Hugging Face's logo

hmbert-tiny
/
flair-hipe-2022-hipe2020-fr

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
e7fa3ea
•
1 Parent(s): 4fff24d

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.1697657143.46dc0c540dd0.3108.9 +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 sha256:42fcbcf7cbd47b9651c70bf36449cb3c3b612957c3ec5320943da1d719b414eb
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 19:26:09 0.0000 1.1200 0.3651 0.2733 0.1930 0.2263 0.1348
3
+ 2 19:26:40 0.0000 0.3944 0.2797 0.4438 0.4633 0.4533 0.3169
4
+ 3 19:27:10 0.0000 0.3243 0.2566 0.4598 0.5109 0.4840 0.3447
5
+ 4 19:27:41 0.0000 0.2808 0.2500 0.4986 0.5229 0.5105 0.3671
6
+ 5 19:28:11 0.0000 0.2542 0.2327 0.5035 0.5727 0.5359 0.3943
7
+ 6 19:28:42 0.0000 0.2325 0.2366 0.5217 0.5790 0.5489 0.4060
8
+ 7 19:29:13 0.0000 0.2189 0.2305 0.5327 0.5916 0.5607 0.4165
9
+ 8 19:29:44 0.0000 0.2038 0.2288 0.5249 0.6094 0.5640 0.4189
10
+ 9 19:30:15 0.0000 0.1978 0.2332 0.5372 0.6002 0.5669 0.4228
11
+ 10 19:30:46 0.0000 0.1932 0.2344 0.5345 0.6031 0.5667 0.4227
runs/events.out.tfevents.1697657143.46dc0c540dd0.3108.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:401151bf18af51ac0083eb8f4d6777a352d915cb132b74d2435afd5cf417e18c
3
+ size 825716
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 19:25:43,700 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 19:25:43,700 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 19:25:43,700 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 19:25:43,700 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
52
+ - NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
53
+ 2023-10-18 19:25:43,700 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 19:25:43,700 Train: 5901 sentences
55
+ 2023-10-18 19:25:43,700 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 19:25:43,701 Training Params:
58
+ 2023-10-18 19:25:43,701 - learning_rate: "5e-05"
59
+ 2023-10-18 19:25:43,701 - mini_batch_size: "4"
60
+ 2023-10-18 19:25:43,701 - max_epochs: "10"
61
+ 2023-10-18 19:25:43,701 - shuffle: "True"
62
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 19:25:43,701 Plugins:
64
+ 2023-10-18 19:25:43,701 - TensorboardLogger
65
+ 2023-10-18 19:25:43,701 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 19:25:43,701 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 19:25:43,701 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 19:25:43,701 Computation:
71
+ 2023-10-18 19:25:43,701 - compute on device: cuda:0
72
+ 2023-10-18 19:25:43,701 - embedding storage: none
73
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 19:25:43,701 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
75
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 19:25:43,701 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 19:25:43,701 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 19:25:46,078 epoch 1 - iter 147/1476 - loss 3.45082650 - time (sec): 2.38 - samples/sec: 6855.90 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-18 19:25:48,564 epoch 1 - iter 294/1476 - loss 2.95264139 - time (sec): 4.86 - samples/sec: 6711.01 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-18 19:25:50,700 epoch 1 - iter 441/1476 - loss 2.30518090 - time (sec): 7.00 - samples/sec: 7409.64 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-18 19:25:52,905 epoch 1 - iter 588/1476 - loss 1.93520085 - time (sec): 9.20 - samples/sec: 7380.55 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-18 19:25:55,256 epoch 1 - iter 735/1476 - loss 1.69071046 - time (sec): 11.55 - samples/sec: 7222.26 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-18 19:25:57,580 epoch 1 - iter 882/1476 - loss 1.53189135 - time (sec): 13.88 - samples/sec: 7155.68 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-18 