Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbyt5-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
d477832
·
1 Parent(s): 34bd27f

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.1697068183.de2e83fddbee.1120.18 +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 sha256:228705614876cbb6ecd26d06307683039cb600ed07b1e2a5fddacb9ff3479de5
3
+ size 870817519
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:babe67ea5b72d4448ec259ea39384094c5496afe9eb0e442fda693257f5c92fb
3
+ size 870817636
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 23:59:15 0.0001 1.1386 0.1807 0.5818 0.6190 0.5999 0.4652
3
+ 2 00:08:48 0.0001 0.1342 0.1144 0.7436 0.7850 0.7637 0.6404
4
+ 3 00:18:24 0.0001 0.0724 0.1281 0.7609 0.8054 0.7826 0.6585
5
+ 4 00:28:11 0.0001 0.0511 0.1459 0.7936 0.8109 0.8022 0.6906
6
+ 5 00:38:02 0.0001 0.0377 0.1630 0.7900 0.8136 0.8016 0.6858
7
+ 6 00:47:42 0.0001 0.0287 0.1803 0.7792 0.8068 0.7928 0.6731
8
+ 7 00:57:08 0.0001 0.0202 0.1907 0.7668 0.8095 0.7876 0.6648
9
+ 8 01:06:38 0.0000 0.0152 0.2064 0.7640 0.7973 0.7803 0.6555
10
+ 9 01:16:03 0.0000 0.0110 0.2116 0.7720 0.8109 0.7910 0.6719
11
+ 10 01:25:08 0.0000 0.0085 0.2086 0.7646 0.8041 0.7838 0.6611
runs/events.out.tfevents.1697068183.de2e83fddbee.1120.18 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf4208df9f0d2b7bc6fcc3f3fc92a9ac30786e041ba8b626e578dfa46d5e911f
3
+ size 999862
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-11 23:49:43,888 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-11 23:49:43,890 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=17, bias=True)
68
+ (loss_function): CrossEntropyLoss()
69
+ )"
70
+ 2023-10-11 23:49:43,890 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-11 23:49:43,890 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
72
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
73
+ 2023-10-11 23:49:43,890 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-11 23:49:43,891 Train: 7142 sentences
75
+ 2023-10-11 23:49:43,891 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-11 23:49:43,891 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-11 23:49:43,891 Training Params:
78
+ 2023-10-11 23:49:43,891 - learning_rate: "0.00015"
79
+ 2023-10-11 23:49:43,891 - mini_batch_size: "4"
80
+ 2023-10-11 23:49:43,891 - max_epochs: "10"
81
+ 2023-10-11 23:49:43,891 - shuffle: "True"
82
+ 2023-10-11 23:49:43,891 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-11 23:49:43,891 Plugins:
84
+ 2023-10-11 23:49:43,891 - TensorboardLogger
85
+ 2023-10-11 23:49:43,891 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-11 23:49:43,891 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-11 23:49:43,891 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-11 23:49:43,891 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-11 23:49:43,892 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-11 23:49:43,892 Computation:
91
+ 2023-10-11 23:49:43,892 - compute on device: cuda:0
92
+ 2023-10-11 23:49:43,892 - embedding storage: none
93
+ 2023-10-11 23:49:43,892 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-11 23:49:43,892 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
95
