Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbert_64k-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
d1bb402
•
1 Parent(s): 629ad88

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +244 -0
training.log ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-25 16:26:03,207 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 16:26:03,208 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(64001, 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-25 16:26:03,208 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 16:26:03,208 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
52
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
53
+ 2023-10-25 16:26:03,208 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 16:26:03,208 Train: 7142 sentences
55
+ 2023-10-25 16:26:03,208 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 16:26:03,208 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 16:26:03,208 Training Params:
58
+ 2023-10-25 16:26:03,208 - learning_rate: "3e-05"
59
+ 2023-10-25 16:26:03,208 - mini_batch_size: "4"
60
+ 2023-10-25 16:26:03,208 - max_epochs: "10"
61
+ 2023-10-25 16:26:03,208 - shuffle: "True"
62
+ 2023-10-25 16:26:03,208 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 16:26:03,208 Plugins:
64
+ 2023-10-25 16:26:03,208 - TensorboardLogger
65
+ 2023-10-25 16:26:03,208 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 16:26:03,208 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 16:26:03,208 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 16:26:03,209 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 16:26:03,209 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 16:26:03,209 Computation:
71
+ 2023-10-25 16:26:03,209 - compute on device: cuda:0
72
+ 2023-10-25 16:26:03,209 - embedding storage: none
73
+ 2023-10-25 16:26:03,209 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 16:26:03,209 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
75
+ 2023-10-25 16:26:03,209 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 16:26:03,209 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 16:26:03,209 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 16:26:12,778 epoch 1 - iter 178/1786 - loss 1.72389527 - time (sec): 9.57 - samples/sec: 2499.38 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-25 16:26:22,214 epoch 1 - iter 356/1786 - loss 1.11489437 - time (sec): 19.00 - samples/sec: 2559.80 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-25 16:26:31,700 epoch 1 - iter 534/1786 - loss 0.85501815 - time (sec): 28.49 - samples/sec: 2565.38 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-25 16:26:41,250 epoch 1 - iter 712/1786 - loss 0.69063084 - time (sec): 38.04 - samples/sec: 2617.48 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-25 16:26:50,769 epoch 1 - iter 890/1786 - loss 0.59262743 - time (sec): 47.56 - samples/sec: 2593.45 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-25 16:27:00,269 epoch 1 - iter 1068/1786 - loss 0.52253362 - time (sec): 57.06 - samples/sec: 2598.06 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-25 16:27:09,548 epoch 1 - iter 1246/1786 - loss 0.46852649 - time (sec): 66.34 - samples/sec: 2629.99 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-25 16:27:18,901 epoch 1 - iter 1424/1786 - loss 0.42908150 - time (sec): 75.69 - samples/sec: 2630.06 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-25 16:27:28,243 epoch 1 - iter 1602/1786 - loss 0.39816297 - time (sec): 85.03 - samples/sec: 2627.16 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-25 16:27:37,690 epoch 1 - iter 1780/1786 - loss 0.37429285 - time (sec): 94.48 - samples/sec: 2626.94 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-25 16:27:37,985 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 16:27:37,985 EPOCH 1 done: loss 0.3737 - lr: 0.000030
90
+ 2023-10-25 16:27:42,018 DEV : loss 0.11665168404579163 - f1-score (micro avg) 0.7065
91
+ 2023-10-25 16:27:42,041 saving best model
92
