Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbyt5-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
4971493
•
1 Parent(s): 885f2b4

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697317753.d3463e005216.2433.5 +3 -0
  5. test.tsv +0 -0
  6. training.log +265 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac35f669199bb854af06036f74dba23077049dec3b20c5e8ecc7e3dd91b55390
3
+ size 870841135
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
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 21:12:15 0.0002 1.8272 0.4691 0.0000 0.0000 0.0000 0.0000
3
+ 2 21:15:21 0.0001 0.3593 0.2348 0.5315 0.4621 0.4944 0.3460
4
+ 3 21:18:29 0.0001 0.1959 0.1630 0.6512 0.7154 0.6818 0.5443
5
+ 4 21:21:41 0.0001 0.1079 0.1377 0.7217 0.7725 0.7462 0.6187
6
+ 5 21:24:55 0.0001 0.0672 0.1688 0.7465 0.7459 0.7462 0.6131
7
+ 6 21:28:03 0.0001 0.0434 0.1666 0.7630 0.7701 0.7665 0.6392
8
+ 7 21:31:15 0.0001 0.0303 0.1883 0.7777 0.7686 0.7731 0.6459
9
+ 8 21:34:26 0.0000 0.0223 0.1866 0.7685 0.7787 0.7736 0.6480
10
+ 9 21:37:34 0.0000 0.0176 0.1906 0.7453 0.7826 0.7635 0.6356
11
+ 10 21:40:43 0.0000 0.0142 0.1945 0.7618 0.7803 0.7710 0.6435
runs/events.out.tfevents.1697317753.d3463e005216.2433.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68e95fc5ee6543df24127899c88c64f5a3c22c17285bbff53fb56345b619b6f1
3
+ size 253592
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 21:09:13,480 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 21:09:13,481 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=21, bias=True)
68
+ (loss_function): CrossEntropyLoss()
69
+ )"
70
+ 2023-10-14 21:09:13,481 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-14 21:09:13,481 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
72
+ - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
73
+ 2023-10-14 21:09:13,481 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-14 21:09:13,481 Train: 3575 sentences
75
+ 2023-10-14 21:09:13,481 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-14 21:09:13,481 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-14 21:09:13,481 Training Params:
78
+ 2023-10-14 21:09:13,481 - learning_rate: "0.00016"
79
+ 2023-10-14 21:09:13,481 - mini_batch_size: "8"
80
+ 2023-10-14 21:09:13,481 - max_epochs: "10"
81
+ 2023-10-14 21:09:13,481 - shuffle: "True"
82
+ 2023-10-14 21:09:13,481 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-14 21:09:13,481 Plugins:
84
+ 2023-10-14 21:09:13,481 - TensorboardLogger
85
+ 2023-10-14 21:09:13,481 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-14 21:09:13,481 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 21:09:13,481 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-14 21:09:13,481 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-14 21:09:13,482 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-14 21:09:13,482 Computation:
91
+ 2023-10-14 21:09:13,482 - compute on device: cuda:0
92
+ 2023-10-14 21:09:13,482 - embedding storage: none
93
+ 2023-10-14 21:09:13,482 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-14 21:09:13,482 Model training base path: "hmbench-hipe2020/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2"
95
