Hugging Face's logo

stefan-it
/
hmbench-topres19th-en-hmbyt5-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
English
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
9da7dcb
·
1 Parent(s): 04809c4

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.1697199110.6d4c7681f95b.3224.4 +3 -0
  6. test.tsv +0 -0
  7. training.log +262 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dc0206d7f36a4465f2efb6cfd28b51f89db372b750cb0da71d5389463f1b40f
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 sha256:613f7c226ad90ad008efd9c7e728624c024953e5043db4e5ff220abc9194fea2
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 12:19:20 0.0001 1.1083 0.1016 0.0000 0.0000 0.0000 0.0000
3
+ 2 12:26:30 0.0001 0.1042 0.0559 0.7965 0.7764 0.7863 0.6691
4
+ 3 12:33:47 0.0001 0.0611 0.0515 0.7782 0.8143 0.7959 0.6772
5
+ 4 12:41:06 0.0001 0.0403 0.0635 0.7669 0.8608 0.8111 0.6915
6
+ 5 12:48:23 0.0001 0.0259 0.0732 0.7895 0.8228 0.8058 0.6890
7
+ 6 12:55:28 0.0001 0.0174 0.0744 0.7565 0.8650 0.8071 0.6949
8
+ 7 13:02:31 0.0001 0.0125 0.0863 0.7727 0.8608 0.8144 0.7010
9
+ 8 13:09:37 0.0000 0.0086 0.0887 0.7977 0.8650 0.8300 0.7270
10
+ 9 13:16:45 0.0000 0.0063 0.0927 0.7821 0.8481 0.8138 0.7003
11
+ 10 13:24:07 0.0000 0.0045 0.0917 0.7812 0.8439 0.8114 0.6993
runs/events.out.tfevents.1697199110.6d4c7681f95b.3224.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5670489d08ebedf838b1f37e1ed840ee9dbf2a70cc51ec67f689acf985c86006
3
+ size 434848
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 12:11:50,968 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 12:11:50,971 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
+ 2023-10-13 12:11:50,971 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-13 12:11:50,971 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
+ 2023-10-13 12:11:50,971 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-13 12:11:50,971 Train: 6183 sentences
75
+ 2023-10-13 12:11:50,971 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-13 12:11:50,971 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-13 12:11:50,971 Training Params:
78
+ 2023-10-13 12:11:50,972 - learning_rate: "0.00015"
79
+ 2023-10-13 12:11:50,972 - mini_batch_size: "8"
80
+ 2023-10-13 12:11:50,972 - max_epochs: "10"
81
+ 2023-10-13 12:11:50,972 - shuffle: "True"
82
+ 2023-10-13 12:11:50,972 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-13 12:11:50,972 Plugins:
84
+ 2023-10-13 12:11:50,972 - TensorboardLogger
85
+ 2023-10-13 12:11:50,972 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-13 12:11:50,972 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 12:11:50,972 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-13 12:11:50,972 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-13 12:11:50,972 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-13 12:11:50,972 Computation:
91
+ 2023-10-13 12:11:50,972 - compute on device: cuda:0
92
+ 2023-10-13 12:11:50,973 - embedding storage: none
93
+ 2023-10-13 12:11:50,973 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-13 12:11:50,973 Model training base path: "hmbench-topres19th/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
95
