Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbyt5-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
Dutch
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
87d3875
·
1 Parent(s): c16d6d4

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.1697101951.c8b2203b18a8.2408.1 +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 sha256:36a9530a7d5017d7c8392d0e13a7c6599c0a2c2d33a664d66ef7a7bb8e1f74ab
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 sha256:2707c8573728c17ff7208008c095e46170b902c51dfa6842b1340b5e48c8e338
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 09:19:21 0.0002 1.0851 0.1975 0.4971 0.2645 0.3452 0.2284
3
+ 2 09:25:55 0.0001 0.1208 0.1081 0.8095 0.7066 0.7546 0.6151
4
+ 3 09:32:44 0.0001 0.0723 0.0697 0.8636 0.8636 0.8636 0.7712
5
+ 4 09:39:25 0.0001 0.0472 0.0707 0.8717 0.8492 0.8603 0.7647
6
+ 5 09:46:03 0.0001 0.0334 0.0800 0.8959 0.8264 0.8598 0.7626
7
+ 6 09:52:56 0.0001 0.0262 0.0937 0.8733 0.8399 0.8562 0.7598
8
+ 7 10:00:23 0.0001 0.0204 0.1118 0.9060 0.8264 0.8644 0.7700
9
+ 8 10:07:06 0.0000 0.0169 0.1114 0.8881 0.8523 0.8698 0.7827
10
+ 9 10:13:51 0.0000 0.0130 0.1235 0.8917 0.8419 0.8661 0.7762
11
+ 10 10:20:36 0.0000 0.0116 0.1270 0.8961 0.8378 0.8660 0.7746
runs/events.out.tfevents.1697101951.c8b2203b18a8.2408.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df619740211c863b86f4bc750d5d1603f1e8d152ebab70580a9f79ec94644c60
3
+ size 407048
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-12 09:12:31,013 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-12 09:12:31,015 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
+ 2023-10-12 09:12:31,016 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-12 09:12:31,016 MultiCorpus: 5777 train + 722 dev + 723 test sentences
72
+ - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
73
+ 2023-10-12 09:12:31,016 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-12 09:12:31,016 Train: 5777 sentences
75
+ 2023-10-12 09:12:31,016 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-12 09:12:31,016 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-12 09:12:31,016 Training Params:
78
+ 2023-10-12 09:12:31,016 - learning_rate: "0.00016"
79
+ 2023-10-12 09:12:31,016 - mini_batch_size: "8"
80
+ 2023-10-12 09:12:31,016 - max_epochs: "10"
81
+ 2023-10-12 09:12:31,016 - shuffle: "True"
82
+ 2023-10-12 09:12:31,017 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-12 09:12:31,017 Plugins:
84
+ 2023-10-12 09:12:31,017 - TensorboardLogger
85
+ 2023-10-12 09:12:31,017 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-12 09:12:31,017 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-12 09:12:31,017 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-12 09:12:31,017 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-12 09:12:31,017 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-12 09:12:31,017 Computation:
91
+ 2023-10-12 09:12:31,017 - compute on device: cuda:0
92
+ 2023-10-12 09:12:31,017 - embedding storage: none
93
+ 2023-10-12 09:12:31,017 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-12 09:12:31,017 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3"
95