+ 2023-10-11 23:49:43,892 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-11 23:49:43,892 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-11 23:49:43,892 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-11 23:50:42,826 epoch 1 - iter 178/1786 - loss 2.80679544 - time (sec): 58.93 - samples/sec: 459.62 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-11 23:51:38,946 epoch 1 - iter 356/1786 - loss 2.65034125 - time (sec): 115.05 - samples/sec: 459.46 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-11 23:52:36,349 epoch 1 - iter 534/1786 - loss 2.36581802 - time (sec): 172.45 - samples/sec: 462.09 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-11 23:53:31,134 epoch 1 - iter 712/1786 - loss 2.08798123 - time (sec): 227.24 - samples/sec: 461.30 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-11 23:54:26,815 epoch 1 - iter 890/1786 - loss 1.83005658 - time (sec): 282.92 - samples/sec: 457.29 - lr: 0.000075 - momentum: 0.000000
103
+ 2023-10-11 23:55:20,922 epoch 1 - iter 1068/1786 - loss 1.63577264 - time (sec): 337.03 - samples/sec: 454.06 - lr: 0.000090 - momentum: 0.000000
104
+ 2023-10-11 23:56:15,319 epoch 1 - iter 1246/1786 - loss 1.47180576 - time (sec): 391.43 - samples/sec: 452.29 - lr: 0.000105 - momentum: 0.000000
105
+ 2023-10-11 23:57:09,254 epoch 1 - iter 1424/1786 - loss 1.34808408 - time (sec): 445.36 - samples/sec: 447.88 - lr: 0.000120 - momentum: 0.000000
106
+ 2023-10-11 23:58:01,774 epoch 1 - iter 1602/1786 - loss 1.23634249 - time (sec): 497.88 - samples/sec: 449.14 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-11 23:58:54,318 epoch 1 - iter 1780/1786 - loss 1.14122061 - time (sec): 550.42 - samples/sec: 450.73 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-11 23:58:55,874 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-11 23:58:55,875 EPOCH 1 done: loss 1.1386 - lr: 0.000149
110
+ 2023-10-11 23:59:15,310 DEV : loss 0.18069760501384735 - f1-score (micro avg) 0.5999
111
+ 2023-10-11 23:59:15,342 saving best model
112
+ 2023-10-11 23:59:16,229 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-12 00:00:10,169 epoch 2 - iter 178/1786 - loss 0.20711620 - time (sec): 53.94 - samples/sec: 462.66 - lr: 0.000148 - momentum: 0.000000
114
+ 2023-10-12 00:01:04,653 epoch 2 - iter 356/1786 - loss 0.19247005 - time (sec): 108.42 - samples/sec: 464.22 - lr: 0.000147 - momentum: 0.000000
115
+ 2023-10-12 00:02:00,536 epoch 2 - iter 534/1786 - loss 0.17794878 - time (sec): 164.30 - samples/sec: 458.93 - lr: 0.000145 - momentum: 0.000000
116
+ 2023-10-12 00:02:55,523 epoch 2 - iter 712/1786 - loss 0.16748993 - time (sec): 219.29 - samples/sec: 455.31 - lr: 0.000143 - momentum: 0.000000
117
+ 2023-10-12 00:03:52,876 epoch 2 - iter 890/1786 - loss 0.15565609 - time (sec): 276.64 - samples/sec: 456.14 - lr: 0.000142 - momentum: 0.000000
118
+ 2023-10-12 00:04:47,012 epoch 2 - iter 1068/1786 - loss 0.15126686 - time (sec): 330.78 - samples/sec: 451.88 - lr: 0.000140 - momentum: 0.000000
119
+ 2023-10-12 00:05:41,211 epoch 2 - iter 1246/1786 - loss 0.14667868 - time (sec): 384.98 - samples/sec: 450.19 - lr: 0.000138 - momentum: 0.000000
120
+ 2023-10-12 00:06:37,571 epoch 2 - iter 1424/1786 - loss 0.14213535 - time (sec): 441.34 - samples/sec: 450.42 - lr: 0.000137 - momentum: 0.000000
121
