Hugging Face's logo

hmbyt5-preliminary
/
flair-hipe-2022-topres19th-en

Token Classification
Flair
PyTorch
TensorBoard
English
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
f526174
·
1 Parent(s): 44f2aab

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.1697244053.6d4c7681f95b.3224.14 +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 sha256:0dc2205772d22e94b45de58ec73859fd5e55abe87e14ea838606c265fc4bbc31
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 sha256:9d10056dc010f0bba4d29f82454c39a50606bfb6a7f514cdc76651e4e522a943
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 00:48:28 0.0001 0.9011 0.0813 0.5217 0.6582 0.5821 0.4262
3
+ 2 00:56:01 0.0001 0.0918 0.0590 0.7132 0.7975 0.7530 0.6364
4
+ 3 01:03:38 0.0001 0.0535 0.0566 0.7698 0.8608 0.8127 0.6962
5
+ 4 01:11:14 0.0001 0.0326 0.0654 0.8080 0.8523 0.8296 0.7214
6
+ 5 01:18:53 0.0001 0.0210 0.0730 0.7812 0.8439 0.8114 0.7042
7
+ 6 01:26:29 0.0001 0.0143 0.0767 0.7857 0.8819 0.8310 0.7207
8
+ 7 01:34:01 0.0001 0.0096 0.0880 0.8023 0.8734 0.8364 0.7340
9
+ 8 01:41:54 0.0000 0.0059 0.0929 0.7909 0.8776 0.8320 0.7273
10
+ 9 01:49:20 0.0000 0.0045 0.1013 0.7969 0.8608 0.8276 0.7208
11
+ 10 01:56:55 0.0000 0.0025 0.1049 0.7938 0.8608 0.8259 0.7234
runs/events.out.tfevents.1697244053.6d4c7681f95b.3224.14 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34a30e3ce5f15d9264ab0f80d9217d12e60115ff031fcaa8e9cb542c5c84d45c
3
+ size 864636
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,263 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 00:40:53,759 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 00:40:53,762 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
+ 2023-10-14 00:40:53,762 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-14 00:40:53,762 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
72
+ - NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
73
+ 2023-10-14 00:40:53,762 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-14 00:40:53,763 Train: 6183 sentences
75
+ 2023-10-14 00:40:53,763 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-14 00:40:53,763 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-14 00:40:53,763 Training Params:
78
+ 2023-10-14 00:40:53,763 - learning_rate: "0.00015"
79
+ 2023-10-14 00:40:53,763 - mini_batch_size: "4"
80
+ 2023-10-14 00:40:53,763 - max_epochs: "10"
81
+ 2023-10-14 00:40:53,763 - shuffle: "True"
82
+ 2023-10-14 00:40:53,763 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-14 00:40:53,763 Plugins:
84
+ 2023-10-14 00:40:53,763 - TensorboardLogger
85
+ 2023-10-14 00:40:53,763 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-14 00:40:53,763 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 00:40:53,764 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-14 00:40:53,764 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-14 00:40:53,764 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-14 00:40:53,764 Computation:
91
+ 2023-10-14 00:40:53,764 - compute on device: cuda:0
92
+ 2023-10-14 00:40:53,764 - embedding storage: none
93
+ 2023-10-14 00:40:53,764 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-14 00:40:53,764 Model training base path: "hmbench-topres19th/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-4"
95
