Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbyt5-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
2c1a4d8
•
1 Parent(s): 01cb2bf

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. final-model.pt +3 -0
  4. loss.tsv +11 -0
  5. runs/events.out.tfevents.1697336974.c8b2203b18a8.2923.18 +3 -0
  6. test.tsv +0 -0
  7. training.log +260 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07c9b340c5ffc17f9e5d37ba37f5c55ff2f81df63f91246a0b9ac3cc42ac7d43
3
+ size 870793839
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
final-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d24f1651fd23ed9b74e68bce903a1a031c2733e08d601bb8b480b73c6bfc57ef
3
+ size 870793956
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 02:47:22 0.0001 0.6153 0.1182 0.5371 0.7048 0.6096 0.4480
3
+ 2 03:05:21 0.0001 0.0894 0.1254 0.5407 0.8284 0.6543 0.4983
4
+ 3 03:23:35 0.0001 0.0635 0.1557 0.5748 0.6991 0.6309 0.4696
5
+ 4 03:41:24 0.0001 0.0442 0.2190 0.5597 0.7620 0.6453 0.4861
6
+ 5 03:59:29 0.0001 0.0326 0.2557 0.5604 0.7483 0.6409 0.4777
7
+ 6 04:17:14 0.0001 0.0232 0.2822 0.5468 0.7551 0.6343 0.4741
8
+ 7 04:35:23 0.0001 0.0155 0.3076 0.5633 0.7895 0.6575 0.4993
9
+ 8 04:53:30 0.0000 0.0100 0.3296 0.5677 0.7437 0.6439 0.4854
10
+ 9 05:11:57 0.0000 0.0059 0.3671 0.5620 0.7620 0.6469 0.4886
11
+ 10 05:29:54 0.0000 0.0039 0.3736 0.5640 0.7666 0.6499 0.4905
runs/events.out.tfevents.1697336974.c8b2203b18a8.2923.18 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4732b48ac4404c1e9770e5469d31d48a0af44f2fbc270a512a61aa3d700eca2
3
+ size 2030580
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,260 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 02:29:34,769 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 02:29:34,771 Model: "SequenceTagger(
3
+ (embeddings): ByT5Embeddings(
4
+ (model): T5EncoderModel(
5
+ (shared): Embedding(384, 1472)
6
+ (encoder): T5Stack(
7
+ (embed_tokens): Embedding(384, 1472)
8
+ (block): ModuleList(
9
+ (0): T5Block(
10
+ (layer): ModuleList(
11
+ (0): T5LayerSelfAttention(
12
+ (SelfAttention): T5Attention(
13
+ (q): Linear(in_features=1472, out_features=384, bias=False)
14
+ (k): Linear(in_features=1472, out_features=384, bias=False)
15
+ (v): Linear(in_features=1472, out_features=384, bias=False)
16
+ (o): Linear(in_features=384, out_features=1472, bias=False)
17
+ (relative_attention_bias): Embedding(32, 6)
18
+ )
19
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (1): T5LayerFF(
23
+ (DenseReluDense): T5DenseGatedActDense(
24
+ (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
25
+ (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
26
+ (wo): Linear(in_features=3584, out_features=1472, bias=False)
27
+ (dropout): Dropout(p=0.1, inplace=False)
28
+ (act): NewGELUActivation()
29
+ )
30
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
31
+ (dropout): Dropout(p=0.1, inplace=False)
32
+ )
33
+ )
34
+ )
35
+ (1-11): 11 x T5Block(
36
+ (layer): ModuleList(
37
+ (0): T5LayerSelfAttention(
38
+ (SelfAttention): T5Attention(
39
+ (q): Linear(in_features=1472, out_features=384, bias=False)
40
+ (k): Linear(in_features=1472, out_features=384, bias=False)
41
+ (v): Linear(in_features=1472, out_features=384, bias=False)
