11:04:18,513 epoch 6 - iter 890/1786 - loss 0.04012505 - time (sec): 44.34 - samples/sec: 2778.95 - lr: 0.000025 - momentum: 0.000000
156
+ 2023-10-16 11:04:27,194 epoch 6 - iter 1068/1786 - loss 0.04289492 - time (sec): 53.02 - samples/sec: 2807.61 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-16 11:04:36,047 epoch 6 - iter 1246/1786 - loss 0.04221536 - time (sec): 61.87 - samples/sec: 2818.13 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-16 11:04:44,578 epoch 6 - iter 1424/1786 - loss 0.04182216 - time (sec): 70.40 - samples/sec: 2812.76 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-16 11:04:53,113 epoch 6 - iter 1602/1786 - loss 0.04171130 - time (sec): 78.94 - samples/sec: 2819.50 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-16 11:05:01,768 epoch 6 - iter 1780/1786 - loss 0.04099380 - time (sec): 87.59 - samples/sec: 2833.27 - lr: 0.000022 - momentum: 0.000000
161
+ 2023-10-16 11:05:02,038 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-16 11:05:02,038 EPOCH 6 done: loss 0.0410 - lr: 0.000022
163
+ 2023-10-16 11:05:07,359 DEV : loss 0.16918683052062988 - f1-score (micro avg) 0.7777
164
+ 2023-10-16 11:05:07,384 saving best model
165
+ 2023-10-16 11:05:08,130 ----------------------------------------------------------------------------------------------------
166
+ 2023-10-16 11:05:16,990 epoch 7 - iter 178/1786 - loss 0.02648365 - time (sec): 8.86 - samples/sec: 3052.78 - lr: 0.000022 - momentum: 0.000000
167
+ 2023-10-16 11:05:25,906 epoch 7 - iter 356/1786 - loss 0.02728552 - time (sec): 17.77 - samples/sec: 2869.74 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-16 11:05:34,815 epoch 7 - iter 534/1786 - loss 0.02817086 - time (sec): 26.68 - samples/sec: 2837.46 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-16 11:05:43,583 epoch 7 - iter 712/1786 - loss 0.02776279 - time (sec): 35.45 - samples/sec: 2851.27 - lr: 0.000020 - momentum: 0.000000
170
+ 2023-10-16 11:05:52,253 epoch 7 - iter 890/1786 - loss 0.02711624 - time (sec): 44.12 - samples/sec: 2851.82 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-16 11:06:00,861 epoch 7 - iter 1068/1786 - loss 0.02876745 - time (sec): 52.73 - samples/sec: 2839.91 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-16 11:06:09,636 epoch 7 - iter 1246/1786 - loss 0.02936533 - time (sec): 61.50 - samples/sec: 2815.16 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-16 11:06:18,532 epoch 7 - iter 1424/1786 - loss 0.02812044 - time (sec): 70.40 - samples/sec: 2830.18 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-16 11:06:27,146 epoch 7 - iter 1602/1786 - loss 0.02927136 - time (sec): 79.01 - samples/sec: 2823.45 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-16 11:06:35,809 epoch 7 - iter 1780/1786 - loss 0.02918995 - time (sec): 87.68 - samples/sec: 2825.84 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-16 11:06:36,138 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-16 11:06:36,139 EPOCH 7 done: loss 0.0292 - lr: 0.000017
178
+ 2023-10-16 11:06:40,964 DEV : loss 0.186806321144104 - f1-score (micro avg) 0.7809
179
+ 2023-10-16 11:06:40,981 saving best model
180
+ 2023-10-16 11:06:41,594 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-16 11:06:50,242 epoch 8 - iter 178/1786 - loss 0.02299188 - time (sec): 8.65 - samples/sec: 2815.85 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-16 11:06:58,827 epoch 8 - iter 356/1786 - loss 0.02359001 - time (sec): 17.23 - samples/sec: 2827.78 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-16 11:07:07,629 epoch 8 - iter 534/1786 - loss 0.02254523 - time (sec): 26.03 - samples/sec: 2811.01 - lr: 0.000015 - momentum: 0.000000
184
+ 2023-10-16 11:07:16,444 epoch 8 - iter 712/1786 - loss 0.02199670 - time (sec): 34.85 - samples/sec: 2838.40 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-16 11:07:25,076 epoch 8 - iter 890/1786 - loss 0.02265221 - time (sec): 43.48 - samples/sec: 2842.73 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-16 11:07:33,588 epoch 8 - iter 1068/1786 - loss 0.02375428 - time (sec): 51.99 - samples/sec: 2838.22 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-16 11:07:42,171 epoch 8 - iter 1246/1786 - loss 0.02271185 - time (sec): 60.57 - samples/sec: 2838.00 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-16 11:07:50,996 epoch 8 - iter 1424/1786 - loss 0.02264176 - time (sec): 69.40 - samples/sec: 2830.78 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-16 11:07:59,917 epoch 8 - iter 1602/1786 - loss 0.02278864 - time (sec): 78.32 - samples/sec: 2834.50 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-16 11:08:08,823 epoch 8 - iter 1780/1786 - loss 0.02208015 - time (sec): 87.23 - samples/sec: 2842.31 - lr: 0.000011 - momentum: 0.000000
191