+ 2023-10-12 09:12:31,017 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-12 09:12:31,017 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-12 09:12:31,018 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-12 09:13:10,651 epoch 1 - iter 72/723 - loss 2.54068274 - time (sec): 39.63 - samples/sec: 435.41 - lr: 0.000016 - momentum: 0.000000
99
+ 2023-10-12 09:13:49,781 epoch 1 - iter 144/723 - loss 2.48896229 - time (sec): 78.76 - samples/sec: 437.46 - lr: 0.000032 - momentum: 0.000000
100
+ 2023-10-12 09:14:30,289 epoch 1 - iter 216/723 - loss 2.31703108 - time (sec): 119.27 - samples/sec: 439.33 - lr: 0.000048 - momentum: 0.000000
101
+ 2023-10-12 09:15:08,873 epoch 1 - iter 288/723 - loss 2.11125208 - time (sec): 157.85 - samples/sec: 438.77 - lr: 0.000064 - momentum: 0.000000
102
+ 2023-10-12 09:15:47,731 epoch 1 - iter 360/723 - loss 1.87865228 - time (sec): 196.71 - samples/sec: 443.76 - lr: 0.000079 - momentum: 0.000000
103
+ 2023-10-12 09:16:25,957 epoch 1 - iter 432/723 - loss 1.66306784 - time (sec): 234.94 - samples/sec: 444.23 - lr: 0.000095 - momentum: 0.000000
104
+ 2023-10-12 09:17:05,925 epoch 1 - iter 504/723 - loss 1.45592361 - time (sec): 274.91 - samples/sec: 449.78 - lr: 0.000111 - momentum: 0.000000
105
+ 2023-10-12 09:17:44,352 epoch 1 - iter 576/723 - loss 1.31694027 - time (sec): 313.33 - samples/sec: 446.12 - lr: 0.000127 - momentum: 0.000000
106
+ 2023-10-12 09:18:22,944 epoch 1 - iter 648/723 - loss 1.19195801 - time (sec): 351.92 - samples/sec: 447.36 - lr: 0.000143 - momentum: 0.000000
107
+ 2023-10-12 09:19:01,438 epoch 1 - iter 720/723 - loss 1.08849052 - time (sec): 390.42 - samples/sec: 449.67 - lr: 0.000159 - momentum: 0.000000
108
+ 2023-10-12 09:19:02,618 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-12 09:19:02,618 EPOCH 1 done: loss 1.0851 - lr: 0.000159
110
+ 2023-10-12 09:19:21,607 DEV : loss 0.19754847884178162 - f1-score (micro avg) 0.3452
111
+ 2023-10-12 09:19:21,636 saving best model
112
+ 2023-10-12 09:19:22,482 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-12 09:19:59,860 epoch 2 - iter 72/723 - loss 0.15960703 - time (sec): 37.38 - samples/sec: 474.56 - lr: 0.000158 - momentum: 0.000000
114
+ 2023-10-12 09:20:37,795 epoch 2 - iter 144/723 - loss 0.15205605 - time (sec): 75.31 - samples/sec: 476.10 - lr: 0.000156 - momentum: 0.000000
115
+ 2023-10-12 09:21:15,356 epoch 2 - iter 216/723 - loss 0.14254311 - time (sec): 112.87 - samples/sec: 476.58 - lr: 0.000155 - momentum: 0.000000
116
+ 2023-10-12 09:21:53,960 epoch 2 - iter 288/723 - loss 0.13485595 - time (sec): 151.48 - samples/sec: 482.39 - lr: 0.000153 - momentum: 0.000000
117
+ 2023-10-12 09:22:30,418 epoch 2 - iter 360/723 - loss 0.12904402 - time (sec): 187.93 - samples/sec: 476.01 - lr: 0.000151 - momentum: 0.000000
118
+ 2023-10-12 09:23:07,023 epoch 2 - iter 432/723 - loss 0.12930298 - time (sec): 224.54 - samples/sec: 472.38 - lr: 0.000149 - momentum: 0.000000
119
+ 2023-10-12 09:23:43,890 epoch 2 - iter 504/723 - loss 0.12740399 - time (sec): 261.41 - samples/sec: 472.37 - lr: 0.000148 - momentum: 0.000000
120
+ 2023-10-12 09:24:20,296 epoch 2 - iter 576/723 - loss 0.12436397 - time (sec): 297.81 - samples/sec: 471.01 - lr: 0.000146 - momentum: 0.000000
121
