Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbyt5-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
13abe5f
•
1 Parent(s): 4f10a7b

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.1696989627.de2e83fddbee.1120.4 +3 -0
  6. test.tsv +0 -0
  7. training.log +263 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cba52a9cc2ce7de52d908c9f433ca330b2838de145fae61a15b57ddd3a4bf31
3
+ size 870817519
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 sha256:76fbf068af5a192706eb18d9341d59b1a22128462088ff05eabd285c6d1cfdb8
3
+ size 870817636
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 02:09:20 0.0001 1.3752 0.2779 0.3301 0.1837 0.2360 0.1432
3
+ 2 02:17:51 0.0001 0.1931 0.1034 0.7150 0.7578 0.7358 0.6015
4
+ 3 02:26:37 0.0001 0.0799 0.1001 0.7678 0.7918 0.7796 0.6539
5
+ 4 02:35:24 0.0001 0.0507 0.1174 0.7781 0.8014 0.7895 0.6693
6
+ 5 02:44:09 0.0001 0.0383 0.1292 0.7695 0.8041 0.7864 0.6626
7
+ 6 02:53:31 0.0001 0.0289 0.1372 0.7704 0.8082 0.7888 0.6682
8
+ 7 03:02:27 0.0001 0.0224 0.1543 0.7646 0.8041 0.7838 0.6611
9
+ 8 03:11:44 0.0000 0.0176 0.1711 0.7861 0.8150 0.8003 0.6838
10
+ 9 03:20:59 0.0000 0.0147 0.1781 0.7870 0.8095 0.7981 0.6816
11
+ 10 03:29:51 0.0000 0.0121 0.1803 0.7768 0.8095 0.7928 0.6738
runs/events.out.tfevents.1696989627.de2e83fddbee.1120.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59ab1cb9905ac22e1c2f2a2af405efff5e3ca3a01116d3c1463fa73f61c80b8c
3
+ size 502461
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,263 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-11 02:00:27,642 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-11 02:00:27,645 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=17, bias=True)
68
+ (loss_function): CrossEntropyLoss()
69
+ )"
70
+ 2023-10-11 02:00:27,645 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-11 02:00:27,645 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
72
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
73
+ 2023-10-11 02:00:27,645 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-11 02:00:27,645 Train: 7142 sentences
75
+ 2023-10-11 02:00:27,645 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-11 02:00:27,646 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-11 02:00:27,646 Training Params:
78
+ 2023-10-11 02:00:27,646 - learning_rate: "0.00015"
79
+ 2023-10-11 02:00:27,646 - mini_batch_size: "8"
80
+ 2023-10-11 02:00:27,646 - max_epochs: "10"
81
+ 2023-10-11 02:00:27,646 - shuffle: "True"
82
+ 2023-10-11 02:00:27,646 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-11 02:00:27,646 Plugins:
84
+ 2023-10-11 02:00:27,646 - TensorboardLogger
85
+ 2023-10-11 02:00:27,646 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-11 02:00:27,646 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-11 02:00:27,646 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-11 02:00:27,646 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-11 02:00:27,646 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-11 02:00:27,646 Computation:
91
+ 2023-10-11 02:00:27,646 - compute on device: cuda:0
92
+ 2023-10-11 02:00:27,647 - embedding storage: none
93
+ 2023-10-11 02:00:27,647 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-11 02:00:27,647 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
