Hugging Face's logo

hmbert-tiny
/
flair-hipe-2022-newseye-fr

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
c055566
•
1 Parent(s): 69c7b1b

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.1697744184.46dc0c540dd0.4731.1 +3 -0
  5. test.tsv +0 -0
  6. training.log +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11ed184726d0d78798ec9b07adb864057033b9be8f940a3cab19c0f209aaf481
3
+ size 19048098
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 19:36:56 0.0000 1.1002 0.2989 0.2801 0.1741 0.2148 0.1244
3
+ 2 19:37:30 0.0000 0.3817 0.2360 0.4566 0.3796 0.4146 0.2773
4
+ 3 19:38:03 0.0000 0.3013 0.2129 0.5014 0.4898 0.4955 0.3488
5
+ 4 19:38:36 0.0000 0.2639 0.1954 0.5267 0.5510 0.5386 0.3913
6
+ 5 19:39:09 0.0000 0.2359 0.1977 0.5095 0.5810 0.5429 0.3957
7
+ 6 19:39:43 0.0000 0.2166 0.1919 0.5360 0.5878 0.5607 0.4083
8
+ 7 19:40:15 0.0000 0.2037 0.1908 0.5249 0.5878 0.5546 0.4045
9
+ 8 19:40:49 0.0000 0.1933 0.1924 0.5567 0.6014 0.5782 0.4246
10
+ 9 19:41:22 0.0000 0.1861 0.1915 0.5366 0.6082 0.5702 0.4181
11
+ 10 19:41:55 0.0000 0.1813 0.1935 0.5370 0.6027 0.5679 0.4152
runs/events.out.tfevents.1697744184.46dc0c540dd0.4731.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21e983f3a219248ad22e9661cc10cff051caafe55338257bc71cd3d8e637ce8b
3
+ size 999862
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-19 19:36:24,482 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-19 19:36:24,482 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-19 19:36:24,482 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-19 19:36:24,482 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
+ 2023-10-19 19:36:24,482 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-19 19:36:24,482 Train: 7142 sentences
55
+ 2023-10-19 19:36:24,482 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-19 19:36:24,482 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-19 19:36:24,482 Training Params:
58
+ 2023-10-19 19:36:24,482 - learning_rate: "5e-05"
59
+ 2023-10-19 19:36:24,482 - mini_batch_size: "4"
60
+ 2023-10-19 19:36:24,482 - max_epochs: "10"
61
+ 2023-10-19 19:36:24,482 - shuffle: "True"
62
+ 2023-10-19 19:36:24,482 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-19 19:36:24,482 Plugins:
64
+ 2023-10-19 19:36:24,482 - TensorboardLogger
65
+ 2023-10-19 19:36:24,483 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-19 19:36:24,483 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-19 19:36:24,483 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-19 19:36:24,483 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-19 19:36:24,483 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-19 19:36:24,483 Computation:
71
+ 2023-10-19 19:36:24,483 - compute on device: cuda:0
72
+ 2023-10-19 19:36:24,483 - embedding storage: none
73
+ 2023-10-19 19:36:24,483 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-19 19:36:24,483 Model training base path: "hmbench-newseye/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-19 19:36:24,483 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-19 19:36:24,483 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-19 19:36:24,483 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-19 19:36:27,605 epoch 1 - iter 178/1786 - loss 3.27396970 - time (sec): 3.12 - samples/sec: 8577.89 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-19 19:36:30,669 epoch 1 - iter 356/1786 - loss 2.78021747 - time (sec): 6.19 - samples/sec: 8319.68 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-19 19:36:33,854 epoch 1 - iter 534/1786 - loss 2.20153535 - time (sec): 9.37 - samples/sec: 8225.33 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-19 19:36:36,825 epoch 1 - iter 712/1786 - loss 1.87990522 - time (sec): 12.34 - samples/sec: 8105.74 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-19 