+ 2023-10-14 00:40:53,764 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-14 00:40:53,764 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-14 00:40:53,764 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-14 00:41:37,315 epoch 1 - iter 154/1546 - loss 2.56873151 - time (sec): 43.55 - samples/sec: 288.51 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-14 00:42:21,594 epoch 1 - iter 308/1546 - loss 2.43379623 - time (sec): 87.83 - samples/sec: 291.12 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-14 00:43:04,591 epoch 1 - iter 462/1546 - loss 2.16358210 - time (sec): 130.82 - samples/sec: 293.17 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-14 00:43:47,729 epoch 1 - iter 616/1546 - loss 1.89703690 - time (sec): 173.96 - samples/sec: 286.04 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-14 00:44:31,467 epoch 1 - iter 770/1546 - loss 1.61404026 - time (sec): 217.70 - samples/sec: 285.41 - lr: 0.000075 - momentum: 0.000000
103
+ 2023-10-14 00:45:15,241 epoch 1 - iter 924/1546 - loss 1.39110007 - time (sec): 261.47 - samples/sec: 282.67 - lr: 0.000090 - momentum: 0.000000
104
+ 2023-10-14 00:45:59,472 epoch 1 - iter 1078/1546 - loss 1.22539657 - time (sec): 305.71 - samples/sec: 282.34 - lr: 0.000104 - momentum: 0.000000
105
+ 2023-10-14 00:46:43,473 epoch 1 - iter 1232/1546 - loss 1.09447369 - time (sec): 349.71 - samples/sec: 282.98 - lr: 0.000119 - momentum: 0.000000
106
+ 2023-10-14 00:47:26,775 epoch 1 - iter 1386/1546 - loss 0.98722789 - time (sec): 393.01 - samples/sec: 284.48 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-14 00:48:09,530 epoch 1 - iter 1540/1546 - loss 0.90394722 - time (sec): 435.76 - samples/sec: 284.17 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-14 00:48:11,100 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-14 00:48:11,100 EPOCH 1 done: loss 0.9011 - lr: 0.000149
110
+ 2023-10-14 00:48:28,371 DEV : loss 0.0812714695930481 - f1-score (micro avg) 0.5821
111
+ 2023-10-14 00:48:28,400 saving best model
112
+ 2023-10-14 00:48:29,339 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-14 00:49:11,816 epoch 2 - iter 154/1546 - loss 0.10318258 - time (sec): 42.47 - samples/sec: 258.58 - lr: 0.000148 - momentum: 0.000000
114
+ 2023-10-14 00:49:55,140 epoch 2 - iter 308/1546 - loss 0.10981228 - time (sec): 85.80 - samples/sec: 280.28 - lr: 0.000147 - momentum: 0.000000
115
+ 2023-10-14 00:50:38,438 epoch 2 - iter 462/1546 - loss 0.10677416 - time (sec): 129.10 - samples/sec: 279.42 - lr: 0.000145 - momentum: 0.000000
116
+ 2023-10-14 00:51:21,839 epoch 2 - iter 616/1546 - loss 0.10494738 - time (sec): 172.50 - samples/sec: 283.09 - lr: 0.000143 - momentum: 0.000000
117
+ 2023-10-14 00:52:05,915 epoch 2 - iter 770/1546 - loss 0.10120027 - time (sec): 216.57 - samples/sec: 285.19 - lr: 0.000142 - momentum: 0.000000
118
+ 2023-10-14 00:52:49,780 epoch 2 - iter 924/1546 - loss 0.09745622 - time (sec): 260.44 - samples/sec: 287.52 - lr: 0.000140 - momentum: 0.000000
119
+ 2023-10-14 00:53:33,706 epoch 2 - iter 1078/1546 - loss 0.09614656 - time (sec): 304.36 - samples/sec: 285.46 - lr: 0.000138 - momentum: 0.000000
120
+ 2023-10-14 00:54:16,839 epoch 2 - iter 1232/1546 - loss 0.09374842 - time (sec): 347.50 - samples/sec: 285.24 - lr: 0.000137 - momentum: 0.000000