95
+ 2023-10-11 02:00:27,647 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-11 02:00:27,647 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-11 02:00:27,647 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-11 02:01:19,984 epoch 1 - iter 89/893 - loss 2.83790708 - time (sec): 52.33 - samples/sec: 507.40 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-11 02:02:11,747 epoch 1 - iter 178/893 - loss 2.77852176 - time (sec): 104.10 - samples/sec: 497.39 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-11 02:03:05,897 epoch 1 - iter 267/893 - loss 2.58685480 - time (sec): 158.25 - samples/sec: 490.84 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-11 02:03:55,271 epoch 1 - iter 356/893 - loss 2.36244979 - time (sec): 207.62 - samples/sec: 495.84 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-11 02:04:46,699 epoch 1 - iter 445/893 - loss 2.13959943 - time (sec): 259.05 - samples/sec: 491.26 - lr: 0.000075 - momentum: 0.000000
103
+ 2023-10-11 02:05:38,352 epoch 1 - iter 534/893 - loss 1.91736663 - time (sec): 310.70 - samples/sec: 492.56 - lr: 0.000090 - momentum: 0.000000
104
+ 2023-10-11 02:06:29,081 epoch 1 - iter 623/893 - loss 1.74080020 - time (sec): 361.43 - samples/sec: 492.25 - lr: 0.000104 - momentum: 0.000000
105
+ 2023-10-11 02:07:23,712 epoch 1 - iter 712/893 - loss 1.60422960 - time (sec): 416.06 - samples/sec: 483.34 - lr: 0.000119 - momentum: 0.000000
106
+ 2023-10-11 02:08:12,004 epoch 1 - iter 801/893 - loss 1.48109419 - time (sec): 464.36 - samples/sec: 483.49 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-11 02:08:59,733 epoch 1 - iter 890/893 - loss 1.37704593 - time (sec): 512.08 - samples/sec: 484.80 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-11 02:09:01,007 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-11 02:09:01,008 EPOCH 1 done: loss 1.3752 - lr: 0.000149
110
+ 2023-10-11 02:09:20,678 DEV : loss 0.27791526913642883 - f1-score (micro avg) 0.236
111
+ 2023-10-11 02:09:20,711 saving best model
112
+ 2023-10-11 02:09:21,542 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-11 02:10:10,289 epoch 2 - iter 89/893 - loss 0.31498471 - time (sec): 48.74 - samples/sec: 515.58 - lr: 0.000148 - momentum: 0.000000
114
+ 2023-10-11 02:11:00,221 epoch 2 - iter 178/893 - loss 0.29516314 - time (sec): 98.68 - samples/sec: 522.13 - lr: 0.000147 - momentum: 0.000000
115
+ 2023-10-11 02:11:48,501 epoch 2 - iter 267/893 - loss 0.27648439 - time (sec): 146.96 - samples/sec: 519.52 - lr: 0.000145 - momentum: 0.000000
116
+ 2023-10-11 02:12:36,435 epoch 2 - iter 356/893 - loss 0.25974186 - time (sec): 194.89 - samples/sec: 515.74 - lr: 0.000143 - momentum: 0.000000
117
+ 2023-10-11 02:13:23,642 epoch 2 - iter 445/893 - loss 0.24526095 - time (sec): 242.10 - samples/sec: 514.55 - lr: 0.000142 - momentum: 0.000000
118
+ 2023-10-11 02:14:11,342 epoch 2 - iter 534/893 - loss 0.23418021 - time (sec): 289.80 - samples/sec: 512.27 - lr: 0.000140 - momentum: 0.000000
119
+ 2023-10-11 02:14:58,173 epoch 2 - iter 623/893 - loss 0.22272322 - time (sec): 336.63 - samples/sec: 511.28 - lr: 0.000138 - momentum: 0.000000
120
+ 2023-10-11 02:15:47,390 epoch 2 - iter 712/893 - loss 0.21186423 - time (sec): 385.84 - samples/sec: 511.40 - lr: 0.000137 - momentum: 0.000000