19:25:59,878 epoch 1 - iter 1029/1476 - loss 1.39345233 - time (sec): 16.18 - samples/sec: 7142.25 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-18 19:26:02,193 epoch 1 - iter 1176/1476 - loss 1.29402372 - time (sec): 18.49 - samples/sec: 7083.22 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-18 19:26:04,530 epoch 1 - iter 1323/1476 - loss 1.20850834 - time (sec): 20.83 - samples/sec: 7090.57 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-18 19:26:06,892 epoch 1 - iter 1470/1476 - loss 1.12176032 - time (sec): 23.19 - samples/sec: 7158.61 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-18 19:26:06,982 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 19:26:06,982 EPOCH 1 done: loss 1.1200 - lr: 0.000050
90
+ 2023-10-18 19:26:09,849 DEV : loss 0.3651462495326996 - f1-score (micro avg) 0.2263
91
+ 2023-10-18 19:26:09,874 saving best model
92
+ 2023-10-18 19:26:09,903 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 19:26:12,157 epoch 2 - iter 147/1476 - loss 0.43364940 - time (sec): 2.25 - samples/sec: 6739.81 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-18 19:26:14,566 epoch 2 - iter 294/1476 - loss 0.44810793 - time (sec): 4.66 - samples/sec: 7098.25 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-18 19:26:16,919 epoch 2 - iter 441/1476 - loss 0.45228143 - time (sec): 7.02 - samples/sec: 7106.62 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-18 19:26:19,206 epoch 2 - iter 588/1476 - loss 0.44605435 - time (sec): 9.30 - samples/sec: 7022.19 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-18 19:26:21,537 epoch 2 - iter 735/1476 - loss 0.44203136 - time (sec): 11.63 - samples/sec: 6960.52 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-18 19:26:23,893 epoch 2 - iter 882/1476 - loss 0.42953157 - time (sec): 13.99 - samples/sec: 6888.04 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-18 19:26:26,221 epoch 2 - iter 1029/1476 - loss 0.42536159 - time (sec): 16.32 - samples/sec: 6913.01 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-18 19:26:28,640 epoch 2 - iter 1176/1476 - loss 0.41686008 - time (sec): 18.74 - samples/sec: 6973.24 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-18 19:26:31,053 epoch 2 - iter 1323/1476 - loss 0.40319699 - time (sec): 21.15 - samples/sec: 7020.40 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-18 19:26:33,422 epoch 2 - iter 1470/1476 - loss 0.39495645 - time (sec): 23.52 - samples/sec: 7055.20 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-18 19:26:33,504 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 19:26:33,504 EPOCH 2 done: loss 0.3944 - lr: 0.000044
105
+ 2023-10-18 19:26:40,653 DEV : loss 0.27972713112831116 - f1-score (micro avg) 0.4533
106
+ 2023-10-18 19:26:40,679 saving best model
107
+ 2023-10-18 19:26:40,712 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-18 19:26:42,802 epoch 3 - iter 147/1476 - loss 0.31787561 - time (sec): 2.09 - samples/sec: 7349.79 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-18 19:26:44,882 epoch 3 - iter 294/1476 - loss 0.31847873 - time (sec): 4.17 - samples/sec: 7727.01 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-18 19:26:46,953 epoch 3 - iter 441/1476 - loss 0.32050732 - time (sec): 6.24 - samples/sec: 7800.13 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-18 19:26:49,069 epoch 3 - iter 588/1476 - loss 0.33945722 - time (sec): 8.36 - samples/sec: 7971.13 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-18 19:26:51,403 epoch 3 - iter 735/1476 - loss 0.33243206 - time (sec): 10.69 - samples/sec: 7886.43 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-18 19:26:53,795 epoch 3 - iter 882/1476 - loss 0.33011776 - time (sec): 13.08 - samples/sec: 7712.96 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-18 