19:36:39,787 epoch 1 - iter 890/1786 - loss 1.65052606 - time (sec): 15.30 - samples/sec: 8162.50 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-19 19:36:42,845 epoch 1 - iter 1068/1786 - loss 1.48583748 - time (sec): 18.36 - samples/sec: 8135.08 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-19 19:36:45,942 epoch 1 - iter 1246/1786 - loss 1.35348330 - time (sec): 21.46 - samples/sec: 8120.55 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-19 19:36:48,993 epoch 1 - iter 1424/1786 - loss 1.25058687 - time (sec): 24.51 - samples/sec: 8187.36 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-19 19:36:52,056 epoch 1 - iter 1602/1786 - loss 1.16632694 - time (sec): 27.57 - samples/sec: 8211.05 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-19 19:36:55,021 epoch 1 - iter 1780/1786 - loss 1.10109838 - time (sec): 30.54 - samples/sec: 8128.81 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-19 19:36:55,111 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-19 19:36:55,111 EPOCH 1 done: loss 1.1002 - lr: 0.000050
90
+ 2023-10-19 19:36:56,586 DEV : loss 0.2988516390323639 - f1-score (micro avg) 0.2148
91
+ 2023-10-19 19:36:56,601 saving best model
92
+ 2023-10-19 19:36:56,632 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-19 19:36:59,872 epoch 2 - iter 178/1786 - loss 0.45233865 - time (sec): 3.24 - samples/sec: 7524.97 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-19 19:37:02,958 epoch 2 - iter 356/1786 - loss 0.41531477 - time (sec): 6.33 - samples/sec: 7897.94 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-19 19:37:06,164 epoch 2 - iter 534/1786 - loss 0.42140018 - time (sec): 9.53 - samples/sec: 7904.32 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-19 19:37:09,284 epoch 2 - iter 712/1786 - loss 0.41082544 - time (sec): 12.65 - samples/sec: 7981.69 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-19 19:37:12,285 epoch 2 - iter 890/1786 - loss 0.40282999 - time (sec): 15.65 - samples/sec: 7882.50 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-19 19:37:15,326 epoch 2 - iter 1068/1786 - loss 0.39333372 - time (sec): 18.69 - samples/sec: 7950.62 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-19 19:37:18,468 epoch 2 - iter 1246/1786 - loss 0.39031933 - time (sec): 21.84 - samples/sec: 7946.19 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-19 19:37:21,578 epoch 2 - iter 1424/1786 - loss 0.39082695 - time (sec): 24.95 - samples/sec: 8003.47 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-19 19:37:24,609 epoch 2 - iter 1602/1786 - loss 0.38591567 - time (sec): 27.98 - samples/sec: 7988.62 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-19 19:37:27,778 epoch 2 - iter 1780/1786 - loss 0.38165514 - time (sec): 31.15 - samples/sec: 7964.48 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-19 19:37:27,871 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-19 19:37:27,871 EPOCH 2 done: loss 0.3817 - lr: 0.000044
105
+ 2023-10-19 19:37:30,634 DEV : loss 0.2359563112258911 - f1-score (micro avg) 0.4146
106
+ 2023-10-19 19:37:30,648 saving best model
107
+ 2023-10-19 19:37:30,680 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-19 19:37:33,740 epoch 3 - iter 178/1786 - loss 0.29124349 - time (sec): 3.06 - samples/sec: 7695.75 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-19 19:37:36,723 epoch 3 - iter 356/1786 - loss 0.30538165 - time (sec): 6.04 - samples/sec: 7877.90 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-19 19:37:39,776 epoch 3 - iter 534/1786 - loss 0.31807769 - time (sec): 9.09 - samples/sec: 7900.96 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-19 19:37:42,785 epoch 3 - iter 712/1786 - loss 0.31141018 - time (sec): 12.10 - samples/sec: 7965.42 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-19 