Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
874173c
·
1 Parent(s): 308d4d1

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 sha256:8f9ce099c45041df3832167bd3afda46739ad49fc8d9fa2a9ae60a5f6b2dc48a
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 22:55:19 0.0000 0.2930 0.1094 0.5364 0.6911 0.6040 0.4364
3
+ 2 22:57:18 0.0000 0.0833 0.1233 0.5400 0.7963 0.6436 0.4820
4
+ 3 22:59:16 0.0000 0.0563 0.1788 0.5482 0.7414 0.6304 0.4679
5
+ 4 23:01:14 0.0000 0.0407 0.2125 0.5388 0.7860 0.6394 0.4791
6
+ 5 23:03:12 0.0000 0.0297 0.3090 0.5214 0.8227 0.6383 0.4777
7
+ 6 23:05:09 0.0000 0.0211 0.3297 0.5604 0.7323 0.6349 0.4720
8
+ 7 23:07:10 0.0000 0.0146 0.3629 0.5518 0.7918 0.6504 0.4897
9
+ 8 23:09:13 0.0000 0.0100 0.3726 0.5480 0.7975 0.6496 0.4891
10
+ 9 23:11:10 0.0000 0.0063 0.3872 0.5509 0.7860 0.6478 0.4872
11
+ 10 23:13:08 0.0000 0.0044 0.3975 0.5548 0.7941 0.6532 0.4932
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 22:53:22,080 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 22:53:22,081 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 22:53:22,081 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 22:53:22,081 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
52
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
53
+ 2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 22:53:22,081 Train: 14465 sentences
55
+ 2023-10-14 22:53:22,081 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 22:53:22,081 Training Params:
58
+ 2023-10-14 22:53:22,081 - learning_rate: "3e-05"
59
+ 2023-10-14 22:53:22,081 - mini_batch_size: "8"
60
+ 2023-10-14 22:53:22,081 - max_epochs: "10"
61
+ 2023-10-14 22:53:22,081 - shuffle: "True"
62
+ 2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 22:53:22,081 Plugins:
64
+ 2023-10-14 22:53:22,081 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 22:53:22,082 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 22:53:22,082 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 22:53:22,082 Computation:
70
+ 2023-10-14 22:53:22,082 - compute on device: cuda:0
71
+ 2023-10-14 22:53:22,082 - embedding storage: none
72
+ 2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 22:53:22,082 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 22:53:33,152 epoch 1 - iter 180/1809 - loss 1.67027407 - time (sec): 11.07 - samples/sec: 3457.13 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 22:53:44,247 epoch 1 - iter 360/1809 - loss 0.96696948 - time (sec): 22.16 - samples/sec: 3425.79 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 22:53:55,940 epoch 1 - iter 540/1809 - loss 0.70638013 - time (sec): 33.86 - samples/sec: 3331.88 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 22:54:07,179 epoch 1 - iter 720/1809 - loss 0.55894456 - time (sec): 45.10 - samples/sec: 3377.37 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 22:54:18,106 epoch 1 - iter 900/1809 - loss 0.47409400 - time (sec): 56.02 - samples/sec: 3391.19 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 22:54:29,388 epoch 1 - iter 1080/1809 - loss 0.41428178 - time (sec): 67.31 - samples/sec: 3405.66 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 22:54:40,232 epoch 1 - iter 1260/1809 - loss 0.37349997 - time (sec): 78.15 - samples/sec: 3406.85 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 22:54:51,158 epoch 1 - iter 1440/1809 - loss 0.34066974 - time (sec): 89.08 - samples/sec: 3410.54 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 22:55:02,254 epoch 1 - iter 1620/1809 - loss 0.31439415 - time (sec): 100.17 - samples/sec: 3403.63 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 22:55:13,151 epoch 1 - iter 1800/1809 - loss 0.29381810 - time (sec): 111.07 - samples/sec: 3406.22 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 22:55:13,661 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 22:55:13,661 EPOCH 1 done: loss 0.2930 - lr: 0.000030
88
+ 2023-10-14 22:55:19,299 DEV : loss 0.10940668731927872 - f1-score (micro avg) 0.604
89
+ 2023-10-14 22:55:19,334 saving best model
90