121
+ 2023-10-14 00:54:59,232 epoch 2 - iter 1386/1546 - loss 0.09264200 - time (sec): 389.89 - samples/sec: 284.46 - lr: 0.000135 - momentum: 0.000000
122
+ 2023-10-14 00:55:42,457 epoch 2 - iter 1540/1546 - loss 0.09188658 - time (sec): 433.12 - samples/sec: 285.66 - lr: 0.000133 - momentum: 0.000000
123
+ 2023-10-14 00:55:44,125 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-14 00:55:44,125 EPOCH 2 done: loss 0.0918 - lr: 0.000133
125
+ 2023-10-14 00:56:01,105 DEV : loss 0.05896108224987984 - f1-score (micro avg) 0.753
126
+ 2023-10-14 00:56:01,138 saving best model
127
+ 2023-10-14 00:56:02,107 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-14 00:56:45,802 epoch 3 - iter 154/1546 - loss 0.03817025 - time (sec): 43.69 - samples/sec: 290.81 - lr: 0.000132 - momentum: 0.000000
129
+ 2023-10-14 00:57:30,409 epoch 3 - iter 308/1546 - loss 0.04755001 - time (sec): 88.30 - samples/sec: 285.37 - lr: 0.000130 - momentum: 0.000000
130
+ 2023-10-14 00:58:14,419 epoch 3 - iter 462/1546 - loss 0.05104447 - time (sec): 132.31 - samples/sec: 287.17 - lr: 0.000128 - momentum: 0.000000
131
+ 2023-10-14 00:58:59,278 epoch 3 - iter 616/1546 - loss 0.05222974 - time (sec): 177.17 - samples/sec: 284.35 - lr: 0.000127 - momentum: 0.000000
132
+ 2023-10-14 00:59:42,378 epoch 3 - iter 770/1546 - loss 0.05664685 - time (sec): 220.27 - samples/sec: 283.57 - lr: 0.000125 - momentum: 0.000000
133
+ 2023-10-14 01:00:26,466 epoch 3 - iter 924/1546 - loss 0.05615839 - time (sec): 264.36 - samples/sec: 283.26 - lr: 0.000123 - momentum: 0.000000
134
+ 2023-10-14 01:01:10,213 epoch 3 - iter 1078/1546 - loss 0.05563682 - time (sec): 308.10 - samples/sec: 283.56 - lr: 0.000122 - momentum: 0.000000
135
+ 2023-10-14 01:01:53,935 epoch 3 - iter 1232/1546 - loss 0.05496907 - time (sec): 351.83 - samples/sec: 283.05 - lr: 0.000120 - momentum: 0.000000
136
+ 2023-10-14 01:02:36,567 epoch 3 - iter 1386/1546 - loss 0.05420399 - time (sec): 394.46 - samples/sec: 282.68 - lr: 0.000118 - momentum: 0.000000
137
+ 2023-10-14 01:03:19,488 epoch 3 - iter 1540/1546 - loss 0.05358357 - time (sec): 437.38 - samples/sec: 283.24 - lr: 0.000117 - momentum: 0.000000
138
+ 2023-10-14 01:03:21,125 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-14 01:03:21,126 EPOCH 3 done: loss 0.0535 - lr: 0.000117
140
+ 2023-10-14 01:03:38,668 DEV : loss 0.05659706890583038 - f1-score (micro avg) 0.8127
141
+ 2023-10-14 01:03:38,696 saving best model
142
+ 2023-10-14 01:03:39,676 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-14 01:04:22,928 epoch 4 - iter 154/1546 - loss 0.02880123 - time (sec): 43.25 - samples/sec: 282.77 - lr: 0.000115 - momentum: 0.000000
144
+ 2023-10-14 01:05:05,591 epoch 4 - iter 308/1546 - loss 0.03320371 - time (sec): 85.91 - samples/sec: 281.37 - lr: 0.000113 - momentum: 0.000000
145
+ 2023-10-14 01:05:47,828 epoch 4 - iter 462/1546 - loss 0.03355342 - time (sec): 128.15 - samples/sec: 277.43 - lr: 0.000112 - momentum: 0.000000
146
+ 2023-10-14 01:06:32,557 epoch 4 - iter 616/1546 - loss 0.03292566 - time (sec): 172.88 - samples/sec: 281.16 - lr: 0.000110 - momentum: 0.000000
147
+ 2023-10-14 01:07:16,430 epoch 4 - iter 770/1546 - loss 0.03538356 - time (sec): 216.75 - samples/sec: 283.12 - lr: 0.000108 - momentum: 0.000000