19:37:45,725 epoch 3 - iter 890/1786 - loss 0.31100537 - time (sec): 15.04 - samples/sec: 8024.61 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-19 19:37:48,854 epoch 3 - iter 1068/1786 - loss 0.30996002 - time (sec): 18.17 - samples/sec: 8102.43 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-19 19:37:51,863 epoch 3 - iter 1246/1786 - loss 0.30713704 - time (sec): 21.18 - samples/sec: 8081.12 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-19 19:37:55,026 epoch 3 - iter 1424/1786 - loss 0.30756342 - time (sec): 24.35 - samples/sec: 8084.50 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-19 19:37:58,152 epoch 3 - iter 1602/1786 - loss 0.30206285 - time (sec): 27.47 - samples/sec: 8134.80 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-19 19:38:01,178 epoch 3 - iter 1780/1786 - loss 0.30084992 - time (sec): 30.50 - samples/sec: 8129.64 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-19 19:38:01,277 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-19 19:38:01,277 EPOCH 3 done: loss 0.3013 - lr: 0.000039
120
+ 2023-10-19 19:38:03,640 DEV : loss 0.21294961869716644 - f1-score (micro avg) 0.4955
121
+ 2023-10-19 19:38:03,654 saving best model
122
+ 2023-10-19 19:38:03,687 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-19 19:38:06,205 epoch 4 - iter 178/1786 - loss 0.28951262 - time (sec): 2.52 - samples/sec: 9326.20 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-19 19:38:08,836 epoch 4 - iter 356/1786 - loss 0.27653901 - time (sec): 5.15 - samples/sec: 9375.61 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-19 19:38:11,931 epoch 4 - iter 534/1786 - loss 0.27776588 - time (sec): 8.24 - samples/sec: 9028.31 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-19 19:38:14,936 epoch 4 - iter 712/1786 - loss 0.27643033 - time (sec): 11.25 - samples/sec: 8604.89 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-19 19:38:17,926 epoch 4 - iter 890/1786 - loss 0.27157129 - time (sec): 14.24 - samples/sec: 8501.64 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-19 19:38:20,907 epoch 4 - iter 1068/1786 - loss 0.27366911 - time (sec): 17.22 - samples/sec: 8487.46 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-19 19:38:23,995 epoch 4 - iter 1246/1786 - loss 0.26878768 - time (sec): 20.31 - samples/sec: 8465.61 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-19 19:38:27,060 epoch 4 - iter 1424/1786 - loss 0.26860832 - time (sec): 23.37 - samples/sec: 8435.48 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-19 19:38:30,106 epoch 4 - iter 1602/1786 - loss 0.26521033 - time (sec): 26.42 - samples/sec: 8428.78 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-19 19:38:33,181 epoch 4 - iter 1780/1786 - loss 0.26377235 - time (sec): 29.49 - samples/sec: 8412.91 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-19 19:38:33,280 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-19 19:38:33,280 EPOCH 4 done: loss 0.2639 - lr: 0.000033
135
+ 2023-10-19 19:38:36,104 DEV : loss 0.1954185664653778 - f1-score (micro avg) 0.5386
136
+ 2023-10-19 19:38:36,118 saving best model
137
+ 2023-10-19 19:38:36,150 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-19 19:38:39,326 epoch 5 - iter 178/1786 - loss 0.22782338 - time (sec): 3.17 - samples/sec: 8017.78 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-19 19:38:42,405 epoch 5 - iter 356/1786 - loss 0.24493235 - time (sec): 6.25 - samples/sec: 8072.13 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-19 19:38:45,548 epoch 5 - iter 534/1786 - loss 0.24576962 - time (sec): 9.40 - samples/sec: 8134.03 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-19 19:38:48,433 epoch 5 - iter 712/1786 - loss 0.25073368 - time (sec): 12.28 - samples/sec: 8034.20 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-19 