+ 2023-10-14 22:55:19,823 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 22:55:30,992 epoch 2 - iter 180/1809 - loss 0.08979228 - time (sec): 11.17 - samples/sec: 3317.64 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 22:55:42,224 epoch 2 - iter 360/1809 - loss 0.08989058 - time (sec): 22.40 - samples/sec: 3360.44 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 22:55:53,269 epoch 2 - iter 540/1809 - loss 0.08901176 - time (sec): 33.44 - samples/sec: 3395.78 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 22:56:04,648 epoch 2 - iter 720/1809 - loss 0.08769618 - time (sec): 44.82 - samples/sec: 3400.92 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 22:56:15,731 epoch 2 - iter 900/1809 - loss 0.08657795 - time (sec): 55.91 - samples/sec: 3423.49 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 22:56:26,940 epoch 2 - iter 1080/1809 - loss 0.08614963 - time (sec): 67.11 - samples/sec: 3412.98 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 22:56:37,915 epoch 2 - iter 1260/1809 - loss 0.08650169 - time (sec): 78.09 - samples/sec: 3409.66 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 22:56:48,753 epoch 2 - iter 1440/1809 - loss 0.08591524 - time (sec): 88.93 - samples/sec: 3397.15 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 22:57:00,131 epoch 2 - iter 1620/1809 - loss 0.08467002 - time (sec): 100.31 - samples/sec: 3398.25 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 22:57:10,944 epoch 2 - iter 1800/1809 - loss 0.08346842 - time (sec): 111.12 - samples/sec: 3402.53 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 22:57:11,479 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 22:57:11,479 EPOCH 2 done: loss 0.0833 - lr: 0.000027
103
+ 2023-10-14 22:57:18,912 DEV : loss 0.12328627705574036 - f1-score (micro avg) 0.6436
104
+ 2023-10-14 22:57:18,946 saving best model
105
+ 2023-10-14 22:57:19,426 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 22:57:31,144 epoch 3 - iter 180/1809 - loss 0.05050900 - time (sec): 11.72 - samples/sec: 3299.37 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 22:57:42,023 epoch 3 - iter 360/1809 - loss 0.05640218 - time (sec): 22.59 - samples/sec: 3385.87 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 22:57:52,973 epoch 3 - iter 540/1809 - loss 0.05884620 - time (sec): 33.54 - samples/sec: 3385.83 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-14 22:58:03,993 epoch 3 - iter 720/1809 - loss 0.05715791 - time (sec): 44.57 - samples/sec: 3417.03 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 22:58:15,316 epoch 3 - iter 900/1809 - loss 0.05631954 - time (sec): 55.89 - samples/sec: 3394.29 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 22:58:26,071 epoch 3 - iter 1080/1809 - loss 0.05690696 - time (sec): 66.64 - samples/sec: 3409.30 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-14 22:58:37,282 epoch 3 - iter 1260/1809 - loss 0.05623779 - time (sec): 77.85 - samples/sec: 3404.87 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 22:58:48,305 epoch 3 - iter 1440/1809 - loss 0.05612202 - time (sec): 88.88 - samples/sec: 3399.30 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 22:58:59,489 epoch 3 - iter 1620/1809 - loss 0.05782234 - time (sec): 100.06 - samples/sec: 3403.20 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-14 22:59:10,384 epoch 3 - iter 1800/1809 - loss 0.05647561 - time (sec): 110.96 - samples/sec: 3408.09 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-14 22:59:11,005 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 22:59:11,005 EPOCH 3 done: loss 0.0563 - lr: 0.000023
118
+ 2023-10-14 22:59:16,771 DEV : loss 0.1787535846233368 - f1-score (micro avg) 0.6304
119
+ 2023-10-14 22:59:16,818 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-14 22:59:28,293 epoch 4 - iter 180/1809 - loss 0.03058125 - time (sec): 11.47 - samples/sec: 3393.40 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-14 22:59:40,346 epoch 4 - iter 360/1809 - loss 0.03335145 - time (sec): 23.53 - samples/sec: 3247.71 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 22:59:51,427 epoch 4 - iter 540/1809 - loss 0.03771304 - time (sec): 34.61 - samples/sec: 3305.71 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-14 23:00:02,403 epoch 4 - iter 720/1809 - loss 0.03909361 - time (sec): 45.58 - samples/sec: 3315.77 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 23:00:13,244 epoch 4 - iter 900/1809 - loss 0.03854045 - time (sec): 56.42 - samples/sec: 3339.92 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 23:00:24,493 epoch 4 - iter 1080/1809 - loss 