19:26:56,101 epoch 3 - iter 1029/1476 - loss 0.32825472 - time (sec): 15.39 - samples/sec: 7623.57 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-18 19:26:58,429 epoch 3 - iter 1176/1476 - loss 0.32855013 - time (sec): 17.72 - samples/sec: 7510.53 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-18 19:27:00,729 epoch 3 - iter 1323/1476 - loss 0.32623173 - time (sec): 20.02 - samples/sec: 7447.76 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-18 19:27:03,055 epoch 3 - iter 1470/1476 - loss 0.32429762 - time (sec): 22.34 - samples/sec: 7420.44 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-18 19:27:03,150 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 19:27:03,150 EPOCH 3 done: loss 0.3243 - lr: 0.000039
120
+ 2023-10-18 19:27:10,343 DEV : loss 0.2565878927707672 - f1-score (micro avg) 0.484
121
+ 2023-10-18 19:27:10,369 saving best model
122
+ 2023-10-18 19:27:10,401 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-18 19:27:12,751 epoch 4 - iter 147/1476 - loss 0.31044806 - time (sec): 2.35 - samples/sec: 7943.66 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-18 19:27:15,078 epoch 4 - iter 294/1476 - loss 0.30521810 - time (sec): 4.68 - samples/sec: 7449.04 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-18 19:27:17,467 epoch 4 - iter 441/1476 - loss 0.30286129 - time (sec): 7.06 - samples/sec: 7165.86 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-18 19:27:19,860 epoch 4 - iter 588/1476 - loss 0.29258445 - time (sec): 9.46 - samples/sec: 7050.26 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-18 19:27:22,211 epoch 4 - iter 735/1476 - loss 0.29065507 - time (sec): 11.81 - samples/sec: 7004.00 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-18 19:27:24,484 epoch 4 - iter 882/1476 - loss 0.28962621 - time (sec): 14.08 - samples/sec: 6895.49 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-18 19:27:26,848 epoch 4 - iter 1029/1476 - loss 0.28847591 - time (sec): 16.45 - samples/sec: 7154.21 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-18 19:27:29,219 epoch 4 - iter 1176/1476 - loss 0.28570559 - time (sec): 18.82 - samples/sec: 7139.19 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-18 19:27:31,547 epoch 4 - iter 1323/1476 - loss 0.28152220 - time (sec): 21.14 - samples/sec: 7085.25 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-18 19:27:33,865 epoch 4 - iter 1470/1476 - loss 0.28053742 - time (sec): 23.46 - samples/sec: 7069.46 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-18 19:27:33,955 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-18 19:27:33,955 EPOCH 4 done: loss 0.2808 - lr: 0.000033
135
+ 2023-10-18 19:27:41,133 DEV : loss 0.25001299381256104 - f1-score (micro avg) 0.5105
136
+ 2023-10-18 19:27:41,160 saving best model
137
+ 2023-10-18 19:27:41,194 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-18 19:27:43,589 epoch 5 - iter 147/1476 - loss 0.24307097 - time (sec): 2.39 - samples/sec: 6941.11 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-18 19:27:45,981 epoch 5 - iter 294/1476 - loss 0.25891254 - time (sec): 4.79 - samples/sec: 7440.31 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-18 19:27:48,219 epoch 5 - iter 441/1476 - loss 0.26161265 - time (sec): 7.02 - samples/sec: 7284.10 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-18 19:27:50,608 epoch 5 - iter 588/1476 - loss 0.26397432 - time (sec): 9.41 - samples/sec: 7301.00 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-18 19:27:52,845 epoch 5 - iter 735/1476 - loss 0.26076338 - time (sec): 11.65 - samples/sec: 7331.47 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-18 19:27:55,258 epoch 5 - iter 882/1476 - loss 0.25857998 - time (sec): 14.06 - samples/sec: 7336.21 