+ 2023-10-13 12:11:50,973 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-13 12:11:50,973 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-13 12:11:50,973 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-13 12:12:34,518 epoch 1 - iter 77/773 - loss 2.57123659 - time (sec): 43.54 - samples/sec: 283.97 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-13 12:13:16,883 epoch 1 - iter 154/773 - loss 2.54004423 - time (sec): 85.91 - samples/sec: 278.44 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-13 12:13:59,430 epoch 1 - iter 231/773 - loss 2.38127903 - time (sec): 128.46 - samples/sec: 285.60 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-13 12:14:44,917 epoch 1 - iter 308/773 - loss 2.14419240 - time (sec): 173.94 - samples/sec: 289.56 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-13 12:15:28,830 epoch 1 - iter 385/773 - loss 1.91099630 - time (sec): 217.86 - samples/sec: 288.68 - lr: 0.000075 - momentum: 0.000000
103
+ 2023-10-13 12:16:11,609 epoch 1 - iter 462/773 - loss 1.70018509 - time (sec): 260.63 - samples/sec: 285.81 - lr: 0.000089 - momentum: 0.000000
104
+ 2023-10-13 12:16:53,244 epoch 1 - iter 539/773 - loss 1.51041487 - time (sec): 302.27 - samples/sec: 284.97 - lr: 0.000104 - momentum: 0.000000
105
+ 2023-10-13 12:17:35,038 epoch 1 - iter 616/773 - loss 1.35148076 - time (sec): 344.06 - samples/sec: 285.00 - lr: 0.000119 - momentum: 0.000000
106
+ 2023-10-13 12:18:17,952 epoch 1 - iter 693/773 - loss 1.21911109 - time (sec): 386.98 - samples/sec: 286.83 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-13 12:19:01,031 epoch 1 - iter 770/773 - loss 1.11179986 - time (sec): 430.06 - samples/sec: 287.98 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-13 12:19:02,586 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-13 12:19:02,586 EPOCH 1 done: loss 1.1083 - lr: 0.000149
110
+ 2023-10-13 12:19:20,096 DEV : loss 0.10156270116567612 - f1-score (micro avg) 0.0
111
+ 2023-10-13 12:19:20,135 ----------------------------------------------------------------------------------------------------
112
+ 2023-10-13 12:20:03,733 epoch 2 - iter 77/773 - loss 0.14343911 - time (sec): 43.59 - samples/sec: 277.88 - lr: 0.000148 - momentum: 0.000000
113
+ 2023-10-13 12:20:45,246 epoch 2 - iter 154/773 - loss 0.14111574 - time (sec): 85.11 - samples/sec: 292.96 - lr: 0.000147 - momentum: 0.000000
114
+ 2023-10-13 12:21:24,773 epoch 2 - iter 231/773 - loss 0.13134074 - time (sec): 124.63 - samples/sec: 298.67 - lr: 0.000145 - momentum: 0.000000
115
+ 2023-10-13 12:22:04,927 epoch 2 - iter 308/773 - loss 0.12226088 - time (sec): 164.79 - samples/sec: 303.09 - lr: 0.000143 - momentum: 0.000000
116
+ 2023-10-13 12:22:44,343 epoch 2 - iter 385/773 - loss 0.11876136 - time (sec): 204.20 - samples/sec: 301.47 - lr: 0.000142 - momentum: 0.000000
117
+ 2023-10-13 12:23:25,052 epoch 2 - iter 462/773 - loss 0.11190683 - time (sec): 244.91 - samples/sec: 304.46 - lr: 0.000140 - momentum: 0.000000
118
+ 2023-10-13 12:24:07,027 epoch 2 - iter 539/773 - loss 0.11157185 - time (sec): 286.89 - samples/sec: 304.72 - lr: 0.000138 - momentum: 0.000000
119
+ 2023-10-13 12:24:48,683 epoch 2 - iter 616/773 - loss 0.10933321 - time (sec): 328.54 - samples/sec: 303.00 - lr: 0.000137 - momentum: 0.000000
120
+ 2023-10-13 12:25:30,303 epoch 2 - iter 693/773 - loss 0.10704561 - time (sec): 370.16 - samples/sec: 302.26 - lr: 0.000135 - momentum: 0.000000