0.03909550 - time (sec): 67.67 - samples/sec: 3349.27 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-14 23:00:35,558 epoch 4 - iter 1260/1809 - loss 0.03949444 - time (sec): 78.74 - samples/sec: 3361.35 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 23:00:46,652 epoch 4 - iter 1440/1809 - loss 0.03909637 - time (sec): 89.83 - samples/sec: 3374.67 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 23:00:57,583 epoch 4 - iter 1620/1809 - loss 0.03977330 - time (sec): 100.76 - samples/sec: 3381.39 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-14 23:01:08,579 epoch 4 - iter 1800/1809 - loss 0.04051474 - time (sec): 111.76 - samples/sec: 3382.87 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 23:01:09,100 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-14 23:01:09,100 EPOCH 4 done: loss 0.0407 - lr: 0.000020
132
+ 2023-10-14 23:01:14,916 DEV : loss 0.212530717253685 - f1-score (micro avg) 0.6394
133
+ 2023-10-14 23:01:14,965 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-14 23:01:26,608 epoch 5 - iter 180/1809 - loss 0.03110605 - time (sec): 11.64 - samples/sec: 3244.49 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-14 23:01:37,407 epoch 5 - iter 360/1809 - loss 0.02633661 - time (sec): 22.44 - samples/sec: 3367.39 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-14 23:01:48,476 epoch 5 - iter 540/1809 - loss 0.02512481 - time (sec): 33.51 - samples/sec: 3381.73 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-14 23:01:59,358 epoch 5 - iter 720/1809 - loss 0.02642180 - time (sec): 44.39 - samples/sec: 3378.28 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 23:02:10,472 epoch 5 - iter 900/1809 - loss 0.02681334 - time (sec): 55.51 - samples/sec: 3383.01 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-14 23:02:21,350 epoch 5 - iter 1080/1809 - loss 0.02711812 - time (sec): 66.38 - samples/sec: 3387.39 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-14 23:02:31,971 epoch 5 - iter 1260/1809 - loss 0.02724187 - time (sec): 77.00 - samples/sec: 3399.19 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 23:02:43,223 epoch 5 - iter 1440/1809 - loss 0.02806269 - time (sec): 88.26 - samples/sec: 3411.19 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-14 23:02:55,040 epoch 5 - iter 1620/1809 - loss 0.02837556 - time (sec): 100.07 - samples/sec: 3395.44 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-14 23:03:06,144 epoch 5 - iter 1800/1809 - loss 0.02959314 - time (sec): 111.18 - samples/sec: 3402.73 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 23:03:06,663 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-14 23:03:06,663 EPOCH 5 done: loss 0.0297 - lr: 0.000017
146
+ 2023-10-14 23:03:12,415 DEV : loss 0.30899283289909363 - f1-score (micro avg) 0.6383
147
+ 2023-10-14 23:03:12,460 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-14 23:03:23,454 epoch 6 - iter 180/1809 - loss 0.01975153 - time (sec): 10.99 - samples/sec: 3268.70 - lr: 0.000016 - momentum: 0.000000
149
+ 2023-10-14 23:03:34,423 epoch 6 - iter 360/1809 - loss 0.02159189 - time (sec): 21.96 - samples/sec: 3373.45 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-14 23:03:45,156 epoch 6 - iter 540/1809 - loss 0.02205122 - time (sec): 32.69 - samples/sec: 3395.68 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-14 23:03:55,922 epoch 6 - iter 720/1809 - loss 0.02227128 - time (sec): 43.46 - samples/sec: 3422.32 - lr: 0.000015 - momentum: 0.000000
152
+ 2023-10-14 23:04:06,988 epoch 6 - iter 900/1809 - loss 0.02215978 - time (sec): 54.53 - samples/sec: 3437.87 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-14 23:04:18,180 epoch 6 - iter 1080/1809 - loss 0.02229865 - time (sec): 65.72 - samples/sec: 3431.52 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-14 23:04:29,141 epoch 6 - iter 1260/1809 - loss 0.02131826 - time (sec): 76.68 - samples/sec: 3450.76 - lr: 0.000014 - momentum: 0.000000
155
+ 2023-10-14 23:04:40,072 epoch 6 - iter 1440/1809 - loss 0.02100046 - time (sec): 87.61 - samples/sec: 3471.36 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-14 23:04:50,521 epoch 6 - iter 1620/1809 - loss 0.02171985 - time (sec): 98.06 - samples/sec: 3468.48 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-14 23:05:01,563 epoch 6 - iter 1800/1809 - loss 0.02103705 - time (sec): 109.10 - samples/sec: 3465.50 - lr: 0.000013 - momentum: 0.000000