+ 2023-10-14 21:09:13,482 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-14 21:09:13,482 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-14 21:09:13,482 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-14 21:09:28,751 epoch 1 - iter 44/447 - loss 3.05259660 - time (sec): 15.27 - samples/sec: 513.49 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-14 21:09:46,378 epoch 1 - iter 88/447 - loss 3.03130581 - time (sec): 32.90 - samples/sec: 533.78 - lr: 0.000031 - momentum: 0.000000
100
+ 2023-10-14 21:10:02,233 epoch 1 - iter 132/447 - loss 2.97859953 - time (sec): 48.75 - samples/sec: 536.12 - lr: 0.000047 - momentum: 0.000000
101
+ 2023-10-14 21:10:18,645 epoch 1 - iter 176/447 - loss 2.82805827 - time (sec): 65.16 - samples/sec: 545.79 - lr: 0.000063 - momentum: 0.000000
102
+ 2023-10-14 21:10:34,847 epoch 1 - iter 220/447 - loss 2.65793730 - time (sec): 81.36 - samples/sec: 549.19 - lr: 0.000078 - momentum: 0.000000
103
+ 2023-10-14 21:10:50,453 epoch 1 - iter 264/447 - loss 2.48457680 - time (sec): 96.97 - samples/sec: 548.84 - lr: 0.000094 - momentum: 0.000000
104
+ 2023-10-14 21:11:05,887 epoch 1 - iter 308/447 - loss 2.30687635 - time (sec): 112.40 - samples/sec: 547.47 - lr: 0.000110 - momentum: 0.000000
105
+ 2023-10-14 21:11:21,021 epoch 1 - iter 352/447 - loss 2.13426617 - time (sec): 127.54 - samples/sec: 545.68 - lr: 0.000126 - momentum: 0.000000
106
+ 2023-10-14 21:11:36,177 epoch 1 - iter 396/447 - loss 1.97499593 - time (sec): 142.69 - samples/sec: 542.47 - lr: 0.000141 - momentum: 0.000000
107
+ 2023-10-14 21:11:51,265 epoch 1 - iter 440/447 - loss 1.84836303 - time (sec): 157.78 - samples/sec: 539.83 - lr: 0.000157 - momentum: 0.000000
108
+ 2023-10-14 21:11:53,680 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-14 21:11:53,681 EPOCH 1 done: loss 1.8272 - lr: 0.000157
110
+ 2023-10-14 21:12:15,957 DEV : loss 0.4690864682197571 - f1-score (micro avg) 0.0
111
+ 2023-10-14 21:12:15,982 ----------------------------------------------------------------------------------------------------
112
+ 2023-10-14 21:12:31,700 epoch 2 - iter 44/447 - loss 0.47142257 - time (sec): 15.72 - samples/sec: 549.94 - lr: 0.000158 - momentum: 0.000000
113
+ 2023-10-14 21:12:47,381 epoch 2 - iter 88/447 - loss 0.50496765 - time (sec): 31.40 - samples/sec: 556.84 - lr: 0.000157 - momentum: 0.000000
114
+ 2023-10-14 21:13:02,839 epoch 2 - iter 132/447 - loss 0.48438537 - time (sec): 46.85 - samples/sec: 547.37 - lr: 0.000155 - momentum: 0.000000
115
+ 2023-10-14 21:13:18,182 epoch 2 - iter 176/447 - loss 0.45854379 - time (sec): 62.20 - samples/sec: 544.15 - lr: 0.000153 - momentum: 0.000000
116
+ 2023-10-14 21:13:33,385 epoch 2 - iter 220/447 - loss 0.43111071 - time (sec): 77.40 - samples/sec: 541.65 - lr: 0.000151 - momentum: 0.000000
117
+ 2023-10-14 21:13:48,337 epoch 2 - iter 264/447 - loss 0.41504902 - time (sec): 92.35 - samples/sec: 537.11 - lr: 0.000150 - momentum: 0.000000
118
+ 2023-10-14 21:14:03,857 epoch 2 - iter 308/447 - loss 0.40207151 - time (sec): 107.87 - samples/sec: 538.55 - lr: 0.000148 - momentum: 0.000000
119
+ 2023-10-14 21:14:19,642 epoch 2 - iter 352/447 - loss 0.38384323 - time (sec): 123.66 - samples/sec: 540.62 - lr: 0.000146 - momentum: 0.000000
120
+ 2023-10-14 21:14:35,295 epoch 2 - iter 396/447 - loss 0.37175563 - time (sec): 139.31 - samples/sec: 541.82 - lr: 0.000144 - momentum: 0.000000