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-18 19:27:57,642 epoch 5 - iter 1029/1476 - loss 0.25772969 - time (sec): 16.45 - samples/sec: 7238.18 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-18 19:27:59,941 epoch 5 - iter 1176/1476 - loss 0.25641948 - time (sec): 18.75 - samples/sec: 7173.24 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-18 19:28:02,276 epoch 5 - iter 1323/1476 - loss 0.25486743 - time (sec): 21.08 - samples/sec: 7169.23 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-18 19:28:04,600 epoch 5 - iter 1470/1476 - loss 0.25455365 - time (sec): 23.41 - samples/sec: 7086.76 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-18 19:28:04,696 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-18 19:28:04,696 EPOCH 5 done: loss 0.2542 - lr: 0.000028
150
+ 2023-10-18 19:28:11,973 DEV : loss 0.23268112540245056 - f1-score (micro avg) 0.5359
151
+ 2023-10-18 19:28:11,999 saving best model
152
+ 2023-10-18 19:28:12,037 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-18 19:28:14,319 epoch 6 - iter 147/1476 - loss 0.22702365 - time (sec): 2.28 - samples/sec: 6646.41 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-18 19:28:16,657 epoch 6 - iter 294/1476 - loss 0.22872241 - time (sec): 4.62 - samples/sec: 6983.40 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-18 19:28:19,038 epoch 6 - iter 441/1476 - loss 0.23060571 - time (sec): 7.00 - samples/sec: 7102.09 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-18 19:28:21,408 epoch 6 - iter 588/1476 - loss 0.22128491 - time (sec): 9.37 - samples/sec: 7211.04 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-18 19:28:23,834 epoch 6 - iter 735/1476 - loss 0.23096293 - time (sec): 11.80 - samples/sec: 7258.92 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-18 19:28:26,140 epoch 6 - iter 882/1476 - loss 0.23662672 - time (sec): 14.10 - samples/sec: 7219.87 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-18 19:28:28,585 epoch 6 - iter 1029/1476 - loss 0.23652904 - time (sec): 16.55 - samples/sec: 7056.48 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-18 19:28:30,842 epoch 6 - iter 1176/1476 - loss 0.23746363 - time (sec): 18.80 - samples/sec: 7043.73 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-18 19:28:33,211 epoch 6 - iter 1323/1476 - loss 0.23598701 - time (sec): 21.17 - samples/sec: 7095.16 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-18 19:28:35,508 epoch 6 - iter 1470/1476 - loss 0.23285328 - time (sec): 23.47 - samples/sec: 7060.49 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-18 19:28:35,599 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-18 19:28:35,600 EPOCH 6 done: loss 0.2325 - lr: 0.000022
165
+ 2023-10-18 19:28:42,848 DEV : loss 0.23657557368278503 - f1-score (micro avg) 0.5489
166
+ 2023-10-18 19:28:42,874 saving best model
167
+ 2023-10-18 19:28:42,913 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-18 19:28:45,188 epoch 7 - iter 147/1476 - loss 0.20843898 - time (sec): 2.28 - samples/sec: 6965.88 - lr: 0.000022 - momentum: 0.000000
169
+ 2023-10-18 19:28:47,506 epoch 7 - iter 294/1476 - loss 0.19848750 - time (sec): 4.59 - samples/sec: 6917.41 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-18 19:28:49,793 epoch 7 - iter 441/1476 - loss 0.20213122 - time (sec): 6.88 - samples/sec: 6857.08 - lr: 0.000021 - momentum: 0.000000
171