121
+ 2023-10-13 12:26:11,495 epoch 2 - iter 770/773 - loss 0.10437133 - time (sec): 411.36 - samples/sec: 301.17 - lr: 0.000133 - momentum: 0.000000
122
+ 2023-10-13 12:26:12,989 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-13 12:26:12,990 EPOCH 2 done: loss 0.1042 - lr: 0.000133
124
+ 2023-10-13 12:26:30,440 DEV : loss 0.055891476571559906 - f1-score (micro avg) 0.7863
125
+ 2023-10-13 12:26:30,470 saving best model
126
+ 2023-10-13 12:26:31,427 ----------------------------------------------------------------------------------------------------
127
+ 2023-10-13 12:27:13,239 epoch 3 - iter 77/773 - loss 0.07137218 - time (sec): 41.81 - samples/sec: 304.93 - lr: 0.000132 - momentum: 0.000000
128
+ 2023-10-13 12:27:54,626 epoch 3 - iter 154/773 - loss 0.06623441 - time (sec): 83.20 - samples/sec: 303.37 - lr: 0.000130 - momentum: 0.000000
129
+ 2023-10-13 12:28:36,413 epoch 3 - iter 231/773 - loss 0.06634603 - time (sec): 124.98 - samples/sec: 295.12 - lr: 0.000128 - momentum: 0.000000
130
+ 2023-10-13 12:29:19,074 epoch 3 - iter 308/773 - loss 0.06925348 - time (sec): 167.64 - samples/sec: 298.65 - lr: 0.000127 - momentum: 0.000000
131
+ 2023-10-13 12:29:59,867 epoch 3 - iter 385/773 - loss 0.06673849 - time (sec): 208.44 - samples/sec: 297.41 - lr: 0.000125 - momentum: 0.000000
132
+ 2023-10-13 12:30:41,086 epoch 3 - iter 462/773 - loss 0.06675991 - time (sec): 249.66 - samples/sec: 297.38 - lr: 0.000123 - momentum: 0.000000
133
+ 2023-10-13 12:31:23,319 epoch 3 - iter 539/773 - loss 0.06556065 - time (sec): 291.89 - samples/sec: 298.20 - lr: 0.000122 - momentum: 0.000000
134
+ 2023-10-13 12:32:05,608 epoch 3 - iter 616/773 - loss 0.06289494 - time (sec): 334.18 - samples/sec: 299.27 - lr: 0.000120 - momentum: 0.000000
135
+ 2023-10-13 12:32:46,787 epoch 3 - iter 693/773 - loss 0.06170302 - time (sec): 375.36 - samples/sec: 299.34 - lr: 0.000118 - momentum: 0.000000
136
+ 2023-10-13 12:33:27,330 epoch 3 - iter 770/773 - loss 0.06115844 - time (sec): 415.90 - samples/sec: 297.40 - lr: 0.000117 - momentum: 0.000000
137
+ 2023-10-13 12:33:28,989 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-13 12:33:28,989 EPOCH 3 done: loss 0.0611 - lr: 0.000117
139
+ 2023-10-13 12:33:47,675 DEV : loss 0.051485899835824966 - f1-score (micro avg) 0.7959
140
+ 2023-10-13 12:33:47,708 saving best model
141
+ 2023-10-13 12:33:50,361 ----------------------------------------------------------------------------------------------------
142
+ 2023-10-13 12:34:30,646 epoch 4 - iter 77/773 - loss 0.04947376 - time (sec): 40.28 - samples/sec: 280.87 - lr: 0.000115 - momentum: 0.000000
143
+ 2023-10-13 12:35:12,235 epoch 4 - iter 154/773 - loss 0.04506746 - time (sec): 81.87 - samples/sec: 298.04 - lr: 0.000113 - momentum: 0.000000
144
+ 2023-10-13 12:35:53,297 epoch 4 - iter 231/773 - loss 0.04471770 - time (sec): 122.93 - samples/sec: 293.77 - lr: 0.000112 - momentum: 0.000000
145
+ 2023-10-13 12:36:36,166 epoch 4 - iter 308/773 - loss 0.04502334 - time (sec): 165.80 - samples/sec: 301.74 - lr: 0.000110 - momentum: 0.000000
146
+ 2023-10-13 12:37:17,047 epoch 4 - iter 385/773 - loss 0.04266320 - time (sec): 206.68 - samples/sec: 299.71 - lr: 0.000108 - momentum: 0.000000