19:38:51,659 epoch 5 - iter 890/1786 - loss 0.24792561 - time (sec): 15.51 - samples/sec: 7972.54 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-19 19:38:54,683 epoch 5 - iter 1068/1786 - loss 0.24376175 - time (sec): 18.53 - samples/sec: 8021.19 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-19 19:38:57,803 epoch 5 - iter 1246/1786 - loss 0.24271881 - time (sec): 21.65 - samples/sec: 8010.20 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-19 19:39:00,899 epoch 5 - iter 1424/1786 - loss 0.23911368 - time (sec): 24.75 - samples/sec: 8072.76 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-19 19:39:03,941 epoch 5 - iter 1602/1786 - loss 0.23851321 - time (sec): 27.79 - samples/sec: 8075.32 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-19 19:39:07,099 epoch 5 - iter 1780/1786 - loss 0.23554334 - time (sec): 30.95 - samples/sec: 8021.42 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-19 19:39:07,184 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-19 19:39:07,184 EPOCH 5 done: loss 0.2359 - lr: 0.000028
150
+ 2023-10-19 19:39:09,528 DEV : loss 0.19769060611724854 - f1-score (micro avg) 0.5429
151
+ 2023-10-19 19:39:09,544 saving best model
152
+ 2023-10-19 19:39:09,578 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-19 19:39:12,652 epoch 6 - iter 178/1786 - loss 0.22045055 - time (sec): 3.07 - samples/sec: 8026.99 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-19 19:39:15,743 epoch 6 - iter 356/1786 - loss 0.21078132 - time (sec): 6.16 - samples/sec: 8203.99 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-19 19:39:18,777 epoch 6 - iter 534/1786 - loss 0.21210754 - time (sec): 9.20 - samples/sec: 8270.68 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-19 19:39:21,787 epoch 6 - iter 712/1786 - loss 0.21530597 - time (sec): 12.21 - samples/sec: 8235.79 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-19 19:39:24,845 epoch 6 - iter 890/1786 - loss 0.21713503 - time (sec): 15.27 - samples/sec: 8286.95 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-19 19:39:27,887 epoch 6 - iter 1068/1786 - loss 0.21502049 - time (sec): 18.31 - samples/sec: 8238.44 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-19 19:39:30,932 epoch 6 - iter 1246/1786 - loss 0.21781126 - time (sec): 21.35 - samples/sec: 8159.07 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-19 19:39:33,995 epoch 6 - iter 1424/1786 - loss 0.21736778 - time (sec): 24.42 - samples/sec: 8139.12 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-19 19:39:36,992 epoch 6 - iter 1602/1786 - loss 0.21517526 - time (sec): 27.41 - samples/sec: 8142.54 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-19 19:39:40,261 epoch 6 - iter 1780/1786 - loss 0.21641064 - time (sec): 30.68 - samples/sec: 8090.65 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-19 19:39:40,351 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-19 19:39:40,351 EPOCH 6 done: loss 0.2166 - lr: 0.000022
165
+ 2023-10-19 19:39:43,159 DEV : loss 0.19191302359104156 - f1-score (micro avg) 0.5607
166
+ 2023-10-19 19:39:43,174 saving best model
167
+ 2023-10-19 19:39:43,211 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-19 19:39:46,173 epoch 7 - iter 178/1786 - loss 0.20637887 - time (sec): 2.96 - samples/sec: 7808.24 - lr: 0.000022 - momentum: 0.000000
169
+ 2023-10-19 19:39:49,227 epoch 7 - iter 356/1786 - loss 0.20872566 - time (sec): 6.02 - samples/sec: 7998.94 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-19 19:39:52,303 epoch 7 - iter 534/1786 - loss 0.20004438 - time (sec): 9.09 - samples/sec: 8075.75 - lr: 0.000021 - momentum: 0.000000
171