42
+ (o): Linear(in_features=384, out_features=1472, bias=False)
43
+ )
44
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
45
+ (dropout): Dropout(p=0.1, inplace=False)
46
+ )
47
+ (1): T5LayerFF(
48
+ (DenseReluDense): T5DenseGatedActDense(
49
+ (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
50
+ (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
51
+ (wo): Linear(in_features=3584, out_features=1472, bias=False)
52
+ (dropout): Dropout(p=0.1, inplace=False)
53
+ (act): NewGELUActivation()
54
+ )
55
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
56
+ (dropout): Dropout(p=0.1, inplace=False)
57
+ )
58
+ )
59
+ )
60
+ )
61
+ (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
62
+ (dropout): Dropout(p=0.1, inplace=False)
63
+ )
64
+ )
65
+ )
66
+ (locked_dropout): LockedDropout(p=0.5)
67
+ (linear): Linear(in_features=1472, out_features=13, bias=True)
68
+ (loss_function): CrossEntropyLoss()
69
+ )"
70
+ 2023-10-15 02:29:34,771 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-15 02:29:34,772 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
72
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
73
+ 2023-10-15 02:29:34,772 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-15 02:29:34,772 Train: 14465 sentences
75
+ 2023-10-15 02:29:34,772 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-15 02:29:34,772 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-15 02:29:34,772 Training Params:
78
+ 2023-10-15 02:29:34,772 - learning_rate: "0.00015"
79
+ 2023-10-15 02:29:34,772 - mini_batch_size: "4"
80
+ 2023-10-15 02:29:34,772 - max_epochs: "10"
81
+ 2023-10-15 02:29:34,772 - shuffle: "True"
82
+ 2023-10-15 02:29:34,772 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-15 02:29:34,772 Plugins:
84
+ 2023-10-15 02:29:34,773 - TensorboardLogger
85
+ 2023-10-15 02:29:34,773 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-15 02:29:34,773 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 02:29:34,773 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-15 02:29:34,773 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-15 02:29:34,773 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-15 02:29:34,773 Computation:
91
+ 2023-10-15 02:29:34,773 - compute on device: cuda:0
92
+ 2023-10-15 02:29:34,773 - embedding storage: none
93
+ 2023-10-15 02:29:34,773 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-15 02:29:34,773 Model training base path: "hmbench-letemps/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
95
+ 2023-10-15 02:29:34,773 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-15 02:29:34,773 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-15 02:29:34,774 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-15 02:31:16,649 epoch 1 - iter 361/3617 - loss 2.48429488 - time (sec): 101.87 - samples/sec: 363.76 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-15 02:33:00,526 epoch 1 - iter 722/3617 - loss 2.08511449 - time (sec): 205.75 - samples/sec: 367.56 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-15 02:34:44,602 epoch 1 - iter 1083/3617 - loss 1.63448789 - time (sec): 309.83 - samples/sec: 368.43 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-15 02:36:28,896 epoch 1 - iter 1444/3617 - loss 1.29679939 - time (sec): 414.12 - samples/sec: 368.06 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-15 02:38:12,537 epoch 1 - iter 1805/3617 - loss 1.07970081 - time (sec): 517.76 - samples/sec: 367.99 - lr: 0.000075 - momentum: 0.000000