121
+ 2023-10-11 02:16:37,322 epoch 2 - iter 801/893 - loss 0.20247638 - time (sec): 435.78 - samples/sec: 512.65 - lr: 0.000135 - momentum: 0.000000
122
+ 2023-10-11 02:17:28,585 epoch 2 - iter 890/893 - loss 0.19324619 - time (sec): 487.04 - samples/sec: 509.21 - lr: 0.000133 - momentum: 0.000000
123
+ 2023-10-11 02:17:30,107 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-11 02:17:30,107 EPOCH 2 done: loss 0.1931 - lr: 0.000133
125
+ 2023-10-11 02:17:51,205 DEV : loss 0.10340522974729538 - f1-score (micro avg) 0.7358
126
+ 2023-10-11 02:17:51,239 saving best model
127
+ 2023-10-11 02:17:53,750 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-11 02:18:44,287 epoch 3 - iter 89/893 - loss 0.09071773 - time (sec): 50.53 - samples/sec: 511.46 - lr: 0.000132 - momentum: 0.000000
129
+ 2023-10-11 02:19:34,150 epoch 3 - iter 178/893 - loss 0.08766364 - time (sec): 100.39 - samples/sec: 484.80 - lr: 0.000130 - momentum: 0.000000
130
+ 2023-10-11 02:20:23,940 epoch 3 - iter 267/893 - loss 0.08620654 - time (sec): 150.18 - samples/sec: 492.81 - lr: 0.000128 - momentum: 0.000000
131
+ 2023-10-11 02:21:12,569 epoch 3 - iter 356/893 - loss 0.08710514 - time (sec): 198.81 - samples/sec: 494.07 - lr: 0.000127 - momentum: 0.000000
132
+ 2023-10-11 02:22:00,699 epoch 3 - iter 445/893 - loss 0.08228171 - time (sec): 246.94 - samples/sec: 497.77 - lr: 0.000125 - momentum: 0.000000
133
+ 2023-10-11 02:22:49,313 epoch 3 - iter 534/893 - loss 0.08245604 - time (sec): 295.56 - samples/sec: 501.25 - lr: 0.000123 - momentum: 0.000000
134
+ 2023-10-11 02:23:41,764 epoch 3 - iter 623/893 - loss 0.08060727 - time (sec): 348.01 - samples/sec: 502.27 - lr: 0.000122 - momentum: 0.000000
135
+ 2023-10-11 02:24:33,176 epoch 3 - iter 712/893 - loss 0.08031657 - time (sec): 399.42 - samples/sec: 499.87 - lr: 0.000120 - momentum: 0.000000
136
+ 2023-10-11 02:25:23,560 epoch 3 - iter 801/893 - loss 0.07999967 - time (sec): 449.80 - samples/sec: 497.60 - lr: 0.000118 - momentum: 0.000000
137
+ 2023-10-11 02:26:13,606 epoch 3 - iter 890/893 - loss 0.07986772 - time (sec): 499.85 - samples/sec: 496.50 - lr: 0.000117 - momentum: 0.000000
138
+ 2023-10-11 02:26:15,048 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-11 02:26:15,048 EPOCH 3 done: loss 0.0799 - lr: 0.000117
140
+ 2023-10-11 02:26:37,235 DEV : loss 0.1001172587275505 - f1-score (micro avg) 0.7796
141
+ 2023-10-11 02:26:37,265 saving best model
142
+ 2023-10-11 02:26:39,818 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-11 02:27:30,471 epoch 4 - iter 89/893 - loss 0.05380223 - time (sec): 50.65 - samples/sec: 488.72 - lr: 0.000115 - momentum: 0.000000
144
+ 2023-10-11 02:28:20,599 epoch 4 - iter 178/893 - loss 0.04978763 - time (sec): 100.78 - samples/sec: 474.69 - lr: 0.000113 - momentum: 0.000000
145
+ 2023-10-11 02:29:12,869 epoch 4 - iter 267/893 - loss 0.05029604 - time (sec): 153.05 - samples/sec: 482.96 - lr: 0.000112 - momentum: 0.000000
146
+ 2023-10-11 02:30:04,806 epoch 4 - iter 356/893 - loss 0.05128141 - time (sec): 204.98 - samples/sec: 490.90 - lr: 0.000110 - momentum: 0.000000
147
+ 2023-10-11 02:30:53,688 epoch 4 - iter 445/893 - loss 0.05039331 - time (sec): 253.87 - samples/sec: 487.56 - lr: 0.000108 - momentum: 0.000000