+ 2023-10-19 19:39:55,213 epoch 7 - iter 712/1786 - loss 0.20338323 - time (sec): 12.00 - samples/sec: 8150.91 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-19 19:39:58,054 epoch 7 - iter 890/1786 - loss 0.20370059 - time (sec): 14.84 - samples/sec: 8295.45 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-19 19:40:01,168 epoch 7 - iter 1068/1786 - loss 0.19908535 - time (sec): 17.96 - samples/sec: 8292.89 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-19 19:40:04,243 epoch 7 - iter 1246/1786 - loss 0.20158080 - time (sec): 21.03 - samples/sec: 8271.76 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-19 19:40:07,312 epoch 7 - iter 1424/1786 - loss 0.20291764 - time (sec): 24.10 - samples/sec: 8212.95 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-19 19:40:10,437 epoch 7 - iter 1602/1786 - loss 0.20298768 - time (sec): 27.22 - samples/sec: 8170.91 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-19 19:40:13,465 epoch 7 - iter 1780/1786 - loss 0.20413992 - time (sec): 30.25 - samples/sec: 8202.27 - lr: 0.000017 - momentum: 0.000000
178
+ 2023-10-19 19:40:13,563 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-19 19:40:13,563 EPOCH 7 done: loss 0.2037 - lr: 0.000017
180
+ 2023-10-19 19:40:15,934 DEV : loss 0.19076649844646454 - f1-score (micro avg) 0.5546
181
+ 2023-10-19 19:40:15,948 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-19 19:40:19,099 epoch 8 - iter 178/1786 - loss 0.20287556 - time (sec): 3.15 - samples/sec: 8037.68 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-19 19:40:22,162 epoch 8 - iter 356/1786 - loss 0.20229720 - time (sec): 6.21 - samples/sec: 8081.69 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-19 19:40:25,163 epoch 8 - iter 534/1786 - loss 0.19898193 - time (sec): 9.21 - samples/sec: 7978.92 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-19 19:40:28,253 epoch 8 - iter 712/1786 - loss 0.19369768 - time (sec): 12.30 - samples/sec: 8044.90 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-19 19:40:31,293 epoch 8 - iter 890/1786 - loss 0.19392117 - time (sec): 15.34 - samples/sec: 8021.72 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-19 19:40:34,380 epoch 8 - iter 1068/1786 - loss 0.19084573 - time (sec): 18.43 - samples/sec: 8149.87 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-19 19:40:37,408 epoch 8 - iter 1246/1786 - loss 0.19257386 - time (sec): 21.46 - samples/sec: 8073.89 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-19 19:40:40,545 epoch 8 - iter 1424/1786 - loss 0.19073891 - time (sec): 24.60 - samples/sec: 8070.27 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-19 19:40:43,699 epoch 8 - iter 1602/1786 - loss 0.19251240 - time (sec): 27.75 - samples/sec: 8085.32 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-19 19:40:46,825 epoch 8 - iter 1780/1786 - loss 0.19379490 - time (sec): 30.88 - samples/sec: 8023.17 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-19 19:40:46,927 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-19 19:40:46,927 EPOCH 8 done: loss 0.1933 - lr: 0.000011
194
+ 2023-10-19 19:40:49,730 DEV : loss 0.19240671396255493 - f1-score (micro avg) 0.5782
195
+ 2023-10-19 19:40:49,743 saving best model
196
+ 2023-10-19 19:40:49,776 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-19 19:40:52,886 epoch 9 - iter 178/1786 - loss 0.18684976 - time (sec): 3.11 - samples/sec: 7832.90 - lr: 0.000011 - momentum: 0.000000
198