103
+ 2023-10-15 02:39:51,982 epoch 1 - iter 2166/3617 - loss 0.93329345 - time (sec): 617.20 - samples/sec: 369.51 - lr: 0.000090 - momentum: 0.000000
104
+ 2023-10-15 02:41:34,705 epoch 1 - iter 2527/3617 - loss 0.82257414 - time (sec): 719.93 - samples/sec: 370.28 - lr: 0.000105 - momentum: 0.000000
105
+ 2023-10-15 02:43:15,591 epoch 1 - iter 2888/3617 - loss 0.74390282 - time (sec): 820.82 - samples/sec: 367.41 - lr: 0.000120 - momentum: 0.000000
106
+ 2023-10-15 02:44:58,546 epoch 1 - iter 3249/3617 - loss 0.67241668 - time (sec): 923.77 - samples/sec: 368.85 - lr: 0.000135 - momentum: 0.000000
107
+ 2023-10-15 02:46:41,614 epoch 1 - iter 3610/3617 - loss 0.61659329 - time (sec): 1026.84 - samples/sec: 369.06 - lr: 0.000150 - momentum: 0.000000
108
+ 2023-10-15 02:46:43,525 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-15 02:46:43,525 EPOCH 1 done: loss 0.6153 - lr: 0.000150
110
+ 2023-10-15 02:47:22,767 DEV : loss 0.1182025894522667 - f1-score (micro avg) 0.6096
111
+ 2023-10-15 02:47:22,825 saving best model
112
+ 2023-10-15 02:47:23,873 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-15 02:49:12,405 epoch 2 - iter 361/3617 - loss 0.10486429 - time (sec): 108.53 - samples/sec: 339.21 - lr: 0.000148 - momentum: 0.000000
114
+ 2023-10-15 02:50:57,235 epoch 2 - iter 722/3617 - loss 0.09736275 - time (sec): 213.36 - samples/sec: 347.69 - lr: 0.000147 - momentum: 0.000000
115
+ 2023-10-15 02:52:39,885 epoch 2 - iter 1083/3617 - loss 0.09580660 - time (sec): 316.01 - samples/sec: 353.51 - lr: 0.000145 - momentum: 0.000000
116
+ 2023-10-15 02:54:23,253 epoch 2 - iter 1444/3617 - loss 0.09592234 - time (sec): 419.38 - samples/sec: 358.09 - lr: 0.000143 - momentum: 0.000000
117
+ 2023-10-15 02:56:02,945 epoch 2 - iter 1805/3617 - loss 0.09622099 - time (sec): 519.07 - samples/sec: 359.97 - lr: 0.000142 - momentum: 0.000000
118
+ 2023-10-15 02:57:43,319 epoch 2 - iter 2166/3617 - loss 0.09617073 - time (sec): 619.44 - samples/sec: 361.51 - lr: 0.000140 - momentum: 0.000000
119
+ 2023-10-15 02:59:25,512 epoch 2 - iter 2527/3617 - loss 0.09429772 - time (sec): 721.64 - samples/sec: 362.05 - lr: 0.000138 - momentum: 0.000000
120
+ 2023-10-15 03:01:07,359 epoch 2 - iter 2888/3617 - loss 0.09188935 - time (sec): 823.48 - samples/sec: 365.16 - lr: 0.000137 - momentum: 0.000000
121
+ 2023-10-15 03:02:50,899 epoch 2 - iter 3249/3617 - loss 0.09018743 - time (sec): 927.02 - samples/sec: 366.46 - lr: 0.000135 - momentum: 0.000000
122
+ 2023-10-15 03:04:38,101 epoch 2 - iter 3610/3617 - loss 0.08943780 - time (sec): 1034.23 - samples/sec: 367.05 - lr: 0.000133 - momentum: 0.000000
123
+ 2023-10-15 03:04:39,630 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-15 03:04:39,631 EPOCH 2 done: loss 0.0894 - lr: 0.000133
125
+ 2023-10-15 03:05:21,648 DEV : loss 0.12542375922203064 - f1-score (micro avg) 0.6543
126