147
+ 2023-10-13 12:37:59,706 epoch 4 - iter 462/773 - loss 0.04112529 - time (sec): 249.34 - samples/sec: 299.60 - lr: 0.000107 - momentum: 0.000000
148
+ 2023-10-13 12:38:41,524 epoch 4 - iter 539/773 - loss 0.04114812 - time (sec): 291.16 - samples/sec: 299.15 - lr: 0.000105 - momentum: 0.000000
149
+ 2023-10-13 12:39:22,331 epoch 4 - iter 616/773 - loss 0.03933986 - time (sec): 331.97 - samples/sec: 300.79 - lr: 0.000103 - momentum: 0.000000
150
+ 2023-10-13 12:40:03,518 epoch 4 - iter 693/773 - loss 0.04060396 - time (sec): 373.15 - samples/sec: 299.37 - lr: 0.000102 - momentum: 0.000000
151
+ 2023-10-13 12:40:46,318 epoch 4 - iter 770/773 - loss 0.04042546 - time (sec): 415.95 - samples/sec: 297.67 - lr: 0.000100 - momentum: 0.000000
152
+ 2023-10-13 12:40:47,905 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-13 12:40:47,905 EPOCH 4 done: loss 0.0403 - lr: 0.000100
154
+ 2023-10-13 12:41:06,017 DEV : loss 0.06352678686380386 - f1-score (micro avg) 0.8111
155
+ 2023-10-13 12:41:06,046 saving best model
156
+ 2023-10-13 12:41:08,698 ----------------------------------------------------------------------------------------------------
157
+ 2023-10-13 12:41:50,692 epoch 5 - iter 77/773 - loss 0.02612534 - time (sec): 41.99 - samples/sec: 291.71 - lr: 0.000098 - momentum: 0.000000
158
+ 2023-10-13 12:42:31,783 epoch 5 - iter 154/773 - loss 0.02825120 - time (sec): 83.08 - samples/sec: 293.80 - lr: 0.000097 - momentum: 0.000000
159
+ 2023-10-13 12:43:13,014 epoch 5 - iter 231/773 - loss 0.02760483 - time (sec): 124.31 - samples/sec: 303.77 - lr: 0.000095 - momentum: 0.000000
160
+ 2023-10-13 12:43:55,603 epoch 5 - iter 308/773 - loss 0.02738820 - time (sec): 166.90 - samples/sec: 303.64 - lr: 0.000093 - momentum: 0.000000
161
+ 2023-10-13 12:44:37,368 epoch 5 - iter 385/773 - loss 0.02615460 - time (sec): 208.67 - samples/sec: 298.58 - lr: 0.000092 - momentum: 0.000000
162
+ 2023-10-13 12:45:20,043 epoch 5 - iter 462/773 - loss 0.02553182 - time (sec): 251.34 - samples/sec: 297.00 - lr: 0.000090 - momentum: 0.000000
163
+ 2023-10-13 12:46:01,461 epoch 5 - iter 539/773 - loss 0.02485120 - time (sec): 292.76 - samples/sec: 295.15 - lr: 0.000088 - momentum: 0.000000
164
+ 2023-10-13 12:46:43,286 epoch 5 - iter 616/773 - loss 0.02594767 - time (sec): 334.58 - samples/sec: 297.42 - lr: 0.000087 - momentum: 0.000000
165
+ 2023-10-13 12:47:22,820 epoch 5 - iter 693/773 - loss 0.02558115 - time (sec): 374.12 - samples/sec: 298.56 - lr: 0.000085 - momentum: 0.000000
166
+ 2023-10-13 12:48:04,090 epoch 5 - iter 770/773 - loss 0.02571670 - time (sec): 415.39 - samples/sec: 297.82 - lr: 0.000083 - momentum: 0.000000
167
+ 2023-10-13 12:48:05,685 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-13 12:48:05,685 EPOCH 5 done: loss 0.0259 - lr: 0.000083
169
+ 2023-10-13 12:48:23,336 DEV : loss 0.07320675998926163 - f1-score (micro avg) 0.8058
170
+ 2023-10-13 12:48:23,366 ----------------------------------------------------------------------------------------------------
171
+ 2023-10-13 12:49:03,912 epoch 6 - iter 77/773 - loss 0.01551022 - time (sec): 40.54 - samples/sec: 276.10 - lr: 0.000082 - momentum: 0.000000
172
+ 2023-10-13 12:49:46,020 epoch 6 - iter 154/773 - loss 0.01573350 - time (sec): 82.65 - samples/sec: 292.58 - lr: 0.000080 - momentum: 0.000000