+ 2023-10-16 11:08:09,133 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-16 11:08:09,133 EPOCH 8 done: loss 0.0221 - lr: 0.000011
193
+ 2023-10-16 11:08:13,486 DEV : loss 0.19474762678146362 - f1-score (micro avg) 0.796
194
+ 2023-10-16 11:08:13,506 saving best model
195
+ 2023-10-16 11:08:14,028 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-16 11:08:23,840 epoch 9 - iter 178/1786 - loss 0.01798579 - time (sec): 9.81 - samples/sec: 2549.51 - lr: 0.000011 - momentum: 0.000000
197
+ 2023-10-16 11:08:32,940 epoch 9 - iter 356/1786 - loss 0.01872623 - time (sec): 18.91 - samples/sec: 2662.25 - lr: 0.000010 - momentum: 0.000000
198
+ 2023-10-16 11:08:42,168 epoch 9 - iter 534/1786 - loss 0.01727464 - time (sec): 28.14 - samples/sec: 2677.48 - lr: 0.000009 - momentum: 0.000000
199
+ 2023-10-16 11:08:50,782 epoch 9 - iter 712/1786 - loss 0.01690762 - time (sec): 36.75 - samples/sec: 2698.90 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-16 11:08:59,465 epoch 9 - iter 890/1786 - loss 0.01729150 - time (sec): 45.43 - samples/sec: 2718.02 - lr: 0.000008 - momentum: 0.000000
201
+ 2023-10-16 11:09:08,235 epoch 9 - iter 1068/1786 - loss 0.01759655 - time (sec): 54.20 - samples/sec: 2743.24 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-16 11:09:17,081 epoch 9 - iter 1246/1786 - loss 0.01836818 - time (sec): 63.05 - samples/sec: 2762.68 - lr: 0.000007 - momentum: 0.000000
203
+ 2023-10-16 11:09:25,811 epoch 9 - iter 1424/1786 - loss 0.01880668 - time (sec): 71.78 - samples/sec: 2792.07 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-16 11:09:34,295 epoch 9 - iter 1602/1786 - loss 0.01846836 - time (sec): 80.26 - samples/sec: 2780.26 - lr: 0.000006 - momentum: 0.000000
205
+ 2023-10-16 11:09:43,041 epoch 9 - iter 1780/1786 - loss 0.01794289 - time (sec): 89.01 - samples/sec: 2784.51 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-16 11:09:43,337 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-16 11:09:43,338 EPOCH 9 done: loss 0.0180 - lr: 0.000006
208
+ 2023-10-16 11:09:48,154 DEV : loss 0.1844618022441864 - f1-score (micro avg) 0.8048
209
+ 2023-10-16 11:09:48,181 saving best model
210
+ 2023-10-16 11:09:48,726 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-16 11:09:58,642 epoch 10 - iter 178/1786 - loss 0.01185496 - time (sec): 9.91 - samples/sec: 2381.04 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-16 11:10:08,696 epoch 10 - iter 356/1786 - loss 0.01135159 - time (sec): 19.97 - samples/sec: 2548.54 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-16 11:10:18,592 epoch 10 - iter 534/1786 - loss 0.01376928 - time (sec): 29.86 - samples/sec: 2545.00 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-16 11:10:28,392 epoch 10 - iter 712/1786 - loss 0.01268921 - time (sec): 39.66 - samples/sec: 2517.19 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-16 11:10:38,278 epoch 10 - iter 890/1786 - loss 0.01251690 - time (sec): 49.55 - samples/sec: 2514.27 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-16 11:10:48,164 epoch 10 - iter 1068/1786 - loss 0.01347864 - time (sec): 59.44 - samples/sec: 2506.01 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-16 11:10:58,148 epoch 10 - iter 1246/1786 - loss 0.01278922 - time (sec): 69.42 - samples/sec: 2507.14 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-16 11:11:07,753 epoch 10 - iter 1424/1786 - loss 0.01267902 - time (sec): 79.02 - samples/sec: 2494.79 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-16 11:11:17,680 epoch 10 - iter 1602/1786 - loss 0.01194308 - time (sec): 88.95 - samples/sec: 2510.87 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-16 11:11:27,393 epoch 10 - iter 1780/1786 - loss 0.01186500 - time (sec): 98.66 - samples/sec: 2515.20 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-16 11:11:27,704 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-16 11:11:27,704 EPOCH 10 done: loss 0.0118 - lr: 0.000000
223
+ 2023-10-16 11:11:32,731 DEV : loss 0.18512549996376038 - f1-score (micro avg) 0.8024
224
+ 2023-10-16 11:11:33,164 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-16 11:11:33,165 Loading model from best epoch ...
226
+ 2023-10-16 11:11:34,828 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
227
+ 2023-10-16 11:11:44,735
228
+ Results:
229
+ - F-score (micro) 0.6916
230
+ - F-score (macro) 0.5961
231
+ - Accuracy 0.5475
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ LOC 0.6627 0.7196 0.6900 1095
237
+ PER 0.7735 0.7628 0.7682 1012
238
+ ORG 0.5198 0.5154 0.5176 357
239
+ HumanProd 0.3167 0.5758 0.4086 33
240
+
241
+ micro avg 0.6778 0.7060 0.6916 2497
242
+ macro avg 0.5682 0.6434 0.5961 2497
243
+ weighted avg 0.6826 0.7060 0.6933 2497
244
+
245
+ 2023-10-16 11:11:44,735 ----------------------------------------------------------------------------------------------------