+ 2023-10-25 16:27:42,503 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 16:27:51,329 epoch 2 - iter 178/1786 - loss 0.10461983 - time (sec): 8.82 - samples/sec: 2853.63 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-25 16:28:00,430 epoch 2 - iter 356/1786 - loss 0.11400338 - time (sec): 17.93 - samples/sec: 2630.89 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-25 16:28:09,657 epoch 2 - iter 534/1786 - loss 0.11338586 - time (sec): 27.15 - samples/sec: 2665.75 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-25 16:28:19,268 epoch 2 - iter 712/1786 - loss 0.11187233 - time (sec): 36.76 - samples/sec: 2721.39 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-25 16:28:28,609 epoch 2 - iter 890/1786 - loss 0.11112092 - time (sec): 46.10 - samples/sec: 2702.62 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-25 16:28:37,715 epoch 2 - iter 1068/1786 - loss 0.11480554 - time (sec): 55.21 - samples/sec: 2693.47 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-25 16:28:47,250 epoch 2 - iter 1246/1786 - loss 0.11362112 - time (sec): 64.75 - samples/sec: 2683.58 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-25 16:28:56,729 epoch 2 - iter 1424/1786 - loss 0.11345575 - time (sec): 74.22 - samples/sec: 2663.61 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-25 16:29:06,404 epoch 2 - iter 1602/1786 - loss 0.11331499 - time (sec): 83.90 - samples/sec: 2661.42 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-25 16:29:15,745 epoch 2 - iter 1780/1786 - loss 0.11389365 - time (sec): 93.24 - samples/sec: 2659.65 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-25 16:29:16,043 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 16:29:16,043 EPOCH 2 done: loss 0.1141 - lr: 0.000027
105
+ 2023-10-25 16:29:20,360 DEV : loss 0.15264025330543518 - f1-score (micro avg) 0.7495
106
+ 2023-10-25 16:29:20,382 saving best model
107
+ 2023-10-25 16:29:21,051 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-25 16:29:30,883 epoch 3 - iter 178/1786 - loss 0.07629665 - time (sec): 9.83 - samples/sec: 2419.01 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-25 16:29:40,271 epoch 3 - iter 356/1786 - loss 0.06863531 - time (sec): 19.22 - samples/sec: 2509.35 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-25 16:29:49,295 epoch 3 - iter 534/1786 - loss 0.07758255 - time (sec): 28.24 - samples/sec: 2575.16 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-25 16:29:58,201 epoch 3 - iter 712/1786 - loss 0.07961424 - time (sec): 37.15 - samples/sec: 2586.45 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-25 16:30:07,236 epoch 3 - iter 890/1786 - loss 0.08178772 - time (sec): 46.18 - samples/sec: 2596.65 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-25 16:30:16,358 epoch 3 - iter 1068/1786 - loss 0.08404606 - time (sec): 55.30 - samples/sec: 2628.21 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-25 16:30:25,496 epoch 3 - iter 1246/1786 - loss 0.08188804 - time (sec): 64.44 - samples/sec: 2647.28 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-25 16:30:35,432 epoch 3 - iter 1424/1786 - loss 0.08055218 - time (sec): 74.38 - samples/sec: 2653.16 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-25 16:30:44,086 epoch 3 - iter 1602/1786 - loss 0.07964001 - time (sec): 83.03 - samples/sec: 2683.32 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-25 16:30:52,968 epoch 3 - iter 1780/1786 - loss 0.07759944 - time (sec): 91.91 - samples/sec: 2697.45 - lr: 0.000023 - momentum: 0.000000
118
+ 2023-10-25 16:30:53,255 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-25 16:30:53,255 EPOCH 3 done: loss 0.0776 - lr: 0.000023
120
+ 2023-10-25 16:30:57,163 DEV : loss 0.13464441895484924 - f1-score (micro avg) 0.7748
121
+ 2023-10-25 16:30:57,184 saving best model
122
+ 2023-10-25 16:30:57,839 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-25 16:31:06,592 epoch 4 - iter 178/1786 - loss 0.04440627 - time (sec): 8.75 - samples/sec: 2715.40 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-25 16:31:15,474 epoch 4 - iter 356/1786 - loss 0.05450741 - time (sec): 17.63 - samples/sec: 2733.26 - lr: 0.000023 - momentum: 0.000000
125
+ 2023-10-25 16:31:24,257 epoch 4 - iter 534/1786 - loss 0.05779653 - time (sec): 26.41 - samples/sec: 2703.26 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-25 16:31:33,177 epoch 4 - iter 712/1786 - loss 0.05478759 - time (sec): 35.33 - samples/sec: 2760.99 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-25 16:31:41,855 epoch 4 - iter 890/1786 - loss 0.05436695 - time (sec): 44.01 - samples/sec: 2764.41 - lr: 0.000022 - momentum: 0.000000
128