173
+ 2023-10-13 12:50:27,182 epoch 6 - iter 231/773 - loss 0.01748381 - time (sec): 123.81 - samples/sec: 293.12 - lr: 0.000078 - momentum: 0.000000
174
+ 2023-10-13 12:51:06,849 epoch 6 - iter 308/773 - loss 0.01658042 - time (sec): 163.48 - samples/sec: 296.68 - lr: 0.000077 - momentum: 0.000000
175
+ 2023-10-13 12:51:45,807 epoch 6 - iter 385/773 - loss 0.01809232 - time (sec): 202.44 - samples/sec: 299.61 - lr: 0.000075 - momentum: 0.000000
176
+ 2023-10-13 12:52:27,059 epoch 6 - iter 462/773 - loss 0.01736914 - time (sec): 243.69 - samples/sec: 302.60 - lr: 0.000073 - momentum: 0.000000
177
+ 2023-10-13 12:53:08,010 epoch 6 - iter 539/773 - loss 0.01676794 - time (sec): 284.64 - samples/sec: 302.43 - lr: 0.000072 - momentum: 0.000000
178
+ 2023-10-13 12:53:48,567 epoch 6 - iter 616/773 - loss 0.01830890 - time (sec): 325.20 - samples/sec: 304.50 - lr: 0.000070 - momentum: 0.000000
179
+ 2023-10-13 12:54:28,837 epoch 6 - iter 693/773 - loss 0.01814792 - time (sec): 365.47 - samples/sec: 304.81 - lr: 0.000068 - momentum: 0.000000
180
+ 2023-10-13 12:55:09,144 epoch 6 - iter 770/773 - loss 0.01745267 - time (sec): 405.78 - samples/sec: 304.88 - lr: 0.000067 - momentum: 0.000000
181
+ 2023-10-13 12:55:10,695 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-13 12:55:10,695 EPOCH 6 done: loss 0.0174 - lr: 0.000067
183
+ 2023-10-13 12:55:28,242 DEV : loss 0.07442227005958557 - f1-score (micro avg) 0.8071
184
+ 2023-10-13 12:55:28,272 ----------------------------------------------------------------------------------------------------
185
+ 2023-10-13 12:56:09,091 epoch 7 - iter 77/773 - loss 0.00999803 - time (sec): 40.82 - samples/sec: 302.86 - lr: 0.000065 - momentum: 0.000000
186
+ 2023-10-13 12:56:48,526 epoch 7 - iter 154/773 - loss 0.01303046 - time (sec): 80.25 - samples/sec: 303.67 - lr: 0.000063 - momentum: 0.000000
187
+ 2023-10-13 12:57:27,898 epoch 7 - iter 231/773 - loss 0.01221184 - time (sec): 119.62 - samples/sec: 305.15 - lr: 0.000062 - momentum: 0.000000
188
+ 2023-10-13 12:58:09,432 epoch 7 - iter 308/773 - loss 0.01126403 - time (sec): 161.16 - samples/sec: 310.29 - lr: 0.000060 - momentum: 0.000000
189
+ 2023-10-13 12:58:50,442 epoch 7 - iter 385/773 - loss 0.01243846 - time (sec): 202.17 - samples/sec: 309.82 - lr: 0.000058 - momentum: 0.000000
190
+ 2023-10-13 12:59:31,439 epoch 7 - iter 462/773 - loss 0.01232868 - time (sec): 243.17 - samples/sec: 306.15 - lr: 0.000057 - momentum: 0.000000
191
+ 2023-10-13 13:00:12,190 epoch 7 - iter 539/773 - loss 0.01239029 - time (sec): 283.92 - samples/sec: 304.08 - lr: 0.000055 - momentum: 0.000000
192
+ 2023-10-13 13:00:52,186 epoch 7 - iter 616/773 - loss 0.01259808 - time (sec): 323.91 - samples/sec: 305.64 - lr: 0.000054 - momentum: 0.000000
193
+ 2023-10-13 13:01:32,889 epoch 7 - iter 693/773 - loss 0.01254284 - time (sec): 364.62 - samples/sec: 306.14 - lr: 0.000052 - momentum: 0.000000
194
+ 2023-10-13 13:02:13,324 epoch 7 - iter 770/773 - loss 0.01237555 - time (sec): 405.05 - samples/sec: 306.02 - lr: 0.000050 - momentum: 0.000000
195
+ 2023-10-13 13:02:14,776 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-13 13:02:14,776 EPOCH 7 done: loss 0.0125 - lr: 0.000050