+ 2023-10-12 09:24:58,056 epoch 2 - iter 648/723 - loss 0.12313707 - time (sec): 335.57 - samples/sec: 471.28 - lr: 0.000144 - momentum: 0.000000
122
+ 2023-10-12 09:25:34,923 epoch 2 - iter 720/723 - loss 0.12108946 - time (sec): 372.44 - samples/sec: 471.44 - lr: 0.000142 - momentum: 0.000000
123
+ 2023-10-12 09:25:36,090 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-12 09:25:36,091 EPOCH 2 done: loss 0.1208 - lr: 0.000142
125
+ 2023-10-12 09:25:55,610 DEV : loss 0.10809068381786346 - f1-score (micro avg) 0.7546
126
+ 2023-10-12 09:25:55,639 saving best model
127
+ 2023-10-12 09:25:58,153 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-12 09:26:35,794 epoch 3 - iter 72/723 - loss 0.09524519 - time (sec): 37.64 - samples/sec: 465.61 - lr: 0.000140 - momentum: 0.000000
129
+ 2023-10-12 09:27:12,312 epoch 3 - iter 144/723 - loss 0.08802830 - time (sec): 74.15 - samples/sec: 463.82 - lr: 0.000139 - momentum: 0.000000
130
+ 2023-10-12 09:27:50,862 epoch 3 - iter 216/723 - loss 0.07792497 - time (sec): 112.70 - samples/sec: 462.77 - lr: 0.000137 - momentum: 0.000000
131
+ 2023-10-12 09:28:27,774 epoch 3 - iter 288/723 - loss 0.08033417 - time (sec): 149.62 - samples/sec: 455.30 - lr: 0.000135 - momentum: 0.000000
132
+ 2023-10-12 09:29:07,289 epoch 3 - iter 360/723 - loss 0.07718260 - time (sec): 189.13 - samples/sec: 457.93 - lr: 0.000133 - momentum: 0.000000
133
+ 2023-10-12 09:29:45,291 epoch 3 - iter 432/723 - loss 0.07571314 - time (sec): 227.13 - samples/sec: 458.11 - lr: 0.000132 - momentum: 0.000000
134
+ 2023-10-12 09:30:25,618 epoch 3 - iter 504/723 - loss 0.07655001 - time (sec): 267.46 - samples/sec: 457.54 - lr: 0.000130 - momentum: 0.000000
135
+ 2023-10-12 09:31:03,813 epoch 3 - iter 576/723 - loss 0.07485966 - time (sec): 305.66 - samples/sec: 455.69 - lr: 0.000128 - momentum: 0.000000
136
+ 2023-10-12 09:31:42,269 epoch 3 - iter 648/723 - loss 0.07373807 - time (sec): 344.11 - samples/sec: 456.31 - lr: 0.000126 - momentum: 0.000000
137
+ 2023-10-12 09:32:22,158 epoch 3 - iter 720/723 - loss 0.07223758 - time (sec): 384.00 - samples/sec: 457.61 - lr: 0.000125 - momentum: 0.000000
138
+ 2023-10-12 09:32:23,294 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-12 09:32:23,294 EPOCH 3 done: loss 0.0723 - lr: 0.000125
140
+ 2023-10-12 09:32:44,912 DEV : loss 0.06969450414180756 - f1-score (micro avg) 0.8636
141
+ 2023-10-12 09:32:44,945 saving best model
142
+ 2023-10-12 09:32:47,473 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-12 09:33:25,242 epoch 4 - iter 72/723 - loss 0.04406291 - time (sec): 37.76 - samples/sec: 481.38 - lr: 0.000123 - momentum: 0.000000
144
+ 2023-10-12 09:34:02,574 epoch 4 - iter 144/723 - loss 0.04218829 - time (sec): 75.10 - samples/sec: 472.91 - lr: 0.000121 - momentum: 0.000000
145
+ 2023-10-12 09:34:41,277 epoch 4 - iter 216/723 - loss 0.05044030 - time (sec): 113.80 - samples/sec: 467.34 - lr: 0.000119 - momentum: 0.000000
146
+ 2023-10-12 09:35:20,827 epoch 4 - iter 288/723 - loss 0.04888086 - time (sec): 153.35 - samples/sec: 465.31 - lr: 0.000117 - momentum: 0.000000
147
+ 2023-10-12 09:35:58,524 epoch 4 - iter 360/723 - loss 0.04697769 - time (sec): 191.05 - samples/sec: 460.95 - lr: 0.000116 - momentum: 0.000000
148