+ 2023-10-25 16:31:50,747 epoch 4 - iter 1068/1786 - loss 0.05371101 - time (sec): 52.90 - samples/sec: 2780.40 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-25 16:31:59,923 epoch 4 - iter 1246/1786 - loss 0.05398691 - time (sec): 62.08 - samples/sec: 2776.06 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-25 16:32:09,269 epoch 4 - iter 1424/1786 - loss 0.05442093 - time (sec): 71.43 - samples/sec: 2779.56 - lr: 0.000021 - momentum: 0.000000
131
+ 2023-10-25 16:32:18,063 epoch 4 - iter 1602/1786 - loss 0.05445676 - time (sec): 80.22 - samples/sec: 2774.76 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-25 16:32:27,167 epoch 4 - iter 1780/1786 - loss 0.05567026 - time (sec): 89.32 - samples/sec: 2778.43 - lr: 0.000020 - momentum: 0.000000
133
+ 2023-10-25 16:32:27,482 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-25 16:32:27,482 EPOCH 4 done: loss 0.0557 - lr: 0.000020
135
+ 2023-10-25 16:32:32,322 DEV : loss 0.1592852622270584 - f1-score (micro avg) 0.7807
136
+ 2023-10-25 16:32:32,358 saving best model
137
+ 2023-10-25 16:32:33,008 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-25 16:32:42,076 epoch 5 - iter 178/1786 - loss 0.03505898 - time (sec): 9.07 - samples/sec: 2675.18 - lr: 0.000020 - momentum: 0.000000
139
+ 2023-10-25 16:32:51,082 epoch 5 - iter 356/1786 - loss 0.03624807 - time (sec): 18.07 - samples/sec: 2659.88 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-25 16:33:00,526 epoch 5 - iter 534/1786 - loss 0.03535087 - time (sec): 27.52 - samples/sec: 2648.84 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-25 16:33:10,149 epoch 5 - iter 712/1786 - loss 0.03748850 - time (sec): 37.14 - samples/sec: 2613.01 - lr: 0.000019 - momentum: 0.000000
142
+ 2023-10-25 16:33:18,985 epoch 5 - iter 890/1786 - loss 0.03909736 - time (sec): 45.97 - samples/sec: 2621.81 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-25 16:33:27,925 epoch 5 - iter 1068/1786 - loss 0.03966780 - time (sec): 54.91 - samples/sec: 2632.78 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-25 16:33:37,092 epoch 5 - iter 1246/1786 - loss 0.03878930 - time (sec): 64.08 - samples/sec: 2662.85 - lr: 0.000018 - momentum: 0.000000
145
+ 2023-10-25 16:33:45,813 epoch 5 - iter 1424/1786 - loss 0.03882227 - time (sec): 72.80 - samples/sec: 2682.22 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-25 16:33:55,003 epoch 5 - iter 1602/1786 - loss 0.03840917 - time (sec): 81.99 - samples/sec: 2715.84 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-25 16:34:03,749 epoch 5 - iter 1780/1786 - loss 0.03930397 - time (sec): 90.74 - samples/sec: 2734.76 - lr: 0.000017 - momentum: 0.000000
148
+ 2023-10-25 16:34:04,047 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-25 16:34:04,047 EPOCH 5 done: loss 0.0395 - lr: 0.000017
150
+ 2023-10-25 16:34:07,972 DEV : loss 0.18587026000022888 - f1-score (micro avg) 0.762
151
+ 2023-10-25 16:34:07,993 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-25 16:34:17,461 epoch 6 - iter 178/1786 - loss 0.02228299 - time (sec): 9.46 - samples/sec: 2513.38 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-25 16:34:26,900 epoch 6 - iter 356/1786 - loss 0.02707967 - time (sec): 18.90 - samples/sec: 2582.19 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-25 16:34:36,267 epoch 6 - iter 534/1786 - loss 0.02969054 - time (sec): 28.27 - samples/sec: 2599.89 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-25 16:34:45,851 epoch 6 - iter 712/1786 - loss 0.02850474 - time (sec): 37.86 - samples/sec: 2596.22 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-25 16:34:55,626 epoch 6 - iter 890/1786 - loss 0.02952048 - time (sec): 47.63 - samples/sec: 2593.98 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-25 16:35:05,298 epoch 6 - iter 1068/1786 - loss 0.02959778 - time (sec): 57.30 - samples/sec: 2579.41 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-25 16:35:14,998 epoch 6 - iter 1246/1786 - loss 0.02950384 - time (sec): 67.00 - samples/sec: 2578.56 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-25 16:35:24,826 epoch 6 - iter 1424/1786 - loss 0.02899248 - time (sec): 76.83 - samples/sec: 2587.24 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-25 16:35:34,276 epoch 6 - iter 1602/1786 - loss 0.02952807 - time (sec): 86.28 - samples/sec: 2584.11 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-25 16:35:43,735 epoch 6 - iter 1780/1786 - loss 0.02944428 - time (sec): 95.74 - samples/sec: 2590.73 - lr: 0.000013 - momentum: 0.000000