148
+ 2023-10-11 02:31:42,720 epoch 4 - iter 534/893 - loss 0.05100630 - time (sec): 302.90 - samples/sec: 490.66 - lr: 0.000107 - momentum: 0.000000
149
+ 2023-10-11 02:32:32,575 epoch 4 - iter 623/893 - loss 0.05103879 - time (sec): 352.75 - samples/sec: 494.73 - lr: 0.000105 - momentum: 0.000000
150
+ 2023-10-11 02:33:22,895 epoch 4 - iter 712/893 - loss 0.05096798 - time (sec): 403.07 - samples/sec: 493.14 - lr: 0.000103 - momentum: 0.000000
151
+ 2023-10-11 02:34:12,828 epoch 4 - iter 801/893 - loss 0.05049788 - time (sec): 453.01 - samples/sec: 492.51 - lr: 0.000102 - momentum: 0.000000
152
+ 2023-10-11 02:35:02,250 epoch 4 - iter 890/893 - loss 0.05080113 - time (sec): 502.43 - samples/sec: 494.14 - lr: 0.000100 - momentum: 0.000000
153
+ 2023-10-11 02:35:03,636 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-11 02:35:03,636 EPOCH 4 done: loss 0.0507 - lr: 0.000100
155
+ 2023-10-11 02:35:24,945 DEV : loss 0.117433100938797 - f1-score (micro avg) 0.7895
156
+ 2023-10-11 02:35:24,976 saving best model
157
+ 2023-10-11 02:35:27,592 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-11 02:36:18,704 epoch 5 - iter 89/893 - loss 0.04326464 - time (sec): 51.11 - samples/sec: 486.91 - lr: 0.000098 - momentum: 0.000000
159
+ 2023-10-11 02:37:07,649 epoch 5 - iter 178/893 - loss 0.04043555 - time (sec): 100.05 - samples/sec: 475.08 - lr: 0.000097 - momentum: 0.000000
160
+ 2023-10-11 02:37:58,633 epoch 5 - iter 267/893 - loss 0.04063361 - time (sec): 151.04 - samples/sec: 479.21 - lr: 0.000095 - momentum: 0.000000
161
+ 2023-10-11 02:38:47,397 epoch 5 - iter 356/893 - loss 0.04065561 - time (sec): 199.80 - samples/sec: 487.14 - lr: 0.000093 - momentum: 0.000000
162
+ 2023-10-11 02:39:35,821 epoch 5 - iter 445/893 - loss 0.04107866 - time (sec): 248.22 - samples/sec: 488.55 - lr: 0.000092 - momentum: 0.000000
163
+ 2023-10-11 02:40:25,648 epoch 5 - iter 534/893 - loss 0.03902400 - time (sec): 298.05 - samples/sec: 490.74 - lr: 0.000090 - momentum: 0.000000
164
+ 2023-10-11 02:41:16,041 epoch 5 - iter 623/893 - loss 0.03913140 - time (sec): 348.45 - samples/sec: 495.76 - lr: 0.000088 - momentum: 0.000000
165
+ 2023-10-11 02:42:05,861 epoch 5 - iter 712/893 - loss 0.03940436 - time (sec): 398.26 - samples/sec: 496.17 - lr: 0.000087 - momentum: 0.000000
166
+ 2023-10-11 02:42:54,853 epoch 5 - iter 801/893 - loss 0.03813737 - time (sec): 447.26 - samples/sec: 496.93 - lr: 0.000085 - momentum: 0.000000
167
+ 2023-10-11 02:43:45,529 epoch 5 - iter 890/893 - loss 0.03835348 - time (sec): 497.93 - samples/sec: 498.20 - lr: 0.000083 - momentum: 0.000000
168
+ 2023-10-11 02:43:47,040 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-11 02:43:47,041 EPOCH 5 done: loss 0.0383 - lr: 0.000083
170
+ 2023-10-11 02:44:09,474 DEV : loss 0.12915697693824768 - f1-score (micro avg) 0.7864
171
+ 2023-10-11 02:44:09,504 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-11 02:45:03,744 epoch 6 - iter 89/893 - loss 0.02671786 - time (sec): 54.24 - samples/sec: 456.47 - lr: 0.000082 - momentum: 0.000000
173
+ 2023-10-11 02:45:56,595 epoch 6 - iter 178/893 - loss 0.02772095 - time (sec): 107.09 - samples/sec: 465.30 - lr: 0.000080 - momentum: 0.000000