+ 2023-10-12 09:36:36,190 epoch 4 - iter 432/723 - loss 0.04817827 - time (sec): 228.71 - samples/sec: 461.05 - lr: 0.000114 - momentum: 0.000000
149
+ 2023-10-12 09:37:14,593 epoch 4 - iter 504/723 - loss 0.04752541 - time (sec): 267.12 - samples/sec: 464.52 - lr: 0.000112 - momentum: 0.000000
150
+ 2023-10-12 09:37:50,742 epoch 4 - iter 576/723 - loss 0.04722134 - time (sec): 303.26 - samples/sec: 465.08 - lr: 0.000110 - momentum: 0.000000
151
+ 2023-10-12 09:38:27,389 epoch 4 - iter 648/723 - loss 0.04692030 - time (sec): 339.91 - samples/sec: 466.31 - lr: 0.000109 - momentum: 0.000000
152
+ 2023-10-12 09:39:03,653 epoch 4 - iter 720/723 - loss 0.04727199 - time (sec): 376.18 - samples/sec: 466.91 - lr: 0.000107 - momentum: 0.000000
153
+ 2023-10-12 09:39:04,778 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-12 09:39:04,779 EPOCH 4 done: loss 0.0472 - lr: 0.000107
155
+ 2023-10-12 09:39:25,632 DEV : loss 0.07068605720996857 - f1-score (micro avg) 0.8603
156
+ 2023-10-12 09:39:25,665 ----------------------------------------------------------------------------------------------------
157
+ 2023-10-12 09:40:02,917 epoch 5 - iter 72/723 - loss 0.02748124 - time (sec): 37.25 - samples/sec: 455.60 - lr: 0.000105 - momentum: 0.000000
158
+ 2023-10-12 09:40:40,248 epoch 5 - iter 144/723 - loss 0.03666304 - time (sec): 74.58 - samples/sec: 457.23 - lr: 0.000103 - momentum: 0.000000
159
+ 2023-10-12 09:41:17,534 epoch 5 - iter 216/723 - loss 0.03231394 - time (sec): 111.87 - samples/sec: 467.35 - lr: 0.000101 - momentum: 0.000000
160
+ 2023-10-12 09:41:54,565 epoch 5 - iter 288/723 - loss 0.03031252 - time (sec): 148.90 - samples/sec: 471.60 - lr: 0.000100 - momentum: 0.000000
161
+ 2023-10-12 09:42:31,819 epoch 5 - iter 360/723 - loss 0.03210903 - time (sec): 186.15 - samples/sec: 476.16 - lr: 0.000098 - momentum: 0.000000
162
+ 2023-10-12 09:43:08,920 epoch 5 - iter 432/723 - loss 0.03173792 - time (sec): 223.25 - samples/sec: 474.45 - lr: 0.000096 - momentum: 0.000000
163
+ 2023-10-12 09:43:46,282 epoch 5 - iter 504/723 - loss 0.03240803 - time (sec): 260.61 - samples/sec: 470.41 - lr: 0.000094 - momentum: 0.000000
164
+ 2023-10-12 09:44:24,185 epoch 5 - iter 576/723 - loss 0.03345992 - time (sec): 298.52 - samples/sec: 469.45 - lr: 0.000093 - momentum: 0.000000
165
+ 2023-10-12 09:45:01,744 epoch 5 - iter 648/723 - loss 0.03356952 - time (sec): 336.08 - samples/sec: 467.27 - lr: 0.000091 - momentum: 0.000000
166
+ 2023-10-12 09:45:40,826 epoch 5 - iter 720/723 - loss 0.03345778 - time (sec): 375.16 - samples/sec: 468.24 - lr: 0.000089 - momentum: 0.000000
167
+ 2023-10-12 09:45:42,036 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-12 09:45:42,036 EPOCH 5 done: loss 0.0334 - lr: 0.000089
169
+ 2023-10-12 09:46:03,200 DEV : loss 0.08003567904233932 - f1-score (micro avg) 0.8598
170
+ 2023-10-12 09:46:03,230 ----------------------------------------------------------------------------------------------------
171
+ 2023-10-12 09:46:41,057 epoch 6 - iter 72/723 - loss 0.01469662 - time (sec): 37.83 - samples/sec: 476.88 - lr: 0.000087 - momentum: 0.000000
172
+ 2023-10-12 09:47:17,385 epoch 6 - iter 144/723 - loss 0.02012899 - time (sec): 74.15 - samples/sec: 458.62 - lr: 0.000085 - momentum: 0.000000
173