197
+ 2023-10-13 13:02:31,857 DEV : loss 0.08628595620393753 - f1-score (micro avg) 0.8144
198
+ 2023-10-13 13:02:31,886 saving best model
199
+ 2023-10-13 13:02:34,611 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-13 13:03:14,909 epoch 8 - iter 77/773 - loss 0.00855586 - time (sec): 40.29 - samples/sec: 299.53 - lr: 0.000048 - momentum: 0.000000
201
+ 2023-10-13 13:03:55,802 epoch 8 - iter 154/773 - loss 0.01073063 - time (sec): 81.19 - samples/sec: 309.92 - lr: 0.000047 - momentum: 0.000000
202
+ 2023-10-13 13:04:35,209 epoch 8 - iter 231/773 - loss 0.00879026 - time (sec): 120.59 - samples/sec: 308.27 - lr: 0.000045 - momentum: 0.000000
203
+ 2023-10-13 13:05:15,170 epoch 8 - iter 308/773 - loss 0.00919919 - time (sec): 160.55 - samples/sec: 303.49 - lr: 0.000043 - momentum: 0.000000
204
+ 2023-10-13 13:05:55,095 epoch 8 - iter 385/773 - loss 0.00880865 - time (sec): 200.48 - samples/sec: 299.32 - lr: 0.000042 - momentum: 0.000000
205
+ 2023-10-13 13:06:36,217 epoch 8 - iter 462/773 - loss 0.00921083 - time (sec): 241.60 - samples/sec: 302.65 - lr: 0.000040 - momentum: 0.000000
206
+ 2023-10-13 13:07:17,052 epoch 8 - iter 539/773 - loss 0.01001243 - time (sec): 282.44 - samples/sec: 306.16 - lr: 0.000039 - momentum: 0.000000
207
+ 2023-10-13 13:07:58,590 epoch 8 - iter 616/773 - loss 0.00960529 - time (sec): 323.97 - samples/sec: 306.67 - lr: 0.000037 - momentum: 0.000000
208
+ 2023-10-13 13:08:39,476 epoch 8 - iter 693/773 - loss 0.00919135 - time (sec): 364.86 - samples/sec: 306.10 - lr: 0.000035 - momentum: 0.000000
209
+ 2023-10-13 13:09:19,241 epoch 8 - iter 770/773 - loss 0.00868187 - time (sec): 404.63 - samples/sec: 305.60 - lr: 0.000034 - momentum: 0.000000
210
+ 2023-10-13 13:09:20,852 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-13 13:09:20,853 EPOCH 8 done: loss 0.0086 - lr: 0.000034
212
+ 2023-10-13 13:09:37,627 DEV : loss 0.08868994563817978 - f1-score (micro avg) 0.83
213
+ 2023-10-13 13:09:37,655 saving best model
214
+ 2023-10-13 13:09:40,347 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-13 13:10:21,144 epoch 9 - iter 77/773 - loss 0.00335222 - time (sec): 40.79 - samples/sec: 306.70 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-13 13:11:01,271 epoch 9 - iter 154/773 - loss 0.00362335 - time (sec): 80.92 - samples/sec: 313.31 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-13 13:11:41,379 epoch 9 - iter 231/773 - loss 0.00379545 - time (sec): 121.03 - samples/sec: 314.51 - lr: 0.000028 - momentum: 0.000000
218
+ 2023-10-13 13:12:21,676 epoch 9 - iter 308/773 - loss 0.00485971 - time (sec): 161.32 - samples/sec: 310.82 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-13 13:13:01,870 epoch 9 - iter 385/773 - loss 0.00587335 - time (sec): 201.52 - samples/sec: 312.32 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-13 13:13:43,524 epoch 9 - iter 462/773 - loss 0.00613505 - time (sec): 243.17 - samples/sec: 309.28 - lr: 0.000024 - momentum: 0.000000
221
+ 2023-10-13 13:14:24,212 epoch 9 - iter 539/773 - loss 0.00629327 - time (sec): 283.86 - samples/sec: 305.77 - lr: 0.000022 - momentum: 0.000000
222