174
+ 2023-10-11 02:46:48,856 epoch 6 - iter 267/893 - loss 0.02637075 - time (sec): 159.35 - samples/sec: 467.27 - lr: 0.000078 - momentum: 0.000000
175
+ 2023-10-11 02:47:41,548 epoch 6 - iter 356/893 - loss 0.02780804 - time (sec): 212.04 - samples/sec: 467.98 - lr: 0.000077 - momentum: 0.000000
176
+ 2023-10-11 02:48:35,284 epoch 6 - iter 445/893 - loss 0.02807407 - time (sec): 265.78 - samples/sec: 467.16 - lr: 0.000075 - momentum: 0.000000
177
+ 2023-10-11 02:49:31,168 epoch 6 - iter 534/893 - loss 0.02808160 - time (sec): 321.66 - samples/sec: 466.96 - lr: 0.000073 - momentum: 0.000000
178
+ 2023-10-11 02:50:23,488 epoch 6 - iter 623/893 - loss 0.02768363 - time (sec): 373.98 - samples/sec: 466.32 - lr: 0.000072 - momentum: 0.000000
179
+ 2023-10-11 02:51:17,788 epoch 6 - iter 712/893 - loss 0.02797056 - time (sec): 428.28 - samples/sec: 465.85 - lr: 0.000070 - momentum: 0.000000
180
+ 2023-10-11 02:52:12,329 epoch 6 - iter 801/893 - loss 0.02832246 - time (sec): 482.82 - samples/sec: 466.61 - lr: 0.000068 - momentum: 0.000000
181
+ 2023-10-11 02:53:07,807 epoch 6 - iter 890/893 - loss 0.02898230 - time (sec): 538.30 - samples/sec: 461.25 - lr: 0.000067 - momentum: 0.000000
182
+ 2023-10-11 02:53:09,228 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-11 02:53:09,229 EPOCH 6 done: loss 0.0289 - lr: 0.000067
184
+ 2023-10-11 02:53:31,359 DEV : loss 0.13720029592514038 - f1-score (micro avg) 0.7888
185
+ 2023-10-11 02:53:31,390 ----------------------------------------------------------------------------------------------------
186
+ 2023-10-11 02:54:24,860 epoch 7 - iter 89/893 - loss 0.02389918 - time (sec): 53.47 - samples/sec: 511.70 - lr: 0.000065 - momentum: 0.000000
187
+ 2023-10-11 02:55:17,199 epoch 7 - iter 178/893 - loss 0.02442693 - time (sec): 105.81 - samples/sec: 491.45 - lr: 0.000063 - momentum: 0.000000
188
+ 2023-10-11 02:56:07,039 epoch 7 - iter 267/893 - loss 0.02393469 - time (sec): 155.65 - samples/sec: 486.40 - lr: 0.000062 - momentum: 0.000000
189
+ 2023-10-11 02:56:57,236 epoch 7 - iter 356/893 - loss 0.02364784 - time (sec): 205.84 - samples/sec: 485.70 - lr: 0.000060 - momentum: 0.000000
190
+ 2023-10-11 02:57:47,574 epoch 7 - iter 445/893 - loss 0.02319517 - time (sec): 256.18 - samples/sec: 485.39 - lr: 0.000058 - momentum: 0.000000
191
+ 2023-10-11 02:58:39,926 epoch 7 - iter 534/893 - loss 0.02254515 - time (sec): 308.53 - samples/sec: 485.71 - lr: 0.000057 - momentum: 0.000000
192
+ 2023-10-11 02:59:31,138 epoch 7 - iter 623/893 - loss 0.02241681 - time (sec): 359.75 - samples/sec: 482.79 - lr: 0.000055 - momentum: 0.000000
193
+ 2023-10-11 03:00:26,161 epoch 7 - iter 712/893 - loss 0.02214750 - time (sec): 414.77 - samples/sec: 482.36 - lr: 0.000053 - momentum: 0.000000
194
+ 2023-10-11 03:01:16,103 epoch 7 - iter 801/893 - loss 0.02204018 - time (sec): 464.71 - samples/sec: 483.40 - lr: 0.000052 - momentum: 0.000000
195
+ 2023-10-11 03:02:04,996 epoch 7 - iter 890/893 - loss 0.02241433 - time (sec): 513.60 - samples/sec: 482.95 - lr: 0.000050 - momentum: 0.000000
196
+ 2023-10-11 03:02:06,525 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-11 03:02:06,526 EPOCH 7 done: loss 0.0224 - lr: 0.000050
198