148
+ 2023-10-14 01:08:00,152 epoch 4 - iter 924/1546 - loss 0.03499173 - time (sec): 260.47 - samples/sec: 282.29 - lr: 0.000107 - momentum: 0.000000
149
+ 2023-10-14 01:08:43,630 epoch 4 - iter 1078/1546 - loss 0.03358414 - time (sec): 303.95 - samples/sec: 282.46 - lr: 0.000105 - momentum: 0.000000
150
+ 2023-10-14 01:09:26,021 epoch 4 - iter 1232/1546 - loss 0.03409530 - time (sec): 346.34 - samples/sec: 282.45 - lr: 0.000103 - momentum: 0.000000
151
+ 2023-10-14 01:10:10,598 epoch 4 - iter 1386/1546 - loss 0.03269230 - time (sec): 390.92 - samples/sec: 284.77 - lr: 0.000102 - momentum: 0.000000
152
+ 2023-10-14 01:10:54,547 epoch 4 - iter 1540/1546 - loss 0.03249523 - time (sec): 434.87 - samples/sec: 284.58 - lr: 0.000100 - momentum: 0.000000
153
+ 2023-10-14 01:10:56,205 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-14 01:10:56,206 EPOCH 4 done: loss 0.0326 - lr: 0.000100
155
+ 2023-10-14 01:11:14,223 DEV : loss 0.06543166935443878 - f1-score (micro avg) 0.8296
156
+ 2023-10-14 01:11:14,257 saving best model
157
+ 2023-10-14 01:11:16,978 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-14 01:12:01,955 epoch 5 - iter 154/1546 - loss 0.02269561 - time (sec): 44.97 - samples/sec: 277.46 - lr: 0.000098 - momentum: 0.000000
159
+ 2023-10-14 01:12:45,258 epoch 5 - iter 308/1546 - loss 0.01877881 - time (sec): 88.28 - samples/sec: 282.06 - lr: 0.000097 - momentum: 0.000000
160
+ 2023-10-14 01:13:28,365 epoch 5 - iter 462/1546 - loss 0.01896534 - time (sec): 131.38 - samples/sec: 284.70 - lr: 0.000095 - momentum: 0.000000
161
+ 2023-10-14 01:14:12,356 epoch 5 - iter 616/1546 - loss 0.01903863 - time (sec): 175.37 - samples/sec: 282.49 - lr: 0.000093 - momentum: 0.000000
162
+ 2023-10-14 01:14:55,863 epoch 5 - iter 770/1546 - loss 0.01875930 - time (sec): 218.88 - samples/sec: 285.22 - lr: 0.000092 - momentum: 0.000000
163
+ 2023-10-14 01:15:39,889 epoch 5 - iter 924/1546 - loss 0.01806369 - time (sec): 262.91 - samples/sec: 285.58 - lr: 0.000090 - momentum: 0.000000
164
+ 2023-10-14 01:16:24,044 epoch 5 - iter 1078/1546 - loss 0.01918173 - time (sec): 307.06 - samples/sec: 285.19 - lr: 0.000088 - momentum: 0.000000
165
+ 2023-10-14 01:17:07,380 epoch 5 - iter 1232/1546 - loss 0.01910152 - time (sec): 350.40 - samples/sec: 282.90 - lr: 0.000087 - momentum: 0.000000
166
+ 2023-10-14 01:17:50,259 epoch 5 - iter 1386/1546 - loss 0.02046787 - time (sec): 393.28 - samples/sec: 283.93 - lr: 0.000085 - momentum: 0.000000
167
+ 2023-10-14 01:18:34,363 epoch 5 - iter 1540/1546 - loss 0.02101615 - time (sec): 437.38 - samples/sec: 282.89 - lr: 0.000083 - momentum: 0.000000
168
+ 2023-10-14 01:18:36,049 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-14 01:18:36,049 EPOCH 5 done: loss 0.0210 - lr: 0.000083
170
+ 2023-10-14 01:18:53,025 DEV : loss 0.07295508682727814 - f1-score (micro avg) 0.8114
171
+ 2023-10-14 01:18:53,055 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-14 01:19:37,032 epoch 6 - iter 154/1546 - loss 0.01805609 - time (sec): 43.98 - samples/sec: 280.23 - lr: 0.000082 - momentum: 0.000000
173
+ 2023-10-14 01:20:20,915 epoch 6 - iter 308/1546 - loss 0.01358401 - time (sec): 87.86 - samples/sec: 282.61 - lr: 0.000080 - momentum: 0.000000
174