+ 2023-10-15 03:05:21,705 saving best model
127
+ 2023-10-15 03:05:26,943 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-15 03:07:15,684 epoch 3 - iter 361/3617 - loss 0.05585430 - time (sec): 108.74 - samples/sec: 351.94 - lr: 0.000132 - momentum: 0.000000
129
+ 2023-10-15 03:09:00,770 epoch 3 - iter 722/3617 - loss 0.05518947 - time (sec): 213.82 - samples/sec: 358.14 - lr: 0.000130 - momentum: 0.000000
130
+ 2023-10-15 03:10:44,972 epoch 3 - iter 1083/3617 - loss 0.05811195 - time (sec): 318.02 - samples/sec: 361.04 - lr: 0.000128 - momentum: 0.000000
131
+ 2023-10-15 03:12:32,005 epoch 3 - iter 1444/3617 - loss 0.06086072 - time (sec): 425.06 - samples/sec: 357.88 - lr: 0.000127 - momentum: 0.000000
132
+ 2023-10-15 03:14:15,820 epoch 3 - iter 1805/3617 - loss 0.06371809 - time (sec): 528.87 - samples/sec: 357.94 - lr: 0.000125 - momentum: 0.000000
133
+ 2023-10-15 03:16:03,404 epoch 3 - iter 2166/3617 - loss 0.06313824 - time (sec): 636.46 - samples/sec: 358.12 - lr: 0.000123 - momentum: 0.000000
134
+ 2023-10-15 03:17:51,966 epoch 3 - iter 2527/3617 - loss 0.06300600 - time (sec): 745.02 - samples/sec: 356.80 - lr: 0.000122 - momentum: 0.000000
135
+ 2023-10-15 03:19:35,189 epoch 3 - iter 2888/3617 - loss 0.06337488 - time (sec): 848.24 - samples/sec: 358.47 - lr: 0.000120 - momentum: 0.000000
136
+ 2023-10-15 03:21:14,694 epoch 3 - iter 3249/3617 - loss 0.06393407 - time (sec): 947.75 - samples/sec: 360.63 - lr: 0.000118 - momentum: 0.000000
137
+ 2023-10-15 03:22:53,465 epoch 3 - iter 3610/3617 - loss 0.06345144 - time (sec): 1046.52 - samples/sec: 362.59 - lr: 0.000117 - momentum: 0.000000
138
+ 2023-10-15 03:22:55,028 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-15 03:22:55,029 EPOCH 3 done: loss 0.0635 - lr: 0.000117
140
+ 2023-10-15 03:23:35,022 DEV : loss 0.15569539368152618 - f1-score (micro avg) 0.6309
141
+ 2023-10-15 03:23:35,087 ----------------------------------------------------------------------------------------------------
142
+ 2023-10-15 03:25:16,482 epoch 4 - iter 361/3617 - loss 0.04588584 - time (sec): 101.39 - samples/sec: 372.12 - lr: 0.000115 - momentum: 0.000000
143
+ 2023-10-15 03:27:03,414 epoch 4 - iter 722/3617 - loss 0.04406940 - time (sec): 208.32 - samples/sec: 359.05 - lr: 0.000113 - momentum: 0.000000
144
+ 2023-10-15 03:28:42,853 epoch 4 - iter 1083/3617 - loss 0.04196868 - time (sec): 307.76 - samples/sec: 362.60 - lr: 0.000112 - momentum: 0.000000
145
+ 2023-10-15 03:30:30,301 epoch 4 - iter 1444/3617 - loss 0.04194387 - time (sec): 415.21 - samples/sec: 362.93 - lr: 0.000110 - momentum: 0.000000
146
+ 2023-10-15 03:32:15,993 epoch 4 - iter 1805/3617 - loss 0.04269980 - time (sec): 520.90 - samples/sec: 360.89 - lr: 0.000108 - momentum: 0.000000
147
+ 2023-10-15 03:34:00,324 epoch 4 - iter 2166/3617 - loss 0.04366510 - time (sec): 625.23 - samples/sec: 362.28 - lr: 0.000107 - momentum: 0.000000
148
+ 2023-10-15 03:35:41,692 epoch 4 - iter 2527/3617 - loss 0.04278926 - time (sec): 726.60 - samples/sec: 363.24 - lr: 0.000105 - momentum: 0.000000
149
+ 2023-10-15 03:37:19,930 epoch 4 - iter 2888/3617 - loss 0.04289369 - time (sec): 824.84 - samples/sec: 366.97 - lr: 0.000103 - momentum: 0.000000