+ 2023-10-11 03:02:27,764 DEV : loss 0.15427739918231964 - f1-score (micro avg) 0.7838
199
+ 2023-10-11 03:02:27,794 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-11 03:03:17,261 epoch 8 - iter 89/893 - loss 0.01644759 - time (sec): 49.46 - samples/sec: 486.76 - lr: 0.000048 - momentum: 0.000000
201
+ 2023-10-11 03:04:05,512 epoch 8 - iter 178/893 - loss 0.01439801 - time (sec): 97.72 - samples/sec: 497.24 - lr: 0.000047 - momentum: 0.000000
202
+ 2023-10-11 03:04:58,009 epoch 8 - iter 267/893 - loss 0.01392832 - time (sec): 150.21 - samples/sec: 489.72 - lr: 0.000045 - momentum: 0.000000
203
+ 2023-10-11 03:05:52,634 epoch 8 - iter 356/893 - loss 0.01599008 - time (sec): 204.84 - samples/sec: 484.90 - lr: 0.000043 - momentum: 0.000000
204
+ 2023-10-11 03:06:48,250 epoch 8 - iter 445/893 - loss 0.01659639 - time (sec): 260.45 - samples/sec: 474.24 - lr: 0.000042 - momentum: 0.000000
205
+ 2023-10-11 03:07:44,160 epoch 8 - iter 534/893 - loss 0.01829444 - time (sec): 316.36 - samples/sec: 468.84 - lr: 0.000040 - momentum: 0.000000
206
+ 2023-10-11 03:08:36,417 epoch 8 - iter 623/893 - loss 0.01815970 - time (sec): 368.62 - samples/sec: 468.62 - lr: 0.000038 - momentum: 0.000000
207
+ 2023-10-11 03:09:32,746 epoch 8 - iter 712/893 - loss 0.01780340 - time (sec): 424.95 - samples/sec: 466.93 - lr: 0.000037 - momentum: 0.000000
208
+ 2023-10-11 03:10:25,528 epoch 8 - iter 801/893 - loss 0.01774196 - time (sec): 477.73 - samples/sec: 466.21 - lr: 0.000035 - momentum: 0.000000
209
+ 2023-10-11 03:11:20,029 epoch 8 - iter 890/893 - loss 0.01749401 - time (sec): 532.23 - samples/sec: 465.57 - lr: 0.000033 - momentum: 0.000000
210
+ 2023-10-11 03:11:21,784 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-11 03:11:21,784 EPOCH 8 done: loss 0.0176 - lr: 0.000033
212
+ 2023-10-11 03:11:44,400 DEV : loss 0.17111782729625702 - f1-score (micro avg) 0.8003
213
+ 2023-10-11 03:11:44,436 saving best model
214
+ 2023-10-11 03:11:46,991 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-11 03:12:39,531 epoch 9 - iter 89/893 - loss 0.01256651 - time (sec): 52.54 - samples/sec: 491.41 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-11 03:13:33,242 epoch 9 - iter 178/893 - loss 0.01534048 - time (sec): 106.25 - samples/sec: 472.82 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-11 03:14:27,777 epoch 9 - iter 267/893 - loss 0.01378958 - time (sec): 160.78 - samples/sec: 465.55 - lr: 0.000028 - momentum: 0.000000
218
+ 2023-10-11 03:15:20,646 epoch 9 - iter 356/893 - loss 0.01377709 - time (sec): 213.65 - samples/sec: 464.37 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-11 03:16:13,432 epoch 9 - iter 445/893 - loss 0.01258058 - time (sec): 266.44 - samples/sec: 465.09 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-11 03:17:04,781 epoch 9 - iter 534/893 - loss 0.01314215 - time (sec): 317.79 - samples/sec: 467.46 - lr: 0.000023 - momentum: 0.000000
221
+ 2023-10-11 03:17:54,474 epoch 9 - iter 623/893 - loss 0.01350092 - time (sec): 367.48 - samples/sec: 466.69 - lr: 0.000022 - momentum: 0.000000
222
+ 2023-10-11 03:18:48,473 epoch 9 - iter 712/893 - loss 0.01447242 - time (sec): 421.48 - samples/sec: 467.37 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-11 03:19:43,268 epoch 9 - iter 801/893 - loss 0.01452537 - time (sec): 476.27 - samples/sec: 468.86 - lr: 0.000019 - momentum: 0.000000