+ 2023-10-12 00:07:31,406 epoch 2 - iter 1602/1786 - loss 0.13922305 - time (sec): 495.17 - samples/sec: 450.14 - lr: 0.000135 - momentum: 0.000000
122
+ 2023-10-12 00:08:24,847 epoch 2 - iter 1780/1786 - loss 0.13416379 - time (sec): 548.62 - samples/sec: 451.32 - lr: 0.000133 - momentum: 0.000000
123
+ 2023-10-12 00:08:26,740 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-12 00:08:26,741 EPOCH 2 done: loss 0.1342 - lr: 0.000133
125
+ 2023-10-12 00:08:48,065 DEV : loss 0.11438284069299698 - f1-score (micro avg) 0.7637
126
+ 2023-10-12 00:08:48,096 saving best model
127
+ 2023-10-12 00:08:51,147 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-12 00:09:44,510 epoch 3 - iter 178/1786 - loss 0.06683015 - time (sec): 53.36 - samples/sec: 459.96 - lr: 0.000132 - momentum: 0.000000
129
+ 2023-10-12 00:10:39,022 epoch 3 - iter 356/1786 - loss 0.06750141 - time (sec): 107.87 - samples/sec: 464.38 - lr: 0.000130 - momentum: 0.000000
130
+ 2023-10-12 00:11:33,861 epoch 3 - iter 534/1786 - loss 0.07127963 - time (sec): 162.71 - samples/sec: 453.13 - lr: 0.000128 - momentum: 0.000000
131
+ 2023-10-12 00:12:28,893 epoch 3 - iter 712/1786 - loss 0.07159232 - time (sec): 217.74 - samples/sec: 449.53 - lr: 0.000127 - momentum: 0.000000
132
+ 2023-10-12 00:13:25,201 epoch 3 - iter 890/1786 - loss 0.06993220 - time (sec): 274.05 - samples/sec: 451.16 - lr: 0.000125 - momentum: 0.000000
133
+ 2023-10-12 00:14:19,775 epoch 3 - iter 1068/1786 - loss 0.07164423 - time (sec): 328.62 - samples/sec: 454.34 - lr: 0.000123 - momentum: 0.000000
134
+ 2023-10-12 00:15:14,214 epoch 3 - iter 1246/1786 - loss 0.07103782 - time (sec): 383.06 - samples/sec: 454.45 - lr: 0.000122 - momentum: 0.000000
135
+ 2023-10-12 00:16:08,217 epoch 3 - iter 1424/1786 - loss 0.07200320 - time (sec): 437.07 - samples/sec: 451.39 - lr: 0.000120 - momentum: 0.000000
136
+ 2023-10-12 00:17:03,415 epoch 3 - iter 1602/1786 - loss 0.07374063 - time (sec): 492.26 - samples/sec: 450.32 - lr: 0.000118 - momentum: 0.000000
137
+ 2023-10-12 00:17:59,334 epoch 3 - iter 1780/1786 - loss 0.07209817 - time (sec): 548.18 - samples/sec: 452.30 - lr: 0.000117 - momentum: 0.000000
138
+ 2023-10-12 00:18:00,989 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-12 00:18:00,990 EPOCH 3 done: loss 0.0724 - lr: 0.000117
140
+ 2023-10-12 00:18:24,115 DEV : loss 0.1281142383813858 - f1-score (micro avg) 0.7826
141
+ 2023-10-12 00:18:24,147 saving best model
142
+ 2023-10-12 00:18:40,438 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-12 00:19:35,647 epoch 4 - iter 178/1786 - loss 0.05657524 - time (sec): 55.21 - samples/sec: 485.86 - lr: 0.000115 - momentum: 0.000000
144
+ 2023-10-12 00:20:30,635 epoch 4 - iter 356/1786 - loss 0.05425663 - time (sec): 110.19 - samples/sec: 461.11 - lr: 0.000113 - momentum: 0.000000
145
+ 2023-10-12 00:21:25,821 epoch 4 - iter 534/1786 - loss 0.05035837 - time (sec): 165.38 - samples/sec: 458.97 - lr: 0.000112 - momentum: 0.000000
146
+ 2023-10-12 00:22:20,078 epoch 4 - iter 712/1786 - loss 0.05157633 - time (sec): 219.64 - samples/sec: 457.43 - lr: 0.000110 - momentum: 0.000000
147
+ 2023-10-12 00:23:14,198 epoch 4 - iter 890/1786 - loss 0.05225442 - time (sec): 273.76 - samples/sec: 453.18 - lr: 0.000108 - momentum: 0.000000
148