+ 2023-10-18 19:28:52,149 epoch 7 - iter 588/1476 - loss 0.20460332 - time (sec): 9.24 - samples/sec: 6881.99 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-18 19:28:54,545 epoch 7 - iter 735/1476 - loss 0.20278281 - time (sec): 11.63 - samples/sec: 6900.60 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-18 19:28:56,668 epoch 7 - iter 882/1476 - loss 0.20307092 - time (sec): 13.75 - samples/sec: 6982.65 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-18 19:28:58,943 epoch 7 - iter 1029/1476 - loss 0.20304549 - time (sec): 16.03 - samples/sec: 7078.70 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-18 19:29:01,285 epoch 7 - iter 1176/1476 - loss 0.21048904 - time (sec): 18.37 - samples/sec: 7076.45 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-18 19:29:03,662 epoch 7 - iter 1323/1476 - loss 0.21393619 - time (sec): 20.75 - samples/sec: 7103.87 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-18 19:29:06,067 epoch 7 - iter 1470/1476 - loss 0.21913870 - time (sec): 23.15 - samples/sec: 7163.04 - lr: 0.000017 - momentum: 0.000000
178
+ 2023-10-18 19:29:06,157 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-18 19:29:06,157 EPOCH 7 done: loss 0.2189 - lr: 0.000017
180
+ 2023-10-18 19:29:13,450 DEV : loss 0.23048946261405945 - f1-score (micro avg) 0.5607
181
+ 2023-10-18 19:29:13,476 saving best model
182
+ 2023-10-18 19:29:13,514 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-18 19:29:15,943 epoch 8 - iter 147/1476 - loss 0.22164479 - time (sec): 2.43 - samples/sec: 7841.51 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-18 19:29:18,259 epoch 8 - iter 294/1476 - loss 0.21612436 - time (sec): 4.74 - samples/sec: 7213.85 - lr: 0.000016 - momentum: 0.000000
185
+ 2023-10-18 19:29:20,610 epoch 8 - iter 441/1476 - loss 0.20551481 - time (sec): 7.10 - samples/sec: 7205.49 - lr: 0.000015 - momentum: 0.000000
186
+ 2023-10-18 19:29:22,942 epoch 8 - iter 588/1476 - loss 0.20444008 - time (sec): 9.43 - samples/sec: 7104.72 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-18 19:29:25,290 epoch 8 - iter 735/1476 - loss 0.20490690 - time (sec): 11.78 - samples/sec: 7023.69 - lr: 0.000014 - momentum: 0.000000
188
+ 2023-10-18 19:29:27,633 epoch 8 - iter 882/1476 - loss 0.20394011 - time (sec): 14.12 - samples/sec: 6971.94 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-18 19:29:29,954 epoch 8 - iter 1029/1476 - loss 0.20450972 - time (sec): 16.44 - samples/sec: 6981.55 - lr: 0.000013 - momentum: 0.000000
190
+ 2023-10-18 19:29:32,268 epoch 8 - iter 1176/1476 - loss 0.20532453 - time (sec): 18.75 - samples/sec: 6988.05 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-18 19:29:34,611 epoch 8 - iter 1323/1476 - loss 0.20424926 - time (sec): 21.10 - samples/sec: 7020.74 - lr: 0.000012 - momentum: 0.000000
192
+ 2023-10-18 19:29:37,049 epoch 8 - iter 1470/1476 - loss 0.20419251 - time (sec): 23.53 - samples/sec: 7048.45 - lr: 0.000011 - momentum: 0.000000
193
+ 2023-10-18 19:29:37,132 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-18 19:29:37,132 EPOCH 8 done: loss 0.2038 - lr: 0.000011
195
+ 2023-10-18 19:29:44,526 DEV : loss 0.2287674993276596 - f1-score (micro avg) 0.564
196
+ 2023-10-18 19:29:44,553 saving best model
197
+ 2023-10-18 19:29:44,588 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-18 19:29:46,979 epoch 9 - iter 147/1476 - loss 0.24117374 - time (sec): 2.39 - samples/sec: 7758.64 - lr: 0.000011 - momentum: 0.000000
199
+ 2023-10-18 19:29:49,330 epoch 9 - iter 294/1476 - loss 0.20776964 - time (sec): 4.74 - samples/sec: 7452.71 - lr: 0.000010 - momentum: 0.000000
200
+ 2023-10-18 19:29:51,767 epoch 9 - iter 441/1476 - loss 0.19920249 - time (sec): 7.18 - samples/sec: 7490.06 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-18 19:29:54,131 epoch 9 - iter 588/1476 - loss 0.19959637 - time (sec): 9.54 - samples/sec: 7400.95 - lr: 0.000009 - momentum: 0.000000
202
+ 2023-10-18 19:29:56,441 epoch 9 - iter 735/1476 - loss 0.20235990 - time (sec): 11.85 - samples/sec: 7218.41 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-18 19:29:58,781 epoch 9 - iter 882/1476 - loss 0.20008424 - time (sec): 14.19 - samples/sec: 7133.15 - lr: 0.000008 - momentum: 0.000000