224
+ 2023-10-11 03:20:34,952 epoch 9 - iter 890/893 - loss 0.01466660 - time (sec): 527.96 - samples/sec: 469.52 - lr: 0.000017 - momentum: 0.000000
225
+ 2023-10-11 03:20:36,621 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-11 03:20:36,621 EPOCH 9 done: loss 0.0147 - lr: 0.000017
227
+ 2023-10-11 03:20:59,124 DEV : loss 0.17805925011634827 - f1-score (micro avg) 0.7981
228
+ 2023-10-11 03:20:59,158 ----------------------------------------------------------------------------------------------------
229
+ 2023-10-11 03:21:47,938 epoch 10 - iter 89/893 - loss 0.01343074 - time (sec): 48.78 - samples/sec: 481.71 - lr: 0.000015 - momentum: 0.000000
230
+ 2023-10-11 03:22:37,293 epoch 10 - iter 178/893 - loss 0.01307677 - time (sec): 98.13 - samples/sec: 492.76 - lr: 0.000013 - momentum: 0.000000
231
+ 2023-10-11 03:23:27,892 epoch 10 - iter 267/893 - loss 0.01186327 - time (sec): 148.73 - samples/sec: 497.09 - lr: 0.000012 - momentum: 0.000000
232
+ 2023-10-11 03:24:17,105 epoch 10 - iter 356/893 - loss 0.01107913 - time (sec): 197.95 - samples/sec: 494.17 - lr: 0.000010 - momentum: 0.000000
233
+ 2023-10-11 03:25:07,498 epoch 10 - iter 445/893 - loss 0.01167501 - time (sec): 248.34 - samples/sec: 499.17 - lr: 0.000008 - momentum: 0.000000
234
+ 2023-10-11 03:25:59,498 epoch 10 - iter 534/893 - loss 0.01203564 - time (sec): 300.34 - samples/sec: 501.72 - lr: 0.000007 - momentum: 0.000000
235
+ 2023-10-11 03:26:49,245 epoch 10 - iter 623/893 - loss 0.01234827 - time (sec): 350.09 - samples/sec: 496.60 - lr: 0.000005 - momentum: 0.000000
236
+ 2023-10-11 03:27:41,346 epoch 10 - iter 712/893 - loss 0.01235296 - time (sec): 402.19 - samples/sec: 495.65 - lr: 0.000004 - momentum: 0.000000
237
+ 2023-10-11 03:28:36,110 epoch 10 - iter 801/893 - loss 0.01233653 - time (sec): 456.95 - samples/sec: 489.75 - lr: 0.000002 - momentum: 0.000000
238
+ 2023-10-11 03:29:27,337 epoch 10 - iter 890/893 - loss 0.01217264 - time (sec): 508.18 - samples/sec: 488.21 - lr: 0.000000 - momentum: 0.000000
239
+ 2023-10-11 03:29:28,848 ----------------------------------------------------------------------------------------------------
240
+ 2023-10-11 03:29:28,848 EPOCH 10 done: loss 0.0121 - lr: 0.000000
241
+ 2023-10-11 03:29:51,390 DEV : loss 0.1803114116191864 - f1-score (micro avg) 0.7928
242
+ 2023-10-11 03:29:52,270 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-11 03:29:52,272 Loading model from best epoch ...
244
+ 2023-10-11 03:29:56,254 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
245
+ 2023-10-11 03:31:03,930
246
+ Results:
247
+ - F-score (micro) 0.7132
248
+ - F-score (macro) 0.6199
249
+ - Accuracy 0.5691
250
+
251
+ By class:
252
+ precision recall f1-score support
253
+
254
+ LOC 0.7458 0.7342 0.7400 1095
255
+ PER 0.7884 0.7806 0.7845 1012
256
+ ORG 0.4342 0.5910 0.5006 357
257
+ HumanProd 0.3636 0.6061 0.4545 33
258
+
259
+ micro avg 0.6963 0.7309 0.7132 2497
260
+ macro avg 0.5830 0.6780 0.6199 2497
261
+ weighted avg 0.7135 0.7309 0.7200 2497
262
+
263
+ 2023-10-11 03:31:03,930 ----------------------------------------------------------------------------------------------------