+ 2023-10-12 09:47:54,113 epoch 6 - iter 216/723 - loss 0.02320686 - time (sec): 110.88 - samples/sec: 462.61 - lr: 0.000084 - momentum: 0.000000
174
+ 2023-10-12 09:48:30,521 epoch 6 - iter 288/723 - loss 0.02223906 - time (sec): 147.29 - samples/sec: 460.08 - lr: 0.000082 - momentum: 0.000000
175
+ 2023-10-12 09:49:07,665 epoch 6 - iter 360/723 - loss 0.02179582 - time (sec): 184.43 - samples/sec: 466.52 - lr: 0.000080 - momentum: 0.000000
176
+ 2023-10-12 09:49:44,594 epoch 6 - iter 432/723 - loss 0.02302519 - time (sec): 221.36 - samples/sec: 467.19 - lr: 0.000078 - momentum: 0.000000
177
+ 2023-10-12 09:50:23,505 epoch 6 - iter 504/723 - loss 0.02173843 - time (sec): 260.27 - samples/sec: 464.70 - lr: 0.000077 - momentum: 0.000000
178
+ 2023-10-12 09:51:04,453 epoch 6 - iter 576/723 - loss 0.02233847 - time (sec): 301.22 - samples/sec: 460.32 - lr: 0.000075 - momentum: 0.000000
179
+ 2023-10-12 09:51:48,517 epoch 6 - iter 648/723 - loss 0.02374348 - time (sec): 345.28 - samples/sec: 454.70 - lr: 0.000073 - momentum: 0.000000
180
+ 2023-10-12 09:52:32,950 epoch 6 - iter 720/723 - loss 0.02595781 - time (sec): 389.72 - samples/sec: 450.50 - lr: 0.000071 - momentum: 0.000000
181
+ 2023-10-12 09:52:34,302 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-12 09:52:34,303 EPOCH 6 done: loss 0.0262 - lr: 0.000071
183
+ 2023-10-12 09:52:56,496 DEV : loss 0.09371042996644974 - f1-score (micro avg) 0.8562
184
+ 2023-10-12 09:52:56,538 ----------------------------------------------------------------------------------------------------
185
+ 2023-10-12 09:53:40,163 epoch 7 - iter 72/723 - loss 0.02096644 - time (sec): 43.62 - samples/sec: 427.85 - lr: 0.000069 - momentum: 0.000000
186
+ 2023-10-12 09:54:20,242 epoch 7 - iter 144/723 - loss 0.01613838 - time (sec): 83.70 - samples/sec: 417.02 - lr: 0.000068 - momentum: 0.000000
187
+ 2023-10-12 09:55:01,918 epoch 7 - iter 216/723 - loss 0.02027854 - time (sec): 125.38 - samples/sec: 423.27 - lr: 0.000066 - momentum: 0.000000
188
+ 2023-10-12 09:55:46,254 epoch 7 - iter 288/723 - loss 0.02019598 - time (sec): 169.71 - samples/sec: 421.59 - lr: 0.000064 - momentum: 0.000000
189
+ 2023-10-12 09:56:30,235 epoch 7 - iter 360/723 - loss 0.02145432 - time (sec): 213.69 - samples/sec: 418.12 - lr: 0.000062 - momentum: 0.000000
190
+ 2023-10-12 09:57:11,668 epoch 7 - iter 432/723 - loss 0.02070345 - time (sec): 255.13 - samples/sec: 414.64 - lr: 0.000061 - momentum: 0.000000
191
+ 2023-10-12 09:57:54,874 epoch 7 - iter 504/723 - loss 0.02079100 - time (sec): 298.33 - samples/sec: 413.30 - lr: 0.000059 - momentum: 0.000000
192
+ 2023-10-12 09:58:39,294 epoch 7 - iter 576/723 - loss 0.02219735 - time (sec): 342.75 - samples/sec: 410.99 - lr: 0.000057 - momentum: 0.000000
193
+ 2023-10-12 09:59:21,613 epoch 7 - iter 648/723 - loss 0.02096818 - time (sec): 385.07 - samples/sec: 410.00 - lr: 0.000055 - momentum: 0.000000
194
+ 2023-10-12 10:00:01,414 epoch 7 - iter 720/723 - loss 0.02042740 - time (sec): 424.87 - samples/sec: 413.43 - lr: 0.000053 - momentum: 0.000000
195
+ 2023-10-12 10:00:02,702 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-12 10:00:02,702 EPOCH 7 done: loss 0.0204 - lr: 0.000053
197