150
+ 2023-10-15 03:39:01,675 epoch 4 - iter 3249/3617 - loss 0.04430666 - time (sec): 926.59 - samples/sec: 367.47 - lr: 0.000102 - momentum: 0.000000
151
+ 2023-10-15 03:40:43,008 epoch 4 - iter 3610/3617 - loss 0.04419319 - time (sec): 1027.92 - samples/sec: 368.77 - lr: 0.000100 - momentum: 0.000000
152
+ 2023-10-15 03:40:45,014 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-15 03:40:45,014 EPOCH 4 done: loss 0.0442 - lr: 0.000100
154
+ 2023-10-15 03:41:24,744 DEV : loss 0.21898627281188965 - f1-score (micro avg) 0.6453
155
+ 2023-10-15 03:41:24,804 ----------------------------------------------------------------------------------------------------
156
+ 2023-10-15 03:43:08,747 epoch 5 - iter 361/3617 - loss 0.02795933 - time (sec): 103.94 - samples/sec: 359.71 - lr: 0.000098 - momentum: 0.000000
157
+ 2023-10-15 03:44:53,272 epoch 5 - iter 722/3617 - loss 0.02955686 - time (sec): 208.47 - samples/sec: 363.22 - lr: 0.000097 - momentum: 0.000000
158
+ 2023-10-15 03:46:34,182 epoch 5 - iter 1083/3617 - loss 0.02857138 - time (sec): 309.38 - samples/sec: 364.75 - lr: 0.000095 - momentum: 0.000000
159
+ 2023-10-15 03:48:18,240 epoch 5 - iter 1444/3617 - loss 0.02890810 - time (sec): 413.43 - samples/sec: 364.62 - lr: 0.000093 - momentum: 0.000000
160
+ 2023-10-15 03:50:03,634 epoch 5 - iter 1805/3617 - loss 0.03096541 - time (sec): 518.83 - samples/sec: 363.78 - lr: 0.000092 - momentum: 0.000000
161
+ 2023-10-15 03:51:46,202 epoch 5 - iter 2166/3617 - loss 0.03145099 - time (sec): 621.40 - samples/sec: 366.85 - lr: 0.000090 - momentum: 0.000000
162
+ 2023-10-15 03:53:29,190 epoch 5 - iter 2527/3617 - loss 0.03190957 - time (sec): 724.38 - samples/sec: 366.09 - lr: 0.000088 - momentum: 0.000000
163
+ 2023-10-15 03:55:11,030 epoch 5 - iter 2888/3617 - loss 0.03178960 - time (sec): 826.22 - samples/sec: 367.04 - lr: 0.000087 - momentum: 0.000000
164
+ 2023-10-15 03:57:01,899 epoch 5 - iter 3249/3617 - loss 0.03207875 - time (sec): 937.09 - samples/sec: 364.93 - lr: 0.000085 - momentum: 0.000000
165
+ 2023-10-15 03:58:46,580 epoch 5 - iter 3610/3617 - loss 0.03263643 - time (sec): 1041.77 - samples/sec: 363.87 - lr: 0.000083 - momentum: 0.000000
166
+ 2023-10-15 03:58:48,517 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-15 03:58:48,517 EPOCH 5 done: loss 0.0326 - lr: 0.000083
168
+ 2023-10-15 03:59:29,095 DEV : loss 0.2557448446750641 - f1-score (micro avg) 0.6409
169
+ 2023-10-15 03:59:29,163 ----------------------------------------------------------------------------------------------------
170
+ 2023-10-15 04:01:10,615 epoch 6 - iter 361/3617 - loss 0.01796897 - time (sec): 101.45 - samples/sec: 378.63 - lr: 0.000082 - momentum: 0.000000
171
+ 2023-10-15 04:02:47,179 epoch 6 - iter 722/3617 - loss 0.01852486 - time (sec): 198.01 - samples/sec: 387.08 - lr: 0.000080 - momentum: 0.000000
172
+ 2023-10-15 04:04:27,156 epoch 6 - iter 1083/3617 - loss 0.02027581 - time (sec): 297.99 - samples/sec: 385.15 - lr: 0.000078 - momentum: 0.000000
173
+ 2023-10-15 04:06:08,772 epoch 6 - iter 1444/3617 - loss 0.01924828 - time (sec): 399.61 - samples/sec: 382.06 - lr: 0.000077 - momentum: 0.000000
174