121
+ 2023-10-14 21:14:52,726 epoch 2 - iter 440/447 - loss 0.36035038 - time (sec): 156.74 - samples/sec: 542.41 - lr: 0.000143 - momentum: 0.000000
122
+ 2023-10-14 21:14:55,351 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-14 21:14:55,351 EPOCH 2 done: loss 0.3593 - lr: 0.000143
124
+ 2023-10-14 21:15:21,373 DEV : loss 0.2348013073205948 - f1-score (micro avg) 0.4944
125
+ 2023-10-14 21:15:21,399 saving best model
126
+ 2023-10-14 21:15:21,992 ----------------------------------------------------------------------------------------------------
127
+ 2023-10-14 21:15:37,229 epoch 3 - iter 44/447 - loss 0.23794894 - time (sec): 15.24 - samples/sec: 504.20 - lr: 0.000141 - momentum: 0.000000
128
+ 2023-10-14 21:15:53,173 epoch 3 - iter 88/447 - loss 0.22177517 - time (sec): 31.18 - samples/sec: 521.41 - lr: 0.000139 - momentum: 0.000000
129
+ 2023-10-14 21:16:09,291 epoch 3 - iter 132/447 - loss 0.21912354 - time (sec): 47.30 - samples/sec: 524.29 - lr: 0.000137 - momentum: 0.000000
130
+ 2023-10-14 21:16:24,523 epoch 3 - iter 176/447 - loss 0.21408920 - time (sec): 62.53 - samples/sec: 520.39 - lr: 0.000135 - momentum: 0.000000
131
+ 2023-10-14 21:16:42,440 epoch 3 - iter 220/447 - loss 0.21302112 - time (sec): 80.45 - samples/sec: 530.27 - lr: 0.000134 - momentum: 0.000000
132
+ 2023-10-14 21:16:58,045 epoch 3 - iter 264/447 - loss 0.21166677 - time (sec): 96.05 - samples/sec: 533.24 - lr: 0.000132 - momentum: 0.000000
133
+ 2023-10-14 21:17:13,455 epoch 3 - iter 308/447 - loss 0.20964952 - time (sec): 111.46 - samples/sec: 533.42 - lr: 0.000130 - momentum: 0.000000
134
+ 2023-10-14 21:17:29,406 epoch 3 - iter 352/447 - loss 0.20346960 - time (sec): 127.41 - samples/sec: 538.12 - lr: 0.000128 - momentum: 0.000000
135
+ 2023-10-14 21:17:44,559 epoch 3 - iter 396/447 - loss 0.20088993 - time (sec): 142.57 - samples/sec: 535.56 - lr: 0.000127 - momentum: 0.000000
136
+ 2023-10-14 21:18:00,372 epoch 3 - iter 440/447 - loss 0.19704251 - time (sec): 158.38 - samples/sec: 537.79 - lr: 0.000125 - momentum: 0.000000
137
+ 2023-10-14 21:18:02,860 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-14 21:18:02,860 EPOCH 3 done: loss 0.1959 - lr: 0.000125
139
+ 2023-10-14 21:18:29,261 DEV : loss 0.16303199529647827 - f1-score (micro avg) 0.6818
140
+ 2023-10-14 21:18:29,300 saving best model
141
+ 2023-10-14 21:18:34,235 ----------------------------------------------------------------------------------------------------
142
+ 2023-10-14 21:18:49,598 epoch 4 - iter 44/447 - loss 0.11035298 - time (sec): 15.36 - samples/sec: 531.97 - lr: 0.000123 - momentum: 0.000000
143
+ 2023-10-14 21:19:05,243 epoch 4 - iter 88/447 - loss 0.10572054 - time (sec): 31.01 - samples/sec: 540.01 - lr: 0.000121 - momentum: 0.000000
144
+ 2023-10-14 21:19:20,470 epoch 4 - iter 132/447 - loss 0.11181937 - time (sec): 46.23 - samples/sec: 531.53 - lr: 0.000119 - momentum: 0.000000
145
+ 2023-10-14 21:19:35,766 epoch 4 - iter 176/447 - loss 0.11352787 - time (sec): 61.53 - samples/sec: 533.58 - lr: 0.000118 - momentum: 0.000000
146
+ 2023-10-14 21:19:53,677 epoch 4 - iter 220/447 - loss 0.11538210 - time (sec): 79.44 - samples/sec: 532.51 - lr: 0.000116 - momentum: 0.000000
147
+ 2023-10-14 21:20:09,762 epoch 4 - iter 264/447 - loss 0.11371066 - time (sec): 95.53 - samples/sec: 533.46 - lr: 0.000114 - momentum: 0.000000