162
+ 2023-10-25 16:35:44,048 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-25 16:35:44,049 EPOCH 6 done: loss 0.0295 - lr: 0.000013
164
+ 2023-10-25 16:35:49,183 DEV : loss 0.18982860445976257 - f1-score (micro avg) 0.7865
165
+ 2023-10-25 16:35:49,208 saving best model
166
+ 2023-10-25 16:35:49,872 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-25 16:35:59,426 epoch 7 - iter 178/1786 - loss 0.02267264 - time (sec): 9.55 - samples/sec: 2718.81 - lr: 0.000013 - momentum: 0.000000
168
+ 2023-10-25 16:36:08,982 epoch 7 - iter 356/1786 - loss 0.02071867 - time (sec): 19.11 - samples/sec: 2692.77 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-25 16:36:18,619 epoch 7 - iter 534/1786 - loss 0.02250539 - time (sec): 28.74 - samples/sec: 2676.34 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-25 16:36:28,445 epoch 7 - iter 712/1786 - loss 0.02189393 - time (sec): 38.57 - samples/sec: 2645.65 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-25 16:36:37,705 epoch 7 - iter 890/1786 - loss 0.02316080 - time (sec): 47.83 - samples/sec: 2631.90 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-25 16:36:46,646 epoch 7 - iter 1068/1786 - loss 0.02272730 - time (sec): 56.77 - samples/sec: 2667.04 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-25 16:36:55,632 epoch 7 - iter 1246/1786 - loss 0.02207885 - time (sec): 65.76 - samples/sec: 2665.92 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-25 16:37:05,058 epoch 7 - iter 1424/1786 - loss 0.02206815 - time (sec): 75.18 - samples/sec: 2654.28 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-25 16:37:14,475 epoch 7 - iter 1602/1786 - loss 0.02192510 - time (sec): 84.60 - samples/sec: 2655.92 - lr: 0.000010 - momentum: 0.000000
176
+ 2023-10-25 16:37:23,960 epoch 7 - iter 1780/1786 - loss 0.02204271 - time (sec): 94.09 - samples/sec: 2636.41 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-25 16:37:24,276 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-25 16:37:24,276 EPOCH 7 done: loss 0.0220 - lr: 0.000010
179
+ 2023-10-25 16:37:29,189 DEV : loss 0.19655928015708923 - f1-score (micro avg) 0.7919
180
+ 2023-10-25 16:37:29,211 saving best model
181
+ 2023-10-25 16:37:29,862 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-25 16:37:39,385 epoch 8 - iter 178/1786 - loss 0.01876484 - time (sec): 9.52 - samples/sec: 2613.60 - lr: 0.000010 - momentum: 0.000000
183
+ 2023-10-25 16:37:48,866 epoch 8 - iter 356/1786 - loss 0.01519318 - time (sec): 19.00 - samples/sec: 2526.15 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-25 16:37:58,302 epoch 8 - iter 534/1786 - loss 0.01373478 - time (sec): 28.44 - samples/sec: 2626.74 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-25 16:38:07,293 epoch 8 - iter 712/1786 - loss 0.01454254 - time (sec): 37.43 - samples/sec: 2645.54 - lr: 0.000009 - momentum: 0.000000
186
+ 2023-10-25 16:38:16,013 epoch 8 - iter 890/1786 - loss 0.01484691 - time (sec): 46.15 - samples/sec: 2685.45 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-25 16:38:24,785 epoch 8 - iter 1068/1786 - loss 0.01499282 - time (sec): 54.92 - samples/sec: 2733.34 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-25 16:38:33,747 epoch 8 - iter 1246/1786 - loss 0.01424428 - time (sec): 63.88 - samples/sec: 2734.25 - lr: 0.000008 - momentum: 0.000000
189
+ 2023-10-25 16:38:42,790 epoch 8 - iter 1424/1786 - loss 0.01406695 - time (sec): 72.93 - samples/sec: 2738.06 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-25 16:38:51,780 epoch 8 - iter 1602/1786 - loss 0.01437150 - time (sec): 81.92 - samples/sec: 2723.87 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-25 16:39:00,844 epoch 8 - iter 1780/1786 - loss 0.01533311 - time (sec): 90.98 - samples/sec: 2725.94 - lr: 0.000007 - momentum: 0.000000
192
+ 2023-10-25 16:39:01,145 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-25 16:39:01,145 EPOCH 8 done: loss 0.0153 - lr: 0.000007
194
+ 2023-10-25 16:39:05,268 DEV : loss 0.21181654930114746 - f1-score (micro avg) 0.7842
195