+ 2023-10-12 00:24:09,428 epoch 4 - iter 1068/1786 - loss 0.05116186 - time (sec): 328.99 - samples/sec: 453.97 - lr: 0.000107 - momentum: 0.000000
149
+ 2023-10-12 00:25:03,579 epoch 4 - iter 1246/1786 - loss 0.05109624 - time (sec): 383.14 - samples/sec: 451.90 - lr: 0.000105 - momentum: 0.000000
150
+ 2023-10-12 00:25:57,486 epoch 4 - iter 1424/1786 - loss 0.05046609 - time (sec): 437.04 - samples/sec: 451.18 - lr: 0.000103 - momentum: 0.000000
151
+ 2023-10-12 00:26:54,137 epoch 4 - iter 1602/1786 - loss 0.05099487 - time (sec): 493.70 - samples/sec: 453.53 - lr: 0.000102 - momentum: 0.000000
152
+ 2023-10-12 00:27:48,105 epoch 4 - iter 1780/1786 - loss 0.05104060 - time (sec): 547.66 - samples/sec: 453.01 - lr: 0.000100 - momentum: 0.000000
153
+ 2023-10-12 00:27:49,733 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-12 00:27:49,733 EPOCH 4 done: loss 0.0511 - lr: 0.000100
155
+ 2023-10-12 00:28:11,834 DEV : loss 0.14586448669433594 - f1-score (micro avg) 0.8022
156
+ 2023-10-12 00:28:11,868 saving best model
157
+ 2023-10-12 00:28:22,595 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-12 00:29:16,940 epoch 5 - iter 178/1786 - loss 0.03100195 - time (sec): 54.34 - samples/sec: 451.46 - lr: 0.000098 - momentum: 0.000000
159
+ 2023-10-12 00:30:10,437 epoch 5 - iter 356/1786 - loss 0.03243173 - time (sec): 107.84 - samples/sec: 454.11 - lr: 0.000097 - momentum: 0.000000
160
+ 2023-10-12 00:31:04,353 epoch 5 - iter 534/1786 - loss 0.03471857 - time (sec): 161.75 - samples/sec: 459.07 - lr: 0.000095 - momentum: 0.000000
161
+ 2023-10-12 00:31:57,749 epoch 5 - iter 712/1786 - loss 0.03271395 - time (sec): 215.15 - samples/sec: 454.35 - lr: 0.000093 - momentum: 0.000000
162
+ 2023-10-12 00:32:54,440 epoch 5 - iter 890/1786 - loss 0.03439854 - time (sec): 271.84 - samples/sec: 447.74 - lr: 0.000092 - momentum: 0.000000
163
+ 2023-10-12 00:33:50,980 epoch 5 - iter 1068/1786 - loss 0.03404286 - time (sec): 328.38 - samples/sec: 445.72 - lr: 0.000090 - momentum: 0.000000
164
+ 2023-10-12 00:34:49,518 epoch 5 - iter 1246/1786 - loss 0.03481655 - time (sec): 386.92 - samples/sec: 448.76 - lr: 0.000088 - momentum: 0.000000
165
+ 2023-10-12 00:35:44,748 epoch 5 - iter 1424/1786 - loss 0.03575715 - time (sec): 442.15 - samples/sec: 448.84 - lr: 0.000087 - momentum: 0.000000
166
+ 2023-10-12 00:36:41,662 epoch 5 - iter 1602/1786 - loss 0.03690779 - time (sec): 499.06 - samples/sec: 447.33 - lr: 0.000085 - momentum: 0.000000
167
+ 2023-10-12 00:37:37,592 epoch 5 - iter 1780/1786 - loss 0.03763514 - time (sec): 554.99 - samples/sec: 447.04 - lr: 0.000083 - momentum: 0.000000
168
+ 2023-10-12 00:37:39,239 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-12 00:37:39,239 EPOCH 5 done: loss 0.0377 - lr: 0.000083
170
+ 2023-10-12 00:38:02,466 DEV : loss 0.16297270357608795 - f1-score (micro avg) 0.8016
171
+ 2023-10-12 00:38:02,499 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-12 00:38:57,768 epoch 6 - iter 178/1786 - loss 0.02668569 - time (sec): 55.27 - samples/sec: 464.94 - lr: 0.000082 - momentum: 0.000000
173
+ 2023-10-12 00:39:51,559 epoch 6 - iter 356/1786 - loss 0.02601327 - time (sec): 109.06 - samples/sec: 456.49 - lr: 0.000080 - momentum: 0.000000
174