148
+ 2023-10-14 21:20:25,452 epoch 4 - iter 308/447 - loss 0.11258637 - time (sec): 111.22 - samples/sec: 534.23 - lr: 0.000112 - momentum: 0.000000
149
+ 2023-10-14 21:20:41,356 epoch 4 - iter 352/447 - loss 0.11068622 - time (sec): 127.12 - samples/sec: 537.00 - lr: 0.000111 - momentum: 0.000000
150
+ 2023-10-14 21:20:57,025 epoch 4 - iter 396/447 - loss 0.10972902 - time (sec): 142.79 - samples/sec: 536.79 - lr: 0.000109 - momentum: 0.000000
151
+ 2023-10-14 21:21:13,144 epoch 4 - iter 440/447 - loss 0.10846253 - time (sec): 158.91 - samples/sec: 536.32 - lr: 0.000107 - momentum: 0.000000
152
+ 2023-10-14 21:21:15,594 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-14 21:21:15,595 EPOCH 4 done: loss 0.1079 - lr: 0.000107
154
+ 2023-10-14 21:21:41,965 DEV : loss 0.13766774535179138 - f1-score (micro avg) 0.7462
155
+ 2023-10-14 21:21:41,991 saving best model
156
+ 2023-10-14 21:21:45,171 ----------------------------------------------------------------------------------------------------
157
+ 2023-10-14 21:22:00,307 epoch 5 - iter 44/447 - loss 0.07404548 - time (sec): 15.13 - samples/sec: 510.58 - lr: 0.000105 - momentum: 0.000000
158
+ 2023-10-14 21:22:16,591 epoch 5 - iter 88/447 - loss 0.07807763 - time (sec): 31.42 - samples/sec: 530.92 - lr: 0.000103 - momentum: 0.000000
159
+ 2023-10-14 21:22:32,708 epoch 5 - iter 132/447 - loss 0.06752753 - time (sec): 47.53 - samples/sec: 544.30 - lr: 0.000102 - momentum: 0.000000
160
+ 2023-10-14 21:22:48,572 epoch 5 - iter 176/447 - loss 0.06407989 - time (sec): 63.40 - samples/sec: 542.33 - lr: 0.000100 - momentum: 0.000000
161
+ 2023-10-14 21:23:03,676 epoch 5 - iter 220/447 - loss 0.06586669 - time (sec): 78.50 - samples/sec: 538.22 - lr: 0.000098 - momentum: 0.000000
162
+ 2023-10-14 21:23:19,808 epoch 5 - iter 264/447 - loss 0.06564498 - time (sec): 94.63 - samples/sec: 539.62 - lr: 0.000096 - momentum: 0.000000
163
+ 2023-10-14 21:23:35,863 epoch 5 - iter 308/447 - loss 0.06768844 - time (sec): 110.69 - samples/sec: 532.24 - lr: 0.000095 - momentum: 0.000000
164
+ 2023-10-14 21:23:51,575 epoch 5 - iter 352/447 - loss 0.06637806 - time (sec): 126.40 - samples/sec: 532.82 - lr: 0.000093 - momentum: 0.000000
165
+ 2023-10-14 21:24:09,379 epoch 5 - iter 396/447 - loss 0.06890479 - time (sec): 144.21 - samples/sec: 535.15 - lr: 0.000091 - momentum: 0.000000
166
+ 2023-10-14 21:24:25,081 epoch 5 - iter 440/447 - loss 0.06779283 - time (sec): 159.91 - samples/sec: 534.05 - lr: 0.000089 - momentum: 0.000000
167
+ 2023-10-14 21:24:27,437 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-14 21:24:27,438 EPOCH 5 done: loss 0.0672 - lr: 0.000089
169
+ 2023-10-14 21:24:55,530 DEV : loss 0.1687537580728531 - f1-score (micro avg) 0.7462
170
+ 2023-10-14 21:24:55,560 ----------------------------------------------------------------------------------------------------
171
+ 2023-10-14 21:25:11,105 epoch 6 - iter 44/447 - loss 0.04321969 - time (sec): 15.54 - samples/sec: 509.18 - lr: 0.000087 - momentum: 0.000000
172
+ 2023-10-14 21:25:27,055 epoch 6 - iter 88/447 - loss 0.04095358 - time (sec): 31.49 - samples/sec: 516.82 - lr: 0.000086 - momentum: 0.000000
173