158
+ 2023-10-14 23:05:02,125 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-14 23:05:02,125 EPOCH 6 done: loss 0.0211 - lr: 0.000013
160
+ 2023-10-14 23:05:09,525 DEV : loss 0.3297453224658966 - f1-score (micro avg) 0.6349
161
+ 2023-10-14 23:05:09,564 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-14 23:05:21,406 epoch 7 - iter 180/1809 - loss 0.01552725 - time (sec): 11.84 - samples/sec: 3136.20 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-14 23:05:32,981 epoch 7 - iter 360/1809 - loss 0.01402831 - time (sec): 23.42 - samples/sec: 3263.96 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-14 23:05:44,453 epoch 7 - iter 540/1809 - loss 0.01360336 - time (sec): 34.89 - samples/sec: 3314.74 - lr: 0.000012 - momentum: 0.000000
165
+ 2023-10-14 23:05:55,765 epoch 7 - iter 720/1809 - loss 0.01494343 - time (sec): 46.20 - samples/sec: 3294.49 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-14 23:06:06,988 epoch 7 - iter 900/1809 - loss 0.01605299 - time (sec): 57.42 - samples/sec: 3310.58 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-14 23:06:17,896 epoch 7 - iter 1080/1809 - loss 0.01580151 - time (sec): 68.33 - samples/sec: 3335.50 - lr: 0.000011 - momentum: 0.000000
168
+ 2023-10-14 23:06:29,114 epoch 7 - iter 1260/1809 - loss 0.01590065 - time (sec): 79.55 - samples/sec: 3352.47 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-14 23:06:39,900 epoch 7 - iter 1440/1809 - loss 0.01510279 - time (sec): 90.33 - samples/sec: 3348.18 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-14 23:06:51,240 epoch 7 - iter 1620/1809 - loss 0.01461003 - time (sec): 101.67 - samples/sec: 3346.97 - lr: 0.000010 - momentum: 0.000000
171
+ 2023-10-14 23:07:02,417 epoch 7 - iter 1800/1809 - loss 0.01467349 - time (sec): 112.85 - samples/sec: 3351.90 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-14 23:07:02,945 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-14 23:07:02,945 EPOCH 7 done: loss 0.0146 - lr: 0.000010
174
+ 2023-10-14 23:07:10,634 DEV : loss 0.3628890812397003 - f1-score (micro avg) 0.6504
175
+ 2023-10-14 23:07:10,675 saving best model
176
+ 2023-10-14 23:07:11,201 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-14 23:07:22,863 epoch 8 - iter 180/1809 - loss 0.00662021 - time (sec): 11.66 - samples/sec: 3155.29 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-14 23:07:34,630 epoch 8 - iter 360/1809 - loss 0.00795343 - time (sec): 23.43 - samples/sec: 3215.56 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-14 23:07:46,178 epoch 8 - iter 540/1809 - loss 0.00814355 - time (sec): 34.97 - samples/sec: 3244.02 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-14 23:07:57,360 epoch 8 - iter 720/1809 - loss 0.00832031 - time (sec): 46.16 - samples/sec: 3271.35 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-14 23:08:08,704 epoch 8 - iter 900/1809 - loss 0.00816301 - time (sec): 57.50 - samples/sec: 3270.22 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-14 23:08:20,051 epoch 8 - iter 1080/1809 - loss 0.00860007 - time (sec): 68.85 - samples/sec: 3288.75 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-14 23:08:31,136 epoch 8 - iter 1260/1809 - loss 0.00904243 - time (sec): 79.93 - samples/sec: 3313.67 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-14 23:08:42,429 epoch 8 - iter 1440/1809 - loss 0.00953422 - time (sec): 91.23 - samples/sec: 3313.24 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-14 23:08:53,796 epoch 8 - iter 1620/1809 - loss 0.01000057 - time (sec): 102.59 - samples/sec: 3308.69 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-14 23:09:05,437 epoch 8 - iter 1800/1809 - loss 0.01005559 - time (sec): 114.23 - samples/sec: 3312.91 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-14 23:09:05,941 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-14 23:09:05,941 EPOCH 8 done: loss 0.0100 - lr: 0.000007
189
+ 2023-10-14 23:09:13,141 DEV : loss 0.37255582213401794 - f1-score (micro avg) 0.6496
190