+ 2023-10-15 04:07:55,921 epoch 6 - iter 1805/3617 - loss 0.02014624 - time (sec): 506.76 - samples/sec: 379.85 - lr: 0.000075 - momentum: 0.000000
175
+ 2023-10-15 04:09:42,057 epoch 6 - iter 2166/3617 - loss 0.01969253 - time (sec): 612.89 - samples/sec: 376.35 - lr: 0.000073 - momentum: 0.000000
176
+ 2023-10-15 04:11:27,616 epoch 6 - iter 2527/3617 - loss 0.02096559 - time (sec): 718.45 - samples/sec: 372.56 - lr: 0.000072 - momentum: 0.000000
177
+ 2023-10-15 04:13:08,698 epoch 6 - iter 2888/3617 - loss 0.02194795 - time (sec): 819.53 - samples/sec: 371.88 - lr: 0.000070 - momentum: 0.000000
178
+ 2023-10-15 04:14:50,495 epoch 6 - iter 3249/3617 - loss 0.02256127 - time (sec): 921.33 - samples/sec: 371.13 - lr: 0.000068 - momentum: 0.000000
179
+ 2023-10-15 04:16:31,345 epoch 6 - iter 3610/3617 - loss 0.02319045 - time (sec): 1022.18 - samples/sec: 371.20 - lr: 0.000067 - momentum: 0.000000
180
+ 2023-10-15 04:16:32,970 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-15 04:16:32,971 EPOCH 6 done: loss 0.0232 - lr: 0.000067
182
+ 2023-10-15 04:17:14,014 DEV : loss 0.282193660736084 - f1-score (micro avg) 0.6343
183
+ 2023-10-15 04:17:14,078 ----------------------------------------------------------------------------------------------------
184
+ 2023-10-15 04:18:55,799 epoch 7 - iter 361/3617 - loss 0.01551071 - time (sec): 101.72 - samples/sec: 373.52 - lr: 0.000065 - momentum: 0.000000
185
+ 2023-10-15 04:20:41,886 epoch 7 - iter 722/3617 - loss 0.01373232 - time (sec): 207.80 - samples/sec: 362.46 - lr: 0.000063 - momentum: 0.000000
186
+ 2023-10-15 04:22:29,691 epoch 7 - iter 1083/3617 - loss 0.01328330 - time (sec): 315.61 - samples/sec: 360.80 - lr: 0.000062 - momentum: 0.000000
187
+ 2023-10-15 04:24:08,799 epoch 7 - iter 1444/3617 - loss 0.01244965 - time (sec): 414.72 - samples/sec: 364.12 - lr: 0.000060 - momentum: 0.000000
188
+ 2023-10-15 04:25:55,152 epoch 7 - iter 1805/3617 - loss 0.01196745 - time (sec): 521.07 - samples/sec: 361.24 - lr: 0.000058 - momentum: 0.000000
189
+ 2023-10-15 04:27:41,799 epoch 7 - iter 2166/3617 - loss 0.01224501 - time (sec): 627.72 - samples/sec: 360.62 - lr: 0.000057 - momentum: 0.000000
190
+ 2023-10-15 04:29:29,080 epoch 7 - iter 2527/3617 - loss 0.01298380 - time (sec): 735.00 - samples/sec: 358.99 - lr: 0.000055 - momentum: 0.000000
191
+ 2023-10-15 04:31:13,831 epoch 7 - iter 2888/3617 - loss 0.01469824 - time (sec): 839.75 - samples/sec: 360.58 - lr: 0.000053 - momentum: 0.000000
192
+ 2023-10-15 04:32:56,602 epoch 7 - iter 3249/3617 - loss 0.01489068 - time (sec): 942.52 - samples/sec: 362.79 - lr: 0.000052 - momentum: 0.000000
193
+ 2023-10-15 04:34:40,305 epoch 7 - iter 3610/3617 - loss 0.01541473 - time (sec): 1046.22 - samples/sec: 362.20 - lr: 0.000050 - momentum: 0.000000
194
+ 2023-10-15 04:34:42,369 ----------------------------------------------------------------------------------------------------
195
+ 2023-10-15 04:34:42,369 EPOCH 7 done: loss 0.0155 - lr: 0.000050
196
+ 2023-10-15 04:35:22,972 DEV : loss 0.30758461356163025 - f1-score (micro avg) 0.6575
197
+ 2023-10-15 04:35:23,043 saving best model
198
+ 2023-10-15 04:35:26,830 ----------------------------------------------------------------------------------------------------