+ 2023-10-14 21:25:42,575 epoch 6 - iter 132/447 - loss 0.04297395 - time (sec): 47.01 - samples/sec: 524.26 - lr: 0.000084 - momentum: 0.000000
174
+ 2023-10-14 21:25:58,344 epoch 6 - iter 176/447 - loss 0.04445698 - time (sec): 62.78 - samples/sec: 531.96 - lr: 0.000082 - momentum: 0.000000
175
+ 2023-10-14 21:26:13,546 epoch 6 - iter 220/447 - loss 0.04268428 - time (sec): 77.99 - samples/sec: 528.41 - lr: 0.000080 - momentum: 0.000000
176
+ 2023-10-14 21:26:30,859 epoch 6 - iter 264/447 - loss 0.04553797 - time (sec): 95.30 - samples/sec: 529.59 - lr: 0.000079 - momentum: 0.000000
177
+ 2023-10-14 21:26:47,490 epoch 6 - iter 308/447 - loss 0.04719176 - time (sec): 111.93 - samples/sec: 534.28 - lr: 0.000077 - momentum: 0.000000
178
+ 2023-10-14 21:27:03,050 epoch 6 - iter 352/447 - loss 0.04533648 - time (sec): 127.49 - samples/sec: 535.45 - lr: 0.000075 - momentum: 0.000000
179
+ 2023-10-14 21:27:19,160 epoch 6 - iter 396/447 - loss 0.04482017 - time (sec): 143.60 - samples/sec: 537.65 - lr: 0.000073 - momentum: 0.000000
180
+ 2023-10-14 21:27:34,862 epoch 6 - iter 440/447 - loss 0.04379707 - time (sec): 159.30 - samples/sec: 536.84 - lr: 0.000072 - momentum: 0.000000
181
+ 2023-10-14 21:27:37,176 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-14 21:27:37,177 EPOCH 6 done: loss 0.0434 - lr: 0.000072
183
+ 2023-10-14 21:28:03,901 DEV : loss 0.1666191667318344 - f1-score (micro avg) 0.7665
184
+ 2023-10-14 21:28:03,928 saving best model
185
+ 2023-10-14 21:28:06,993 ----------------------------------------------------------------------------------------------------
186
+ 2023-10-14 21:28:22,547 epoch 7 - iter 44/447 - loss 0.03003648 - time (sec): 15.55 - samples/sec: 546.70 - lr: 0.000070 - momentum: 0.000000
187
+ 2023-10-14 21:28:40,163 epoch 7 - iter 88/447 - loss 0.03562673 - time (sec): 33.17 - samples/sec: 552.39 - lr: 0.000068 - momentum: 0.000000
188
+ 2023-10-14 21:28:55,731 epoch 7 - iter 132/447 - loss 0.04006384 - time (sec): 48.74 - samples/sec: 547.83 - lr: 0.000066 - momentum: 0.000000
189
+ 2023-10-14 21:29:11,201 epoch 7 - iter 176/447 - loss 0.03708255 - time (sec): 64.21 - samples/sec: 544.17 - lr: 0.000064 - momentum: 0.000000
190
+ 2023-10-14 21:29:26,968 epoch 7 - iter 220/447 - loss 0.03429430 - time (sec): 79.97 - samples/sec: 545.69 - lr: 0.000063 - momentum: 0.000000
191
+ 2023-10-14 21:29:42,856 epoch 7 - iter 264/447 - loss 0.03272464 - time (sec): 95.86 - samples/sec: 539.78 - lr: 0.000061 - momentum: 0.000000
192
+ 2023-10-14 21:29:59,050 epoch 7 - iter 308/447 - loss 0.03215421 - time (sec): 112.06 - samples/sec: 541.01 - lr: 0.000059 - momentum: 0.000000
193
+ 2023-10-14 21:30:14,596 epoch 7 - iter 352/447 - loss 0.03128111 - time (sec): 127.60 - samples/sec: 539.66 - lr: 0.000057 - momentum: 0.000000
194
+ 2023-10-14 21:30:29,944 epoch 7 - iter 396/447 - loss 0.03125527 - time (sec): 142.95 - samples/sec: 537.71 - lr: 0.000056 - momentum: 0.000000
195
+ 2023-10-14 21:30:45,865 epoch 7 - iter 440/447 - loss 0.03057599 - time (sec): 158.87 - samples/sec: 538.26 - lr: 0.000054 - momentum: 0.000000
196
+ 2023-10-14 21:30:48,160 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-14 21:30:48,161 EPOCH 7 done: loss 0.0303 - lr: 0.000054
198
+ 2023-10-14 21:31:15,339 DEV : loss 0.18828615546226501 - f1-score (micro avg) 0.7731
199