+ 2023-10-12 10:00:23,279 DEV : loss 0.11184699088335037 - f1-score (micro avg) 0.8644
198
+ 2023-10-12 10:00:23,310 saving best model
199
+ 2023-10-12 10:00:26,348 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-12 10:01:04,206 epoch 8 - iter 72/723 - loss 0.00941353 - time (sec): 37.85 - samples/sec: 439.88 - lr: 0.000052 - momentum: 0.000000
201
+ 2023-10-12 10:01:42,366 epoch 8 - iter 144/723 - loss 0.01683680 - time (sec): 76.01 - samples/sec: 453.17 - lr: 0.000050 - momentum: 0.000000
202
+ 2023-10-12 10:02:20,195 epoch 8 - iter 216/723 - loss 0.01621885 - time (sec): 113.84 - samples/sec: 455.09 - lr: 0.000048 - momentum: 0.000000
203
+ 2023-10-12 10:02:58,389 epoch 8 - iter 288/723 - loss 0.01774953 - time (sec): 152.04 - samples/sec: 459.51 - lr: 0.000046 - momentum: 0.000000
204
+ 2023-10-12 10:03:36,210 epoch 8 - iter 360/723 - loss 0.01673376 - time (sec): 189.86 - samples/sec: 464.11 - lr: 0.000045 - momentum: 0.000000
205
+ 2023-10-12 10:04:13,871 epoch 8 - iter 432/723 - loss 0.01870128 - time (sec): 227.52 - samples/sec: 464.71 - lr: 0.000043 - momentum: 0.000000
206
+ 2023-10-12 10:04:52,284 epoch 8 - iter 504/723 - loss 0.01758556 - time (sec): 265.93 - samples/sec: 466.61 - lr: 0.000041 - momentum: 0.000000
207
+ 2023-10-12 10:05:30,404 epoch 8 - iter 576/723 - loss 0.01710052 - time (sec): 304.05 - samples/sec: 467.78 - lr: 0.000039 - momentum: 0.000000
208
+ 2023-10-12 10:06:08,098 epoch 8 - iter 648/723 - loss 0.01689011 - time (sec): 341.75 - samples/sec: 466.57 - lr: 0.000037 - momentum: 0.000000
209
+ 2023-10-12 10:06:45,084 epoch 8 - iter 720/723 - loss 0.01690370 - time (sec): 378.73 - samples/sec: 463.95 - lr: 0.000036 - momentum: 0.000000
210
+ 2023-10-12 10:06:46,164 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-12 10:06:46,164 EPOCH 8 done: loss 0.0169 - lr: 0.000036
212
+ 2023-10-12 10:07:06,450 DEV : loss 0.11143004894256592 - f1-score (micro avg) 0.8698
213
+ 2023-10-12 10:07:06,482 saving best model
214
+ 2023-10-12 10:07:09,971 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-12 10:07:46,794 epoch 9 - iter 72/723 - loss 0.01791029 - time (sec): 36.82 - samples/sec: 467.39 - lr: 0.000034 - momentum: 0.000000
216
+ 2023-10-12 10:08:23,570 epoch 9 - iter 144/723 - loss 0.01713879 - time (sec): 73.59 - samples/sec: 458.06 - lr: 0.000032 - momentum: 0.000000
217
+ 2023-10-12 10:09:02,166 epoch 9 - iter 216/723 - loss 0.01306516 - time (sec): 112.19 - samples/sec: 469.03 - lr: 0.000030 - momentum: 0.000000
218
+ 2023-10-12 10:09:39,707 epoch 9 - iter 288/723 - loss 0.01216452 - time (sec): 149.73 - samples/sec: 467.63 - lr: 0.000028 - momentum: 0.000000
219
+ 2023-10-12 10:10:17,712 epoch 9 - iter 360/723 - loss 0.01176075 - time (sec): 187.74 - samples/sec: 464.93 - lr: 0.000027 - momentum: 0.000000
220
+ 2023-10-12 10:10:54,836 epoch 9 - iter 432/723 - loss 0.01188469 - time (sec): 224.86 - samples/sec: 466.23 - lr: 0.000025 - momentum: 0.000000
221
+ 2023-10-12 10:11:34,174 epoch 9 - iter 504/723 - loss 0.01118095 - time (sec): 264.20 - samples/sec: 464.20 - lr: 0.000023 - momentum: 0.000000
222
+ 2023-10-12 10:12:12,050 epoch 9 - iter 576/723 - loss 0.01081907 - time (sec): 302.07 - samples/sec: 462.46 - lr: 0.000021 - momentum: 0.000000
223