199
+ 2023-10-15 04:37:08,130 epoch 8 - iter 361/3617 - loss 0.00581311 - time (sec): 101.29 - samples/sec: 370.16 - lr: 0.000048 - momentum: 0.000000
200
+ 2023-10-15 04:38:50,789 epoch 8 - iter 722/3617 - loss 0.00742338 - time (sec): 203.95 - samples/sec: 368.68 - lr: 0.000047 - momentum: 0.000000
201
+ 2023-10-15 04:40:33,780 epoch 8 - iter 1083/3617 - loss 0.00920212 - time (sec): 306.94 - samples/sec: 367.84 - lr: 0.000045 - momentum: 0.000000
202
+ 2023-10-15 04:42:18,477 epoch 8 - iter 1444/3617 - loss 0.00960634 - time (sec): 411.63 - samples/sec: 367.76 - lr: 0.000043 - momentum: 0.000000
203
+ 2023-10-15 04:43:59,290 epoch 8 - iter 1805/3617 - loss 0.00956098 - time (sec): 512.45 - samples/sec: 368.89 - lr: 0.000042 - momentum: 0.000000
204
+ 2023-10-15 04:45:43,621 epoch 8 - iter 2166/3617 - loss 0.00946192 - time (sec): 616.78 - samples/sec: 368.64 - lr: 0.000040 - momentum: 0.000000
205
+ 2023-10-15 04:47:29,515 epoch 8 - iter 2527/3617 - loss 0.00973484 - time (sec): 722.67 - samples/sec: 369.13 - lr: 0.000038 - momentum: 0.000000
206
+ 2023-10-15 04:49:15,633 epoch 8 - iter 2888/3617 - loss 0.01000900 - time (sec): 828.79 - samples/sec: 368.05 - lr: 0.000037 - momentum: 0.000000
207
+ 2023-10-15 04:51:01,089 epoch 8 - iter 3249/3617 - loss 0.01011415 - time (sec): 934.25 - samples/sec: 366.34 - lr: 0.000035 - momentum: 0.000000
208
+ 2023-10-15 04:52:44,746 epoch 8 - iter 3610/3617 - loss 0.00982368 - time (sec): 1037.90 - samples/sec: 365.46 - lr: 0.000033 - momentum: 0.000000
209
+ 2023-10-15 04:52:46,498 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-15 04:52:46,498 EPOCH 8 done: loss 0.0100 - lr: 0.000033
211
+ 2023-10-15 04:53:30,565 DEV : loss 0.32959964871406555 - f1-score (micro avg) 0.6439
212
+ 2023-10-15 04:53:30,641 ----------------------------------------------------------------------------------------------------
213
+ 2023-10-15 04:55:18,612 epoch 9 - iter 361/3617 - loss 0.00594922 - time (sec): 107.97 - samples/sec: 353.74 - lr: 0.000032 - momentum: 0.000000
214
+ 2023-10-15 04:57:12,107 epoch 9 - iter 722/3617 - loss 0.00497982 - time (sec): 221.46 - samples/sec: 349.12 - lr: 0.000030 - momentum: 0.000000
215
+ 2023-10-15 04:59:04,948 epoch 9 - iter 1083/3617 - loss 0.00520872 - time (sec): 334.30 - samples/sec: 347.55 - lr: 0.000028 - momentum: 0.000000
216
+ 2023-10-15 05:00:53,029 epoch 9 - iter 1444/3617 - loss 0.00532160 - time (sec): 442.38 - samples/sec: 346.40 - lr: 0.000027 - momentum: 0.000000
217
+ 2023-10-15 05:02:42,252 epoch 9 - iter 1805/3617 - loss 0.00564625 - time (sec): 551.61 - samples/sec: 347.07 - lr: 0.000025 - momentum: 0.000000
218
+ 2023-10-15 05:04:21,445 epoch 9 - iter 2166/3617 - loss 0.00557872 - time (sec): 650.80 - samples/sec: 349.48 - lr: 0.000023 - momentum: 0.000000
219
+ 2023-10-15 05:06:04,807 epoch 9 - iter 2527/3617 - loss 0.00576891 - time (sec): 754.16 - samples/sec: 351.99 - lr: 0.000022 - momentum: 0.000000
220
+ 2023-10-15 05:07:46,313 epoch 9 - iter 2888/3617 - loss 0.00564754 - time (sec): 855.67 - samples/sec: 355.37 - lr: 0.000020 - momentum: 0.000000
221
+ 2023-10-15 05:09:29,508 epoch 9 - iter 3249/3617 - loss 0.00556824 - time (sec): 958.86 - samples/sec: 356.62 - lr: 0.000018 - momentum: 0.000000