+ 2023-10-19 19:40:55,903 epoch 9 - iter 356/1786 - loss 0.19878074 - time (sec): 6.13 - samples/sec: 7964.77 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-19 19:40:58,822 epoch 9 - iter 534/1786 - loss 0.18798472 - time (sec): 9.04 - samples/sec: 8049.85 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-19 19:41:01,517 epoch 9 - iter 712/1786 - loss 0.18858620 - time (sec): 11.74 - samples/sec: 8388.76 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-19 19:41:04,448 epoch 9 - iter 890/1786 - loss 0.18757950 - time (sec): 14.67 - samples/sec: 8347.49 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-19 19:41:07,616 epoch 9 - iter 1068/1786 - loss 0.18804694 - time (sec): 17.84 - samples/sec: 8332.91 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-19 19:41:10,612 epoch 9 - iter 1246/1786 - loss 0.18839630 - time (sec): 20.83 - samples/sec: 8262.66 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-19 19:41:13,620 epoch 9 - iter 1424/1786 - loss 0.18752199 - time (sec): 23.84 - samples/sec: 8305.41 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-19 19:41:16,675 epoch 9 - iter 1602/1786 - loss 0.18922005 - time (sec): 26.90 - samples/sec: 8279.74 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-19 19:41:19,763 epoch 9 - iter 1780/1786 - loss 0.18631733 - time (sec): 29.99 - samples/sec: 8266.86 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-19 19:41:19,855 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-19 19:41:19,856 EPOCH 9 done: loss 0.1861 - lr: 0.000006
209
+ 2023-10-19 19:41:22,241 DEV : loss 0.19148223102092743 - f1-score (micro avg) 0.5702
210
+ 2023-10-19 19:41:22,255 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-19 19:41:25,303 epoch 10 - iter 178/1786 - loss 0.19431345 - time (sec): 3.05 - samples/sec: 8189.90 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-19 19:41:28,461 epoch 10 - iter 356/1786 - loss 0.18949315 - time (sec): 6.21 - samples/sec: 8109.99 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-19 19:41:31,510 epoch 10 - iter 534/1786 - loss 0.18810189 - time (sec): 9.25 - samples/sec: 8210.97 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-19 19:41:34,529 epoch 10 - iter 712/1786 - loss 0.18254160 - time (sec): 12.27 - samples/sec: 8095.09 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-19 19:41:37,560 epoch 10 - iter 890/1786 - loss 0.18671388 - time (sec): 15.30 - samples/sec: 8049.64 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-19 19:41:40,612 epoch 10 - iter 1068/1786 - loss 0.18696598 - time (sec): 18.36 - samples/sec: 8059.17 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-19 19:41:43,694 epoch 10 - iter 1246/1786 - loss 0.18673026 - time (sec): 21.44 - samples/sec: 8064.43 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-19 19:41:46,697 epoch 10 - iter 1424/1786 - loss 0.18361987 - time (sec): 24.44 - samples/sec: 8108.91 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-19 19:41:49,821 epoch 10 - iter 1602/1786 - loss 0.18247854 - time (sec): 27.57 - samples/sec: 8062.99 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-19 19:41:52,926 epoch 10 - iter 1780/1786 - loss 0.18118154 - time (sec): 30.67 - samples/sec: 8093.04 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-19 19:41:53,020 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-19 19:41:53,020 EPOCH 10 done: loss 0.1813 - lr: 0.000000
223
+ 2023-10-19 19:41:55,835 DEV : loss 0.19351665675640106 - f1-score (micro avg) 0.5679
224
+ 2023-10-19 19:41:55,877 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-19 19:41:55,877 Loading model from best epoch ...
226
+ 2023-10-19 19:41:55,952 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
+ 2023-10-19 19:42:00,540
228
+ Results:
229
+ - F-score (micro) 0.457
230
+ - F-score (macro) 0.297
231
+ - Accuracy 0.3051
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ LOC 0.4775 0.5333 0.5039 1095
237
+ PER 0.4620 0.5227 0.4905 1012
238
+ ORG 0.2234 0.1709 0.1937 357
239
+ HumanProd 0.0000 0.0000 0.0000 33
240
+
241
+ micro avg 0.4445 0.4702 0.4570 2497
242
+ macro avg 0.2907 0.3067 0.2970 2497
243
+ weighted avg 0.4286 0.4702 0.4474 2497
244
+
245
+ 2023-10-19 19:42:00,540 ----------------------------------------------------------------------------------------------------