+ 2023-10-12 00:40:48,681 epoch 6 - iter 534/1786 - loss 0.02650022 - time (sec): 166.18 - samples/sec: 464.21 - lr: 0.000078 - momentum: 0.000000
175
+ 2023-10-12 00:41:43,107 epoch 6 - iter 712/1786 - loss 0.02802396 - time (sec): 220.61 - samples/sec: 459.29 - lr: 0.000077 - momentum: 0.000000
176
+ 2023-10-12 00:42:38,563 epoch 6 - iter 890/1786 - loss 0.02839531 - time (sec): 276.06 - samples/sec: 461.31 - lr: 0.000075 - momentum: 0.000000
177
+ 2023-10-12 00:43:36,355 epoch 6 - iter 1068/1786 - loss 0.02844378 - time (sec): 333.85 - samples/sec: 454.37 - lr: 0.000073 - momentum: 0.000000
178
+ 2023-10-12 00:44:32,658 epoch 6 - iter 1246/1786 - loss 0.02807222 - time (sec): 390.16 - samples/sec: 450.56 - lr: 0.000072 - momentum: 0.000000
179
+ 2023-10-12 00:45:29,116 epoch 6 - iter 1424/1786 - loss 0.02797952 - time (sec): 446.62 - samples/sec: 449.93 - lr: 0.000070 - momentum: 0.000000
180
+ 2023-10-12 00:46:23,766 epoch 6 - iter 1602/1786 - loss 0.02796450 - time (sec): 501.27 - samples/sec: 447.34 - lr: 0.000068 - momentum: 0.000000
181
+ 2023-10-12 00:47:17,742 epoch 6 - iter 1780/1786 - loss 0.02870438 - time (sec): 555.24 - samples/sec: 446.07 - lr: 0.000067 - momentum: 0.000000
182
+ 2023-10-12 00:47:19,636 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-12 00:47:19,636 EPOCH 6 done: loss 0.0287 - lr: 0.000067
184
+ 2023-10-12 00:47:42,108 DEV : loss 0.18033917248249054 - f1-score (micro avg) 0.7928
185
+ 2023-10-12 00:47:42,140 ----------------------------------------------------------------------------------------------------
186
+ 2023-10-12 00:48:37,819 epoch 7 - iter 178/1786 - loss 0.02839486 - time (sec): 55.68 - samples/sec: 432.28 - lr: 0.000065 - momentum: 0.000000
187
+ 2023-10-12 00:49:33,964 epoch 7 - iter 356/1786 - loss 0.02271661 - time (sec): 111.82 - samples/sec: 445.36 - lr: 0.000063 - momentum: 0.000000
188
+ 2023-10-12 00:50:28,746 epoch 7 - iter 534/1786 - loss 0.02274955 - time (sec): 166.60 - samples/sec: 441.96 - lr: 0.000062 - momentum: 0.000000
189
+ 2023-10-12 00:51:23,278 epoch 7 - iter 712/1786 - loss 0.02044904 - time (sec): 221.14 - samples/sec: 449.84 - lr: 0.000060 - momentum: 0.000000
190
+ 2023-10-12 00:52:15,708 epoch 7 - iter 890/1786 - loss 0.02037246 - time (sec): 273.57 - samples/sec: 455.10 - lr: 0.000058 - momentum: 0.000000
191
+ 2023-10-12 00:53:08,412 epoch 7 - iter 1068/1786 - loss 0.01965538 - time (sec): 326.27 - samples/sec: 457.62 - lr: 0.000057 - momentum: 0.000000
192
+ 2023-10-12 00:54:02,855 epoch 7 - iter 1246/1786 - loss 0.01962515 - time (sec): 380.71 - samples/sec: 456.00 - lr: 0.000055 - momentum: 0.000000
193
+ 2023-10-12 00:54:57,358 epoch 7 - iter 1424/1786 - loss 0.02031428 - time (sec): 435.22 - samples/sec: 454.87 - lr: 0.000053 - momentum: 0.000000
194
+ 2023-10-12 00:55:52,160 epoch 7 - iter 1602/1786 - loss 0.02032900 - time (sec): 490.02 - samples/sec: 455.60 - lr: 0.000052 - momentum: 0.000000
195
+ 2023-10-12 00:56:45,131 epoch 7 - iter 1780/1786 - loss 0.02028515 - time (sec): 542.99 - samples/sec: 456.49 - lr: 0.000050 - momentum: 0.000000
196
+ 2023-10-12 00:56:46,800 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-12 00:56:46,801 EPOCH 7 done: loss 0.0202 - lr: 0.000050
198
+ 2023-10-12 00:57:08,705 DEV : loss 0.19071684777736664 - f1-score (micro avg) 0.7876
199