+ 2023-10-13 13:15:04,808 epoch 9 - iter 616/773 - loss 0.00630642 - time (sec): 324.46 - samples/sec: 307.27 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-13 13:15:45,380 epoch 9 - iter 693/773 - loss 0.00629380 - time (sec): 365.03 - samples/sec: 306.37 - lr: 0.000019 - momentum: 0.000000
224
+ 2023-10-13 13:16:26,131 epoch 9 - iter 770/773 - loss 0.00636277 - time (sec): 405.78 - samples/sec: 305.61 - lr: 0.000017 - momentum: 0.000000
225
+ 2023-10-13 13:16:27,532 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-13 13:16:27,533 EPOCH 9 done: loss 0.0063 - lr: 0.000017
227
+ 2023-10-13 13:16:45,497 DEV : loss 0.09266868978738785 - f1-score (micro avg) 0.8138
228
+ 2023-10-13 13:16:45,526 ----------------------------------------------------------------------------------------------------
229
+ 2023-10-13 13:17:26,201 epoch 10 - iter 77/773 - loss 0.00190450 - time (sec): 40.67 - samples/sec: 301.95 - lr: 0.000015 - momentum: 0.000000
230
+ 2023-10-13 13:18:07,106 epoch 10 - iter 154/773 - loss 0.00359834 - time (sec): 81.58 - samples/sec: 308.59 - lr: 0.000014 - momentum: 0.000000
231
+ 2023-10-13 13:18:46,933 epoch 10 - iter 231/773 - loss 0.00380266 - time (sec): 121.40 - samples/sec: 308.56 - lr: 0.000012 - momentum: 0.000000
232
+ 2023-10-13 13:19:27,460 epoch 10 - iter 308/773 - loss 0.00362008 - time (sec): 161.93 - samples/sec: 307.67 - lr: 0.000010 - momentum: 0.000000
233
+ 2023-10-13 13:20:09,363 epoch 10 - iter 385/773 - loss 0.00387252 - time (sec): 203.83 - samples/sec: 304.86 - lr: 0.000009 - momentum: 0.000000
234
+ 2023-10-13 13:20:50,798 epoch 10 - iter 462/773 - loss 0.00419240 - time (sec): 245.27 - samples/sec: 300.94 - lr: 0.000007 - momentum: 0.000000
235
+ 2023-10-13 13:21:33,881 epoch 10 - iter 539/773 - loss 0.00456516 - time (sec): 288.35 - samples/sec: 298.57 - lr: 0.000005 - momentum: 0.000000
236
+ 2023-10-13 13:22:19,370 epoch 10 - iter 616/773 - loss 0.00451077 - time (sec): 333.84 - samples/sec: 292.87 - lr: 0.000004 - momentum: 0.000000
237
+ 2023-10-13 13:23:04,395 epoch 10 - iter 693/773 - loss 0.00444498 - time (sec): 378.87 - samples/sec: 293.39 - lr: 0.000002 - momentum: 0.000000
238
+ 2023-10-13 13:23:47,529 epoch 10 - iter 770/773 - loss 0.00454130 - time (sec): 422.00 - samples/sec: 293.04 - lr: 0.000000 - momentum: 0.000000
239
+ 2023-10-13 13:23:49,225 ----------------------------------------------------------------------------------------------------
240
+ 2023-10-13 13:23:49,225 EPOCH 10 done: loss 0.0045 - lr: 0.000000
241
+ 2023-10-13 13:24:07,865 DEV : loss 0.09171322733163834 - f1-score (micro avg) 0.8114
242
+ 2023-10-13 13:24:08,877 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-13 13:24:08,879 Loading model from best epoch ...
244
+ 2023-10-13 13:24:13,748 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
245
+ 2023-10-13 13:25:13,121
246
+ Results:
247
+ - F-score (micro) 0.8083
248
+ - F-score (macro) 0.7183
249
+ - Accuracy 0.6988
250
+
251
+ By class:
252
+ precision recall f1-score support
253
+
254
+ LOC 0.8551 0.8668 0.8609 946
255
+ BUILDING 0.5567 0.5838 0.5699 185
256
+ STREET 0.7000 0.7500 0.7241 56
257
+
258
+ micro avg 0.7997 0.8172 0.8083 1187
259
+ macro avg 0.7039 0.7335 0.7183 1187
260
+ weighted avg 0.8012 0.8172 0.8091 1187
261
+
262
+ 2023-10-13 13:25:13,122 ----------------------------------------------------------------------------------------------------