+ 2023-10-14 23:09:13,188 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-14 23:09:24,752 epoch 9 - iter 180/1809 - loss 0.00494447 - time (sec): 11.56 - samples/sec: 3303.26 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-14 23:09:36,272 epoch 9 - iter 360/1809 - loss 0.00585614 - time (sec): 23.08 - samples/sec: 3278.56 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-14 23:09:47,503 epoch 9 - iter 540/1809 - loss 0.00614073 - time (sec): 34.31 - samples/sec: 3302.47 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-14 23:09:58,681 epoch 9 - iter 720/1809 - loss 0.00551232 - time (sec): 45.49 - samples/sec: 3340.65 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-14 23:10:09,760 epoch 9 - iter 900/1809 - loss 0.00625027 - time (sec): 56.57 - samples/sec: 3357.66 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-14 23:10:20,942 epoch 9 - iter 1080/1809 - loss 0.00619889 - time (sec): 67.75 - samples/sec: 3379.13 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-14 23:10:31,916 epoch 9 - iter 1260/1809 - loss 0.00604495 - time (sec): 78.73 - samples/sec: 3390.61 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-14 23:10:42,708 epoch 9 - iter 1440/1809 - loss 0.00630439 - time (sec): 89.52 - samples/sec: 3391.73 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-14 23:10:53,793 epoch 9 - iter 1620/1809 - loss 0.00656510 - time (sec): 100.60 - samples/sec: 3397.99 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-14 23:11:04,500 epoch 9 - iter 1800/1809 - loss 0.00635203 - time (sec): 111.31 - samples/sec: 3397.12 - lr: 0.000003 - momentum: 0.000000
201
+ 2023-10-14 23:11:05,069 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-14 23:11:05,069 EPOCH 9 done: loss 0.0063 - lr: 0.000003
203
+ 2023-10-14 23:11:10,696 DEV : loss 0.38722819089889526 - f1-score (micro avg) 0.6478
204
+ 2023-10-14 23:11:10,731 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-14 23:11:23,051 epoch 10 - iter 180/1809 - loss 0.00392312 - time (sec): 12.32 - samples/sec: 3093.81 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-14 23:11:33,944 epoch 10 - iter 360/1809 - loss 0.00404755 - time (sec): 23.21 - samples/sec: 3274.69 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-14 23:11:44,860 epoch 10 - iter 540/1809 - loss 0.00406751 - time (sec): 34.13 - samples/sec: 3311.50 - lr: 0.000002 - momentum: 0.000000
208
+ 2023-10-14 23:11:55,986 epoch 10 - iter 720/1809 - loss 0.00488523 - time (sec): 45.25 - samples/sec: 3346.71 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-14 23:12:07,005 epoch 10 - iter 900/1809 - loss 0.00447546 - time (sec): 56.27 - samples/sec: 3372.90 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-14 23:12:17,916 epoch 10 - iter 1080/1809 - loss 0.00409456 - time (sec): 67.18 - samples/sec: 3382.70 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-14 23:12:29,157 epoch 10 - iter 1260/1809 - loss 0.00427865 - time (sec): 78.42 - samples/sec: 3377.11 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-14 23:12:40,344 epoch 10 - iter 1440/1809 - loss 0.00393300 - time (sec): 89.61 - samples/sec: 3387.40 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-14 23:12:51,215 epoch 10 - iter 1620/1809 - loss 0.00442583 - time (sec): 100.48 - samples/sec: 3373.72 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-14 23:13:02,697 epoch 10 - iter 1800/1809 - loss 0.00442660 - time (sec): 111.96 - samples/sec: 3379.21 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-14 23:13:03,201 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-14 23:13:03,201 EPOCH 10 done: loss 0.0044 - lr: 0.000000
217
+ 2023-10-14 23:13:08,887 DEV : loss 0.39746981859207153 - f1-score (micro avg) 0.6532
218
+ 2023-10-14 23:13:08,933 saving best model
219
+ 2023-10-14 23:13:09,734 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-14 23:13:09,736 Loading model from best epoch ...
221
+ 2023-10-14 23:13:11,362 SequenceTagger predicts: Dictionary with 13 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
222
+ 2023-10-14 23:13:20,771
223
+ Results:
224
+ - F-score (micro) 0.6532
225
+ - F-score (macro) 0.5047
226
+ - Accuracy 0.5013
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ loc 0.6327 0.7986 0.7061 591
232
+ pers 0.5670 0.7703 0.6532 357
233
+ org 0.1746 0.1392 0.1549 79
234
+
235
+ micro avg 0.5858 0.7381 0.6532 1027
236
+ macro avg 0.4581 0.5694 0.5047 1027
237
+ weighted avg 0.5746 0.7381 0.6453 1027
238
+
239
+ 2023-10-14 23:13:20,771 ----------------------------------------------------------------------------------------------------