+ 2023-10-14 01:21:04,910 epoch 6 - iter 462/1546 - loss 0.01234981 - time (sec): 131.85 - samples/sec: 286.36 - lr: 0.000078 - momentum: 0.000000
175
+ 2023-10-14 01:21:48,433 epoch 6 - iter 616/1546 - loss 0.01241203 - time (sec): 175.38 - samples/sec: 286.78 - lr: 0.000077 - momentum: 0.000000
176
+ 2023-10-14 01:22:31,732 epoch 6 - iter 770/1546 - loss 0.01401568 - time (sec): 218.67 - samples/sec: 283.93 - lr: 0.000075 - momentum: 0.000000
177
+ 2023-10-14 01:23:15,331 epoch 6 - iter 924/1546 - loss 0.01424007 - time (sec): 262.27 - samples/sec: 282.40 - lr: 0.000073 - momentum: 0.000000
178
+ 2023-10-14 01:23:59,294 epoch 6 - iter 1078/1546 - loss 0.01451504 - time (sec): 306.24 - samples/sec: 282.27 - lr: 0.000072 - momentum: 0.000000
179
+ 2023-10-14 01:24:42,465 epoch 6 - iter 1232/1546 - loss 0.01474960 - time (sec): 349.41 - samples/sec: 281.30 - lr: 0.000070 - momentum: 0.000000
180
+ 2023-10-14 01:25:25,815 epoch 6 - iter 1386/1546 - loss 0.01493525 - time (sec): 392.76 - samples/sec: 281.51 - lr: 0.000068 - momentum: 0.000000
181
+ 2023-10-14 01:26:09,605 epoch 6 - iter 1540/1546 - loss 0.01413867 - time (sec): 436.55 - samples/sec: 283.39 - lr: 0.000067 - momentum: 0.000000
182
+ 2023-10-14 01:26:11,277 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-14 01:26:11,277 EPOCH 6 done: loss 0.0143 - lr: 0.000067
184
+ 2023-10-14 01:26:29,293 DEV : loss 0.07670143991708755 - f1-score (micro avg) 0.831
185
+ 2023-10-14 01:26:29,335 saving best model
186
+ 2023-10-14 01:26:31,945 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-14 01:27:18,143 epoch 7 - iter 154/1546 - loss 0.00975317 - time (sec): 46.19 - samples/sec: 297.57 - lr: 0.000065 - momentum: 0.000000
188
+ 2023-10-14 01:28:00,843 epoch 7 - iter 308/1546 - loss 0.00891956 - time (sec): 88.89 - samples/sec: 292.11 - lr: 0.000063 - momentum: 0.000000
189
+ 2023-10-14 01:28:44,318 epoch 7 - iter 462/1546 - loss 0.00997728 - time (sec): 132.37 - samples/sec: 290.55 - lr: 0.000062 - momentum: 0.000000
190
+ 2023-10-14 01:29:26,892 epoch 7 - iter 616/1546 - loss 0.00930809 - time (sec): 174.94 - samples/sec: 290.48 - lr: 0.000060 - momentum: 0.000000
191
+ 2023-10-14 01:30:09,203 epoch 7 - iter 770/1546 - loss 0.00916108 - time (sec): 217.25 - samples/sec: 287.44 - lr: 0.000058 - momentum: 0.000000
192
+ 2023-10-14 01:30:51,937 epoch 7 - iter 924/1546 - loss 0.00924439 - time (sec): 259.99 - samples/sec: 290.46 - lr: 0.000057 - momentum: 0.000000
193
+ 2023-10-14 01:31:33,959 epoch 7 - iter 1078/1546 - loss 0.01049403 - time (sec): 302.01 - samples/sec: 290.64 - lr: 0.000055 - momentum: 0.000000
194
+ 2023-10-14 01:32:16,483 epoch 7 - iter 1232/1546 - loss 0.01043974 - time (sec): 344.53 - samples/sec: 288.74 - lr: 0.000053 - momentum: 0.000000
195
+ 2023-10-14 01:32:59,452 epoch 7 - iter 1386/1546 - loss 0.00997770 - time (sec): 387.50 - samples/sec: 287.00 - lr: 0.000052 - momentum: 0.000000
196
+ 2023-10-14 01:33:42,672 epoch 7 - iter 1540/1546 - loss 0.00960048 - time (sec): 430.72 - samples/sec: 287.24 - lr: 0.000050 - momentum: 0.000000
197
+ 2023-10-14 01:33:44,315 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-14 01:33:44,316 EPOCH 7 done: loss 0.0096 - lr: 0.000050
199
+ 2023-10-14 01:34:01,530 DEV : loss 0.0880291685461998 - f1-score (micro avg) 0.8364