+ 2023-10-14 21:31:15,366 saving best model
200
+ 2023-10-14 21:31:18,393 ----------------------------------------------------------------------------------------------------
201
+ 2023-10-14 21:31:34,147 epoch 8 - iter 44/447 - loss 0.02130714 - time (sec): 15.75 - samples/sec: 541.91 - lr: 0.000052 - momentum: 0.000000
202
+ 2023-10-14 21:31:49,826 epoch 8 - iter 88/447 - loss 0.02342729 - time (sec): 31.43 - samples/sec: 541.15 - lr: 0.000050 - momentum: 0.000000
203
+ 2023-10-14 21:32:05,348 epoch 8 - iter 132/447 - loss 0.02076442 - time (sec): 46.95 - samples/sec: 533.30 - lr: 0.000048 - momentum: 0.000000
204
+ 2023-10-14 21:32:20,821 epoch 8 - iter 176/447 - loss 0.02165836 - time (sec): 62.43 - samples/sec: 531.70 - lr: 0.000047 - momentum: 0.000000
205
+ 2023-10-14 21:32:37,129 epoch 8 - iter 220/447 - loss 0.02071764 - time (sec): 78.73 - samples/sec: 536.62 - lr: 0.000045 - momentum: 0.000000
206
+ 2023-10-14 21:32:54,732 epoch 8 - iter 264/447 - loss 0.02334353 - time (sec): 96.34 - samples/sec: 537.40 - lr: 0.000043 - momentum: 0.000000
207
+ 2023-10-14 21:33:10,358 epoch 8 - iter 308/447 - loss 0.02194638 - time (sec): 111.96 - samples/sec: 534.11 - lr: 0.000041 - momentum: 0.000000
208
+ 2023-10-14 21:33:25,992 epoch 8 - iter 352/447 - loss 0.02099589 - time (sec): 127.60 - samples/sec: 536.19 - lr: 0.000040 - momentum: 0.000000
209
+ 2023-10-14 21:33:42,092 epoch 8 - iter 396/447 - loss 0.02213319 - time (sec): 143.70 - samples/sec: 539.32 - lr: 0.000038 - momentum: 0.000000
210
+ 2023-10-14 21:33:57,461 epoch 8 - iter 440/447 - loss 0.02255177 - time (sec): 159.07 - samples/sec: 535.75 - lr: 0.000036 - momentum: 0.000000
211
+ 2023-10-14 21:33:59,997 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-14 21:33:59,997 EPOCH 8 done: loss 0.0223 - lr: 0.000036
213
+ 2023-10-14 21:34:26,324 DEV : loss 0.18658936023712158 - f1-score (micro avg) 0.7736
214
+ 2023-10-14 21:34:26,350 saving best model
215
+ 2023-10-14 21:34:27,348 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-14 21:34:45,188 epoch 9 - iter 44/447 - loss 0.01651796 - time (sec): 17.84 - samples/sec: 556.26 - lr: 0.000034 - momentum: 0.000000
217
+ 2023-10-14 21:35:00,545 epoch 9 - iter 88/447 - loss 0.01768868 - time (sec): 33.20 - samples/sec: 541.02 - lr: 0.000032 - momentum: 0.000000
218
+ 2023-10-14 21:35:15,508 epoch 9 - iter 132/447 - loss 0.02368658 - time (sec): 48.16 - samples/sec: 526.96 - lr: 0.000031 - momentum: 0.000000
219
+ 2023-10-14 21:35:31,645 epoch 9 - iter 176/447 - loss 0.02037549 - time (sec): 64.30 - samples/sec: 533.36 - lr: 0.000029 - momentum: 0.000000
220
+ 2023-10-14 21:35:47,262 epoch 9 - iter 220/447 - loss 0.02014087 - time (sec): 79.91 - samples/sec: 530.52 - lr: 0.000027 - momentum: 0.000000
221
+ 2023-10-14 21:36:02,787 epoch 9 - iter 264/447 - loss 0.01849602 - time (sec): 95.44 - samples/sec: 531.50 - lr: 0.000025 - momentum: 0.000000
222
+ 2023-10-14 21:36:18,404 epoch 9 - iter 308/447 - loss 0.01829682 - time (sec): 111.06 - samples/sec: 535.34 - lr: 0.000024 - momentum: 0.000000
223
+ 2023-10-14 21:36:34,355 epoch 9 - iter 352/447 - loss 0.01773948 - time (sec): 127.01 - samples/sec: 536.82 - lr: 0.000022 - momentum: 0.000000
224
+ 2023-10-14 21:36:50,308 epoch 9 - iter 396/447 - loss 0.01801785 - time (sec): 142.96 - samples/sec: 539.52 - lr: 0.000020 - momentum: 0.000000