222
+ 2023-10-15 05:11:13,362 epoch 9 - iter 3610/3617 - loss 0.00589214 - time (sec): 1062.72 - samples/sec: 356.87 - lr: 0.000017 - momentum: 0.000000
223
+ 2023-10-15 05:11:15,161 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-15 05:11:15,161 EPOCH 9 done: loss 0.0059 - lr: 0.000017
225
+ 2023-10-15 05:11:57,377 DEV : loss 0.36710360646247864 - f1-score (micro avg) 0.6469
226
+ 2023-10-15 05:11:57,450 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-15 05:13:41,148 epoch 10 - iter 361/3617 - loss 0.00459603 - time (sec): 103.70 - samples/sec: 363.85 - lr: 0.000015 - momentum: 0.000000
228
+ 2023-10-15 05:15:25,469 epoch 10 - iter 722/3617 - loss 0.00476060 - time (sec): 208.02 - samples/sec: 357.09 - lr: 0.000013 - momentum: 0.000000
229
+ 2023-10-15 05:17:06,915 epoch 10 - iter 1083/3617 - loss 0.00526415 - time (sec): 309.46 - samples/sec: 359.37 - lr: 0.000012 - momentum: 0.000000
230
+ 2023-10-15 05:18:48,330 epoch 10 - iter 1444/3617 - loss 0.00527575 - time (sec): 410.88 - samples/sec: 364.44 - lr: 0.000010 - momentum: 0.000000
231
+ 2023-10-15 05:20:28,943 epoch 10 - iter 1805/3617 - loss 0.00485414 - time (sec): 511.49 - samples/sec: 367.86 - lr: 0.000008 - momentum: 0.000000
232
+ 2023-10-15 05:22:14,000 epoch 10 - iter 2166/3617 - loss 0.00456797 - time (sec): 616.55 - samples/sec: 367.53 - lr: 0.000007 - momentum: 0.000000
233
+ 2023-10-15 05:23:57,638 epoch 10 - iter 2527/3617 - loss 0.00426810 - time (sec): 720.19 - samples/sec: 366.97 - lr: 0.000005 - momentum: 0.000000
234
+ 2023-10-15 05:25:40,310 epoch 10 - iter 2888/3617 - loss 0.00422168 - time (sec): 822.86 - samples/sec: 367.59 - lr: 0.000003 - momentum: 0.000000
235
+ 2023-10-15 05:27:24,538 epoch 10 - iter 3249/3617 - loss 0.00407721 - time (sec): 927.09 - samples/sec: 367.90 - lr: 0.000002 - momentum: 0.000000
236
+ 2023-10-15 05:29:09,771 epoch 10 - iter 3610/3617 - loss 0.00392103 - time (sec): 1032.32 - samples/sec: 367.29 - lr: 0.000000 - momentum: 0.000000
237
+ 2023-10-15 05:29:11,687 ----------------------------------------------------------------------------------------------------
238
+ 2023-10-15 05:29:11,687 EPOCH 10 done: loss 0.0039 - lr: 0.000000
239
+ 2023-10-15 05:29:54,236 DEV : loss 0.3736409842967987 - f1-score (micro avg) 0.6499
240
+ 2023-10-15 05:29:55,360 ----------------------------------------------------------------------------------------------------
241
+ 2023-10-15 05:29:55,362 Loading model from best epoch ...
242
+ 2023-10-15 05:30:00,980 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
243
+ 2023-10-15 05:31:03,256
244
+ Results:
245
+ - F-score (micro) 0.6337
246
+ - F-score (macro) 0.475
247
+ - Accuracy 0.4739
248
+
249
+ By class:
250
+ precision recall f1-score support
251
+
252
+ loc 0.6440 0.7868 0.7083 591
253
+ pers 0.5520 0.6835 0.6108 357
254
+ org 0.1111 0.1013 0.1060 79
255
+
256
+ micro avg 0.5801 0.6981 0.6337 1027
257
+ macro avg 0.4357 0.5238 0.4750 1027
258
+ weighted avg 0.5711 0.6981 0.6281 1027
259
+
260
+ 2023-10-15 05:31:03,257 ----------------------------------------------------------------------------------------------------