+ 2023-10-12 00:57:08,738 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-12 00:58:03,715 epoch 8 - iter 178/1786 - loss 0.01789598 - time (sec): 54.97 - samples/sec: 455.59 - lr: 0.000048 - momentum: 0.000000
201
+ 2023-10-12 00:58:59,441 epoch 8 - iter 356/1786 - loss 0.01800163 - time (sec): 110.70 - samples/sec: 454.42 - lr: 0.000047 - momentum: 0.000000
202
+ 2023-10-12 00:59:55,590 epoch 8 - iter 534/1786 - loss 0.01707568 - time (sec): 166.85 - samples/sec: 450.43 - lr: 0.000045 - momentum: 0.000000
203
+ 2023-10-12 01:00:50,158 epoch 8 - iter 712/1786 - loss 0.01690096 - time (sec): 221.42 - samples/sec: 444.03 - lr: 0.000043 - momentum: 0.000000
204
+ 2023-10-12 01:01:42,049 epoch 8 - iter 890/1786 - loss 0.01620502 - time (sec): 273.31 - samples/sec: 445.86 - lr: 0.000042 - momentum: 0.000000
205
+ 2023-10-12 01:02:35,862 epoch 8 - iter 1068/1786 - loss 0.01592801 - time (sec): 327.12 - samples/sec: 452.71 - lr: 0.000040 - momentum: 0.000000
206
+ 2023-10-12 01:03:29,081 epoch 8 - iter 1246/1786 - loss 0.01640440 - time (sec): 380.34 - samples/sec: 449.84 - lr: 0.000038 - momentum: 0.000000
207
+ 2023-10-12 01:04:24,552 epoch 8 - iter 1424/1786 - loss 0.01591161 - time (sec): 435.81 - samples/sec: 452.42 - lr: 0.000037 - momentum: 0.000000
208
+ 2023-10-12 01:05:20,992 epoch 8 - iter 1602/1786 - loss 0.01561111 - time (sec): 492.25 - samples/sec: 453.96 - lr: 0.000035 - momentum: 0.000000
209
+ 2023-10-12 01:06:14,936 epoch 8 - iter 1780/1786 - loss 0.01528229 - time (sec): 546.20 - samples/sec: 453.87 - lr: 0.000033 - momentum: 0.000000
210
+ 2023-10-12 01:06:16,726 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-12 01:06:16,726 EPOCH 8 done: loss 0.0152 - lr: 0.000033
212
+ 2023-10-12 01:06:38,521 DEV : loss 0.2063598781824112 - f1-score (micro avg) 0.7803
213
+ 2023-10-12 01:06:38,552 ----------------------------------------------------------------------------------------------------
214
+ 2023-10-12 01:07:33,634 epoch 9 - iter 178/1786 - loss 0.01097164 - time (sec): 55.08 - samples/sec: 470.62 - lr: 0.000032 - momentum: 0.000000
215
+ 2023-10-12 01:08:26,848 epoch 9 - iter 356/1786 - loss 0.00871512 - time (sec): 108.29 - samples/sec: 461.16 - lr: 0.000030 - momentum: 0.000000
216
+ 2023-10-12 01:09:20,346 epoch 9 - iter 534/1786 - loss 0.01138957 - time (sec): 161.79 - samples/sec: 459.33 - lr: 0.000028 - momentum: 0.000000
217
+ 2023-10-12 01:10:14,450 epoch 9 - iter 712/1786 - loss 0.01055240 - time (sec): 215.90 - samples/sec: 455.40 - lr: 0.000027 - momentum: 0.000000
218
+ 2023-10-12 01:11:10,367 epoch 9 - iter 890/1786 - loss 0.00994944 - time (sec): 271.81 - samples/sec: 451.87 - lr: 0.000025 - momentum: 0.000000
219
+ 2023-10-12 01:12:06,227 epoch 9 - iter 1068/1786 - loss 0.00965138 - time (sec): 327.67 - samples/sec: 455.01 - lr: 0.000023 - momentum: 0.000000
220
+ 2023-10-12 01:13:00,155 epoch 9 - iter 1246/1786 - loss 0.01059332 - time (sec): 381.60 - samples/sec: 459.44 - lr: 0.000022 - momentum: 0.000000
221
+ 2023-10-12 01:13:53,715 epoch 9 - iter 1424/1786 - loss 0.01083122 - time (sec): 435.16 - samples/sec: 460.45 - lr: 0.000020 - momentum: 0.000000
222
+ 2023-10-12 01:14:47,729 epoch 9 - iter 1602/1786 - loss 0.01115893 - time (sec): 489.17 - samples/sec: 458.87 - lr: 0.000018 - momentum: 0.000000