200
+ 2023-10-14 01:34:01,560 saving best model
201
+ 2023-10-14 01:34:04,174 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-14 01:34:47,366 epoch 8 - iter 154/1546 - loss 0.00390031 - time (sec): 43.19 - samples/sec: 293.43 - lr: 0.000048 - momentum: 0.000000
203
+ 2023-10-14 01:35:29,570 epoch 8 - iter 308/1546 - loss 0.00354778 - time (sec): 85.39 - samples/sec: 283.34 - lr: 0.000047 - momentum: 0.000000
204
+ 2023-10-14 01:36:12,267 epoch 8 - iter 462/1546 - loss 0.00558009 - time (sec): 128.09 - samples/sec: 285.33 - lr: 0.000045 - momentum: 0.000000
205
+ 2023-10-14 01:36:56,668 epoch 8 - iter 616/1546 - loss 0.00613750 - time (sec): 172.49 - samples/sec: 280.88 - lr: 0.000043 - momentum: 0.000000
206
+ 2023-10-14 01:37:43,215 epoch 8 - iter 770/1546 - loss 0.00600191 - time (sec): 219.04 - samples/sec: 278.43 - lr: 0.000042 - momentum: 0.000000
207
+ 2023-10-14 01:38:29,251 epoch 8 - iter 924/1546 - loss 0.00583790 - time (sec): 265.07 - samples/sec: 277.56 - lr: 0.000040 - momentum: 0.000000
208
+ 2023-10-14 01:39:14,114 epoch 8 - iter 1078/1546 - loss 0.00603055 - time (sec): 309.94 - samples/sec: 274.06 - lr: 0.000038 - momentum: 0.000000
209
+ 2023-10-14 01:40:01,383 epoch 8 - iter 1232/1546 - loss 0.00653849 - time (sec): 357.21 - samples/sec: 274.08 - lr: 0.000037 - momentum: 0.000000
210
+ 2023-10-14 01:40:49,090 epoch 8 - iter 1386/1546 - loss 0.00610073 - time (sec): 404.91 - samples/sec: 274.75 - lr: 0.000035 - momentum: 0.000000
211
+ 2023-10-14 01:41:35,518 epoch 8 - iter 1540/1546 - loss 0.00595387 - time (sec): 451.34 - samples/sec: 274.45 - lr: 0.000033 - momentum: 0.000000
212
+ 2023-10-14 01:41:37,155 ----------------------------------------------------------------------------------------------------
213
+ 2023-10-14 01:41:37,155 EPOCH 8 done: loss 0.0059 - lr: 0.000033
214
+ 2023-10-14 01:41:54,940 DEV : loss 0.09291724860668182 - f1-score (micro avg) 0.832
215
+ 2023-10-14 01:41:54,969 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-14 01:42:39,414 epoch 9 - iter 154/1546 - loss 0.00195984 - time (sec): 44.44 - samples/sec: 283.42 - lr: 0.000032 - momentum: 0.000000
217
+ 2023-10-14 01:43:23,371 epoch 9 - iter 308/1546 - loss 0.00226573 - time (sec): 88.40 - samples/sec: 280.80 - lr: 0.000030 - momentum: 0.000000
218
+ 2023-10-14 01:44:07,873 epoch 9 - iter 462/1546 - loss 0.00286662 - time (sec): 132.90 - samples/sec: 284.35 - lr: 0.000028 - momentum: 0.000000
219
+ 2023-10-14 01:44:52,315 epoch 9 - iter 616/1546 - loss 0.00407165 - time (sec): 177.34 - samples/sec: 284.00 - lr: 0.000027 - momentum: 0.000000
220
+ 2023-10-14 01:45:35,291 epoch 9 - iter 770/1546 - loss 0.00463367 - time (sec): 220.32 - samples/sec: 282.55 - lr: 0.000025 - momentum: 0.000000
221
+ 2023-10-14 01:46:18,120 epoch 9 - iter 924/1546 - loss 0.00464126 - time (sec): 263.15 - samples/sec: 283.13 - lr: 0.000023 - momentum: 0.000000
222
+ 2023-10-14 01:46:58,607 epoch 9 - iter 1078/1546 - loss 0.00453876 - time (sec): 303.64 - samples/sec: 288.14 - lr: 0.000022 - momentum: 0.000000
223
+ 2023-10-14 01:47:38,679 epoch 9 - iter 1232/1546 - loss 0.00471355 - time (sec): 343.71 - samples/sec: 290.15 - lr: 0.000020 - momentum: 0.000000
224