+ 2023-10-25 16:39:05,288 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-25 16:39:14,500 epoch 9 - iter 178/1786 - loss 0.00816641 - time (sec): 9.21 - samples/sec: 2907.50 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-25 16:39:23,973 epoch 9 - iter 356/1786 - loss 0.01013131 - time (sec): 18.68 - samples/sec: 2780.63 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-25 16:39:33,578 epoch 9 - iter 534/1786 - loss 0.00954521 - time (sec): 28.29 - samples/sec: 2756.96 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-25 16:39:43,144 epoch 9 - iter 712/1786 - loss 0.00969192 - time (sec): 37.85 - samples/sec: 2674.70 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-25 16:39:52,805 epoch 9 - iter 890/1786 - loss 0.01062753 - time (sec): 47.51 - samples/sec: 2602.39 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-25 16:40:01,825 epoch 9 - iter 1068/1786 - loss 0.01065964 - time (sec): 56.53 - samples/sec: 2596.08 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-25 16:40:10,834 epoch 9 - iter 1246/1786 - loss 0.01011395 - time (sec): 65.54 - samples/sec: 2627.85 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-25 16:40:19,865 epoch 9 - iter 1424/1786 - loss 0.01040764 - time (sec): 74.58 - samples/sec: 2643.30 - lr: 0.000004 - momentum: 0.000000
204
+ 2023-10-25 16:40:28,886 epoch 9 - iter 1602/1786 - loss 0.01055140 - time (sec): 83.60 - samples/sec: 2648.51 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-25 16:40:37,881 epoch 9 - iter 1780/1786 - loss 0.01019071 - time (sec): 92.59 - samples/sec: 2679.20 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-25 16:40:38,162 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-25 16:40:38,163 EPOCH 9 done: loss 0.0102 - lr: 0.000003
208
+ 2023-10-25 16:40:43,043 DEV : loss 0.23426829278469086 - f1-score (micro avg) 0.7848
209
+ 2023-10-25 16:40:43,066 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-25 16:40:52,584 epoch 10 - iter 178/1786 - loss 0.00771099 - time (sec): 9.52 - samples/sec: 2746.65 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-25 16:41:02,050 epoch 10 - iter 356/1786 - loss 0.00805677 - time (sec): 18.98 - samples/sec: 2646.83 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-25 16:41:11,538 epoch 10 - iter 534/1786 - loss 0.00817679 - time (sec): 28.47 - samples/sec: 2656.92 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-25 16:41:20,720 epoch 10 - iter 712/1786 - loss 0.00878410 - time (sec): 37.65 - samples/sec: 2640.29 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-25 16:41:29,664 epoch 10 - iter 890/1786 - loss 0.00887421 - time (sec): 46.60 - samples/sec: 2652.81 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-25 16:41:38,852 epoch 10 - iter 1068/1786 - loss 0.00789479 - time (sec): 55.78 - samples/sec: 2679.34 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-25 16:41:47,525 epoch 10 - iter 1246/1786 - loss 0.00784114 - time (sec): 64.46 - samples/sec: 2718.01 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-25 16:41:56,178 epoch 10 - iter 1424/1786 - loss 0.00742785 - time (sec): 73.11 - samples/sec: 2736.80 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-25 16:42:04,870 epoch 10 - iter 1602/1786 - loss 0.00792667 - time (sec): 81.80 - samples/sec: 2734.65 - lr: 0.000000 - momentum: 0.000000
219
+ 2023-10-25 16:42:13,881 epoch 10 - iter 1780/1786 - loss 0.00755996 - time (sec): 90.81 - samples/sec: 2733.58 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-25 16:42:14,170 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-25 16:42:14,171 EPOCH 10 done: loss 0.0075 - lr: 0.000000
222
+ 2023-10-25 16:42:18,903 DEV : loss 0.23706591129302979 - f1-score (micro avg) 0.7914
223
+ 2023-10-25 16:42:19,357 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-25 16:42:19,358 Loading model from best epoch ...
225
+ 2023-10-25 16:42:21,248 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
226
+ 2023-10-25 16:42:34,839
227
+ Results:
228
+ - F-score (micro) 0.6924
229
+ - F-score (macro) 0.62
230
+ - Accuracy 0.5472
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ LOC 0.7289 0.6703 0.6984 1095
236
+ PER 0.7611 0.7806 0.7707 1012
237
+ ORG 0.4414 0.5910 0.5054 357
238
+ HumanProd 0.3966 0.6970 0.5055 33
239
+
240
+ micro avg 0.6811 0.7040 0.6924 2497
241
+ macro avg 0.5820 0.6847 0.6200 2497
242
+ weighted avg 0.6964 0.7040 0.6976 2497
243
+
244
+ 2023-10-25 16:42:34,839 ----------------------------------------------------------------------------------------------------