223
+ 2023-10-12 01:15:40,190 epoch 9 - iter 1780/1786 - loss 0.01106046 - time (sec): 541.64 - samples/sec: 458.00 - lr: 0.000017 - momentum: 0.000000
224
+ 2023-10-12 01:15:41,822 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-12 01:15:41,822 EPOCH 9 done: loss 0.0110 - lr: 0.000017
226
+ 2023-10-12 01:16:03,451 DEV : loss 0.21163716912269592 - f1-score (micro avg) 0.791
227
+ 2023-10-12 01:16:03,484 ----------------------------------------------------------------------------------------------------
228
+ 2023-10-12 01:16:56,422 epoch 10 - iter 178/1786 - loss 0.00564481 - time (sec): 52.94 - samples/sec: 476.78 - lr: 0.000015 - momentum: 0.000000
229
+ 2023-10-12 01:17:50,049 epoch 10 - iter 356/1786 - loss 0.00861168 - time (sec): 106.56 - samples/sec: 473.86 - lr: 0.000013 - momentum: 0.000000
230
+ 2023-10-12 01:18:41,937 epoch 10 - iter 534/1786 - loss 0.00834080 - time (sec): 158.45 - samples/sec: 478.79 - lr: 0.000012 - momentum: 0.000000
231
+ 2023-10-12 01:19:35,823 epoch 10 - iter 712/1786 - loss 0.00920845 - time (sec): 212.34 - samples/sec: 476.27 - lr: 0.000010 - momentum: 0.000000
232
+ 2023-10-12 01:20:28,835 epoch 10 - iter 890/1786 - loss 0.00915627 - time (sec): 265.35 - samples/sec: 476.63 - lr: 0.000008 - momentum: 0.000000
233
+ 2023-10-12 01:21:19,749 epoch 10 - iter 1068/1786 - loss 0.00855923 - time (sec): 316.26 - samples/sec: 475.74 - lr: 0.000007 - momentum: 0.000000
234
+ 2023-10-12 01:22:11,915 epoch 10 - iter 1246/1786 - loss 0.00883037 - time (sec): 368.43 - samples/sec: 478.28 - lr: 0.000005 - momentum: 0.000000
235
+ 2023-10-12 01:23:02,937 epoch 10 - iter 1424/1786 - loss 0.00832188 - time (sec): 419.45 - samples/sec: 477.38 - lr: 0.000003 - momentum: 0.000000
236
+ 2023-10-12 01:23:54,187 epoch 10 - iter 1602/1786 - loss 0.00826011 - time (sec): 470.70 - samples/sec: 476.61 - lr: 0.000002 - momentum: 0.000000
237
+ 2023-10-12 01:24:45,370 epoch 10 - iter 1780/1786 - loss 0.00851589 - time (sec): 521.88 - samples/sec: 475.55 - lr: 0.000000 - momentum: 0.000000
238
+ 2023-10-12 01:24:46,804 ----------------------------------------------------------------------------------------------------
239
+ 2023-10-12 01:24:46,804 EPOCH 10 done: loss 0.0085 - lr: 0.000000
240
+ 2023-10-12 01:25:08,325 DEV : loss 0.20861276984214783 - f1-score (micro avg) 0.7838
241
+ 2023-10-12 01:25:09,217 ----------------------------------------------------------------------------------------------------
242
+ 2023-10-12 01:25:09,219 Loading model from best epoch ...
243
+ 2023-10-12 01:25:13,155 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
244
+ 2023-10-12 01:26:23,943
245
+ Results:
246
+ - F-score (micro) 0.6972
247
+ - F-score (macro) 0.5865
248
+ - Accuracy 0.5472
249
+
250
+ By class:
251
+ precision recall f1-score support
252
+
253
+ LOC 0.7162 0.6959 0.7059 1095
254
+ PER 0.7693 0.7678 0.7685 1012
255
+ ORG 0.5026 0.5490 0.5248 357
256
+ HumanProd 0.2615 0.5152 0.3469 33
257
+
258
+ micro avg 0.6928 0.7016 0.6972 2497
259
+ macro avg 0.5624 0.6320 0.5865 2497
260
+ weighted avg 0.7012 0.7016 0.7006 2497
261
+
262
+ 2023-10-12 01:26:23,943 ----------------------------------------------------------------------------------------------------