204
+ 2023-10-18 19:30:01,113 epoch 9 - iter 1029/1476 - loss 0.19992171 - time (sec): 16.52 - samples/sec: 7066.75 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-18 19:30:03,499 epoch 9 - iter 1176/1476 - loss 0.20005741 - time (sec): 18.91 - samples/sec: 7012.25 - lr: 0.000007 - momentum: 0.000000
206
+ 2023-10-18 19:30:05,824 epoch 9 - iter 1323/1476 - loss 0.19939868 - time (sec): 21.24 - samples/sec: 6998.13 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-18 19:30:08,170 epoch 9 - iter 1470/1476 - loss 0.19779017 - time (sec): 23.58 - samples/sec: 7031.87 - lr: 0.000006 - momentum: 0.000000
208
+ 2023-10-18 19:30:08,259 ----------------------------------------------------------------------------------------------------
209
+ 2023-10-18 19:30:08,259 EPOCH 9 done: loss 0.1978 - lr: 0.000006
210
+ 2023-10-18 19:30:15,511 DEV : loss 0.23322512209415436 - f1-score (micro avg) 0.5669
211
+ 2023-10-18 19:30:15,537 saving best model
212
+ 2023-10-18 19:30:15,575 ----------------------------------------------------------------------------------------------------
213
+ 2023-10-18 19:30:17,903 epoch 10 - iter 147/1476 - loss 0.17627206 - time (sec): 2.33 - samples/sec: 6928.85 - lr: 0.000005 - momentum: 0.000000
214
+ 2023-10-18 19:30:20,279 epoch 10 - iter 294/1476 - loss 0.19028562 - time (sec): 4.70 - samples/sec: 6917.02 - lr: 0.000004 - momentum: 0.000000
215
+ 2023-10-18 19:30:22,714 epoch 10 - iter 441/1476 - loss 0.20663725 - time (sec): 7.14 - samples/sec: 7221.33 - lr: 0.000004 - momentum: 0.000000
216
+ 2023-10-18 19:30:25,056 epoch 10 - iter 588/1476 - loss 0.20586196 - time (sec): 9.48 - samples/sec: 7218.37 - lr: 0.000003 - momentum: 0.000000
217
+ 2023-10-18 19:30:27,349 epoch 10 - iter 735/1476 - loss 0.19710005 - time (sec): 11.77 - samples/sec: 7161.12 - lr: 0.000003 - momentum: 0.000000
218
+ 2023-10-18 19:30:29,703 epoch 10 - iter 882/1476 - loss 0.19316847 - time (sec): 14.13 - samples/sec: 7033.32 - lr: 0.000002 - momentum: 0.000000
219
+ 2023-10-18 19:30:32,069 epoch 10 - iter 1029/1476 - loss 0.19035206 - time (sec): 16.49 - samples/sec: 7108.95 - lr: 0.000002 - momentum: 0.000000
220
+ 2023-10-18 19:30:34,472 epoch 10 - iter 1176/1476 - loss 0.19179760 - time (sec): 18.90 - samples/sec: 7073.46 - lr: 0.000001 - momentum: 0.000000
221
+ 2023-10-18 19:30:36,816 epoch 10 - iter 1323/1476 - loss 0.19167956 - time (sec): 21.24 - samples/sec: 7103.06 - lr: 0.000001 - momentum: 0.000000
222
+ 2023-10-18 19:30:39,093 epoch 10 - iter 1470/1476 - loss 0.19294142 - time (sec): 23.52 - samples/sec: 7051.22 - lr: 0.000000 - momentum: 0.000000
223
+ 2023-10-18 19:30:39,180 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-18 19:30:39,180 EPOCH 10 done: loss 0.1932 - lr: 0.000000
225
+ 2023-10-18 19:30:46,482 DEV : loss 0.23437848687171936 - f1-score (micro avg) 0.5667
226
+ 2023-10-18 19:30:46,542 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-18 19:30:46,542 Loading model from best epoch ...
228
+ 2023-10-18 19:30:46,624 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-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
229
+ 2023-10-18 19:30:49,805
230
+ Results:
231
+ - F-score (micro) 0.5554
232
+ - F-score (macro) 0.3405
233
+ - Accuracy 0.4067
234
+
235
+ By class:
236
+ precision recall f1-score support
237
+
238
+ loc 0.5913 0.7622 0.6660 858
239
+ pers 0.4338 0.5493 0.4848 537
240
+ org 0.2063 0.0985 0.1333 132
241
+ time 0.4107 0.4259 0.4182 54
242
+ prod 0.0000 0.0000 0.0000 61
243
+
244
+ micro avg 0.5171 0.5999 0.5554 1642
245
+ macro avg 0.3284 0.3672 0.3405 1642
246
+ weighted avg 0.4810 0.5999 0.5310 1642
247
+
248
+ 2023-10-18 19:30:49,805 ----------------------------------------------------------------------------------------------------