+ 2023-10-14 01:48:19,657 epoch 9 - iter 1386/1546 - loss 0.00446859 - time (sec): 384.69 - samples/sec: 292.58 - lr: 0.000018 - momentum: 0.000000
225
+ 2023-10-14 01:49:01,929 epoch 9 - iter 1540/1546 - loss 0.00448726 - time (sec): 426.96 - samples/sec: 289.78 - lr: 0.000017 - momentum: 0.000000
226
+ 2023-10-14 01:49:03,704 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-14 01:49:03,705 EPOCH 9 done: loss 0.0045 - lr: 0.000017
228
+ 2023-10-14 01:49:20,835 DEV : loss 0.10131476074457169 - f1-score (micro avg) 0.8276
229
+ 2023-10-14 01:49:20,865 ----------------------------------------------------------------------------------------------------
230
+ 2023-10-14 01:50:04,766 epoch 10 - iter 154/1546 - loss 0.00119822 - time (sec): 43.90 - samples/sec: 288.50 - lr: 0.000015 - momentum: 0.000000
231
+ 2023-10-14 01:50:47,340 epoch 10 - iter 308/1546 - loss 0.00178207 - time (sec): 86.47 - samples/sec: 277.32 - lr: 0.000013 - momentum: 0.000000
232
+ 2023-10-14 01:51:30,796 epoch 10 - iter 462/1546 - loss 0.00153142 - time (sec): 129.93 - samples/sec: 283.13 - lr: 0.000012 - momentum: 0.000000
233
+ 2023-10-14 01:52:14,216 epoch 10 - iter 616/1546 - loss 0.00162289 - time (sec): 173.35 - samples/sec: 286.02 - lr: 0.000010 - momentum: 0.000000
234
+ 2023-10-14 01:52:57,272 epoch 10 - iter 770/1546 - loss 0.00187635 - time (sec): 216.40 - samples/sec: 285.41 - lr: 0.000008 - momentum: 0.000000
235
+ 2023-10-14 01:53:41,326 epoch 10 - iter 924/1546 - loss 0.00182691 - time (sec): 260.46 - samples/sec: 284.76 - lr: 0.000007 - momentum: 0.000000
236
+ 2023-10-14 01:54:24,005 epoch 10 - iter 1078/1546 - loss 0.00196491 - time (sec): 303.14 - samples/sec: 284.52 - lr: 0.000005 - momentum: 0.000000
237
+ 2023-10-14 01:55:07,582 epoch 10 - iter 1232/1546 - loss 0.00209036 - time (sec): 346.71 - samples/sec: 285.53 - lr: 0.000003 - momentum: 0.000000
238
+ 2023-10-14 01:55:51,488 epoch 10 - iter 1386/1546 - loss 0.00233145 - time (sec): 390.62 - samples/sec: 283.48 - lr: 0.000002 - momentum: 0.000000
239
+ 2023-10-14 01:56:35,504 epoch 10 - iter 1540/1546 - loss 0.00253840 - time (sec): 434.64 - samples/sec: 284.86 - lr: 0.000000 - momentum: 0.000000
240
+ 2023-10-14 01:56:37,127 ----------------------------------------------------------------------------------------------------
241
+ 2023-10-14 01:56:37,127 EPOCH 10 done: loss 0.0025 - lr: 0.000000
242
+ 2023-10-14 01:56:55,037 DEV : loss 0.10494109988212585 - f1-score (micro avg) 0.8259
243
+ 2023-10-14 01:56:55,989 ----------------------------------------------------------------------------------------------------
244
+ 2023-10-14 01:56:55,991 Loading model from best epoch ...
245
+ 2023-10-14 01:57:00,432 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
246
+ 2023-10-14 01:57:55,195
247
+ Results:
248
+ - F-score (micro) 0.7978
249
+ - F-score (macro) 0.713
250
+ - Accuracy 0.6828
251
+
252
+ By class:
253
+ precision recall f1-score support
254
+
255
+ LOC 0.8436 0.8552 0.8493 946
256
+ BUILDING 0.5588 0.5135 0.5352 185
257
+ STREET 0.7414 0.7679 0.7544 56
258
+
259
+ micro avg 0.7978 0.7978 0.7978 1187
260
+ macro avg 0.7146 0.7122 0.7130 1187
261
+ weighted avg 0.7944 0.7978 0.7959 1187
262
+
263
+ 2023-10-14 01:57:55,195 ----------------------------------------------------------------------------------------------------