225
+ 2023-10-14 21:37:05,925 epoch 9 - iter 440/447 - loss 0.01754370 - time (sec): 158.58 - samples/sec: 538.92 - lr: 0.000018 - momentum: 0.000000
226
+ 2023-10-14 21:37:08,236 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-14 21:37:08,236 EPOCH 9 done: loss 0.0176 - lr: 0.000018
228
+ 2023-10-14 21:37:34,203 DEV : loss 0.1905660331249237 - f1-score (micro avg) 0.7635
229
+ 2023-10-14 21:37:34,230 ----------------------------------------------------------------------------------------------------
230
+ 2023-10-14 21:37:49,693 epoch 10 - iter 44/447 - loss 0.00886578 - time (sec): 15.46 - samples/sec: 543.68 - lr: 0.000016 - momentum: 0.000000
231
+ 2023-10-14 21:38:04,706 epoch 10 - iter 88/447 - loss 0.02260186 - time (sec): 30.48 - samples/sec: 526.99 - lr: 0.000015 - momentum: 0.000000
232
+ 2023-10-14 21:38:20,881 epoch 10 - iter 132/447 - loss 0.01836041 - time (sec): 46.65 - samples/sec: 533.38 - lr: 0.000013 - momentum: 0.000000
233
+ 2023-10-14 21:38:36,682 epoch 10 - iter 176/447 - loss 0.01658490 - time (sec): 62.45 - samples/sec: 524.38 - lr: 0.000011 - momentum: 0.000000
234
+ 2023-10-14 21:38:54,679 epoch 10 - iter 220/447 - loss 0.01621745 - time (sec): 80.45 - samples/sec: 531.55 - lr: 0.000009 - momentum: 0.000000
235
+ 2023-10-14 21:39:10,341 epoch 10 - iter 264/447 - loss 0.01459106 - time (sec): 96.11 - samples/sec: 527.46 - lr: 0.000008 - momentum: 0.000000
236
+ 2023-10-14 21:39:26,500 epoch 10 - iter 308/447 - loss 0.01436938 - time (sec): 112.27 - samples/sec: 527.94 - lr: 0.000006 - momentum: 0.000000
237
+ 2023-10-14 21:39:42,850 epoch 10 - iter 352/447 - loss 0.01351933 - time (sec): 128.62 - samples/sec: 530.03 - lr: 0.000004 - momentum: 0.000000
238
+ 2023-10-14 21:39:59,071 epoch 10 - iter 396/447 - loss 0.01384647 - time (sec): 144.84 - samples/sec: 531.61 - lr: 0.000002 - momentum: 0.000000
239
+ 2023-10-14 21:40:14,837 epoch 10 - iter 440/447 - loss 0.01429302 - time (sec): 160.61 - samples/sec: 531.46 - lr: 0.000001 - momentum: 0.000000
240
+ 2023-10-14 21:40:17,225 ----------------------------------------------------------------------------------------------------
241
+ 2023-10-14 21:40:17,226 EPOCH 10 done: loss 0.0142 - lr: 0.000001
242
+ 2023-10-14 21:40:43,527 DEV : loss 0.19451487064361572 - f1-score (micro avg) 0.771
243
+ 2023-10-14 21:40:44,146 ----------------------------------------------------------------------------------------------------
244
+ 2023-10-14 21:40:44,148 Loading model from best epoch ...
245
+ 2023-10-14 21:40:46,758 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
246
+ 2023-10-14 21:41:09,803
247
+ Results:
248
+ - F-score (micro) 0.7291
249
+ - F-score (macro) 0.6237
250
+ - Accuracy 0.5887
251
+
252
+ By class:
253
+ precision recall f1-score support
254
+
255
+ loc 0.8557 0.8658 0.8607 596
256
+ pers 0.6269 0.7417 0.6795 333
257
+ org 0.4020 0.6061 0.4834 132
258
+ prod 0.5918 0.4394 0.5043 66
259
+ time 0.5536 0.6327 0.5905 49
260
+
261
+ micro avg 0.6941 0.7679 0.7291 1176
262
+ macro avg 0.6060 0.6571 0.6237 1176
263
+ weighted avg 0.7126 0.7679 0.7358 1176
264
+
265
+ 2023-10-14 21:41:09,803 ----------------------------------------------------------------------------------------------------