+ 2023-10-12 10:12:50,928 epoch 9 - iter 648/723 - loss 0.01218741 - time (sec): 340.95 - samples/sec: 462.49 - lr: 0.000020 - momentum: 0.000000
224
+ 2023-10-12 10:13:29,451 epoch 9 - iter 720/723 - loss 0.01308210 - time (sec): 379.48 - samples/sec: 462.30 - lr: 0.000018 - momentum: 0.000000
225
+ 2023-10-12 10:13:30,853 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-12 10:13:30,854 EPOCH 9 done: loss 0.0130 - lr: 0.000018
227
+ 2023-10-12 10:13:51,633 DEV : loss 0.12346570193767548 - f1-score (micro avg) 0.8661
228
+ 2023-10-12 10:13:51,663 ----------------------------------------------------------------------------------------------------
229
+ 2023-10-12 10:14:29,981 epoch 10 - iter 72/723 - loss 0.01183784 - time (sec): 38.32 - samples/sec: 458.40 - lr: 0.000016 - momentum: 0.000000
230
+ 2023-10-12 10:15:08,278 epoch 10 - iter 144/723 - loss 0.00969580 - time (sec): 76.61 - samples/sec: 454.26 - lr: 0.000014 - momentum: 0.000000
231
+ 2023-10-12 10:15:46,823 epoch 10 - iter 216/723 - loss 0.00857501 - time (sec): 115.16 - samples/sec: 460.10 - lr: 0.000012 - momentum: 0.000000
232
+ 2023-10-12 10:16:24,331 epoch 10 - iter 288/723 - loss 0.00966868 - time (sec): 152.67 - samples/sec: 456.38 - lr: 0.000011 - momentum: 0.000000
233
+ 2023-10-12 10:17:01,431 epoch 10 - iter 360/723 - loss 0.00912886 - time (sec): 189.77 - samples/sec: 453.30 - lr: 0.000009 - momentum: 0.000000
234
+ 2023-10-12 10:17:40,125 epoch 10 - iter 432/723 - loss 0.00936631 - time (sec): 228.46 - samples/sec: 455.61 - lr: 0.000007 - momentum: 0.000000
235
+ 2023-10-12 10:18:19,178 epoch 10 - iter 504/723 - loss 0.00919697 - time (sec): 267.51 - samples/sec: 457.77 - lr: 0.000005 - momentum: 0.000000
236
+ 2023-10-12 10:18:57,221 epoch 10 - iter 576/723 - loss 0.00981610 - time (sec): 305.56 - samples/sec: 457.01 - lr: 0.000004 - momentum: 0.000000
237
+ 2023-10-12 10:19:35,541 epoch 10 - iter 648/723 - loss 0.01029312 - time (sec): 343.88 - samples/sec: 457.54 - lr: 0.000002 - momentum: 0.000000
238
+ 2023-10-12 10:20:14,421 epoch 10 - iter 720/723 - loss 0.01158386 - time (sec): 382.76 - samples/sec: 459.37 - lr: 0.000000 - momentum: 0.000000
239
+ 2023-10-12 10:20:15,483 ----------------------------------------------------------------------------------------------------
240
+ 2023-10-12 10:20:15,484 EPOCH 10 done: loss 0.0116 - lr: 0.000000
241
+ 2023-10-12 10:20:35,988 DEV : loss 0.12695999443531036 - f1-score (micro avg) 0.866
242
+ 2023-10-12 10:20:36,903 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-12 10:20:36,905 Loading model from best epoch ...
244
+ 2023-10-12 10:20:41,160 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG
245
+ 2023-10-12 10:21:01,016
246
+ Results:
247
+ - F-score (micro) 0.8354
248
+ - F-score (macro) 0.7615
249
+ - Accuracy 0.7266
250
+
251
+ By class:
252
+ precision recall f1-score support
253
+
254
+ PER 0.8275 0.8755 0.8508 482
255
+ LOC 0.9082 0.8210 0.8624 458
256
+ ORG 0.5385 0.6087 0.5714 69
257
+
258
+ micro avg 0.8383 0.8325 0.8354 1009
259
+ macro avg 0.7580 0.7684 0.7615 1009
260
+ weighted avg 0.8443 0.8325 0.8370 1009
261
+
262
+ 2023-10-12 10:21:01,016 ----------------------------------------------------------------------------------------------------