Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbyt5-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
d8a1523
•
1 Parent(s): 43a356b

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.1697040404.de2e83fddbee.1120.13 +3 -0
  6. test.tsv +0 -0
  7. training.log +264 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b01e4693301937934eddef90d7eed0bb8b333a80d679e5a7807e8944390dd279
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 sha256:6ec22615b3639554db16895f92c8377f419ee3a0a7e71af6f802282a78880824
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 16:15:58 0.0002 1.2758 0.2641 0.2921 0.2762 0.2839 0.1935
3
+ 2 16:25:05 0.0001 0.1964 0.1075 0.7376 0.7456 0.7415 0.6096
4
+ 3 16:33:38 0.0001 0.0812 0.1133 0.7150 0.7986 0.7545 0.6199
5
+ 4 16:42:20 0.0001 0.0509 0.1336 0.7370 0.8082 0.7709 0.6429
6
+ 5 16:51:02 0.0001 0.0370 0.1409 0.7824 0.8122 0.7971 0.6776
7
+ 6 16:59:56 0.0001 0.0292 0.1641 0.7795 0.8082 0.7936 0.6727
8
+ 7 17:08:31 0.0001 0.0213 0.1789 0.7839 0.8190 0.8011 0.6810
9
+ 8 17:17:18 0.0000 0.0177 0.1916 0.7819 0.8095 0.7955 0.6738
10
+ 9 17:25:46 0.0000 0.0146 0.1948 0.7749 0.8150 0.7944 0.6715
11
+ 10 17:34:20 0.0000 0.0116 0.1996 0.7683 0.8122 0.7897 0.6663
runs/events.out.tfevents.1697040404.de2e83fddbee.1120.13 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39866306fc1fdf57b823fb75ac41f3a7c36d5f28afce74853a567e9ea699e5b8
3
+ size 502461
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,264 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-11 16:06:44,778 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-11 16:06:44,780 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 16:06:44,780 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-11 16:06:44,781 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 16:06:44,781 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-11 16:06:44,781 Train: 7142 sentences
75
+ 2023-10-11 16:06:44,781 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-11 16:06:44,781 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-11 16:06:44,781 Training Params:
78
+ 2023-10-11 16:06:44,781 - learning_rate: "0.00016"
79
+ 2023-10-11 16:06:44,781 - mini_batch_size: "8"
80
+ 2023-10-11 16:06:44,781 - max_epochs: "10"
81
+ 2023-10-11 16:06:44,781 - shuffle: "True"
82
+ 2023-10-11 16:06:44,781 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-11 16:06:44,781 Plugins:
84
+ 2023-10-11 16:06:44,782 - TensorboardLogger
85
+ 2023-10-11 16:06:44,782 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-11 16:06:44,782 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-11 16:06:44,782 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-11 16:06:44,782 Computation:
91
+ 2023-10-11 16:06:44,782 - compute on device: cuda:0
92
+ 2023-10-11 16:06:44,782 - embedding storage: none
93
+ 2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-11 16:06:44,782 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4"
95
+ 2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-11 16:06:44,782 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-11 16:07:34,507 epoch 1 - iter 89/893 - loss 2.83150238 - time (sec): 49.72 - samples/sec: 469.42 - lr: 0.000016 - momentum: 0.000000
99
+ 2023-10-11 16:08:24,856 epoch 1 - iter 178/893 - loss 2.73922034 - time (sec): 100.07 - samples/sec: 478.79 - lr: 0.000032 - momentum: 0.000000
100
+ 2023-10-11 16:09:15,427 epoch 1 - iter 267/893 - loss 2.54002172 - time (sec): 150.64 - samples/sec: 478.44 - lr: 0.000048 - momentum: 0.000000
101
+ 2023-10-11 16:10:07,265 epoch 1 - iter 356/893 - loss 2.31635472 - time (sec): 202.48 - samples/sec: 478.37 - lr: 0.000064 - momentum: 0.000000
102
+ 2023-10-11 16:10:59,517 epoch 1 - iter 445/893 - loss 2.08548476 - time (sec): 254.73 - samples/sec: 472.32 - lr: 0.000080 - momentum: 0.000000
103
+ 2023-10-11 16:11:54,697 epoch 1 - iter 534/893 - loss 1.85181034 - time (sec): 309.91 - samples/sec: 472.21 - lr: 0.000095 - momentum: 0.000000
104
+ 2023-10-11 16:12:50,107 epoch 1 - iter 623/893 - loss 1.65665303 - time (sec): 365.32 - samples/sec: 473.67 - lr: 0.000111 - momentum: 0.000000
105
+ 2023-10-11 16:13:44,323 epoch 1 - iter 712/893 - loss 1.50734839 - time (sec): 419.54 - samples/sec: 472.30 - lr: 0.000127 - momentum: 0.000000
106
+ 2023-10-11 16:14:37,947 epoch 1 - iter 801/893 - loss 1.38574381 - time (sec): 473.16 - samples/sec: 469.95 - lr: 0.000143 - momentum: 0.000000
107
+ 2023-10-11 16:15:34,092 epoch 1 - iter 890/893 - loss 1.27856971 - time (sec): 529.31 - samples/sec: 468.60 - lr: 0.000159 - momentum: 0.000000
108
+ 2023-10-11 16:15:36,427 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-11 16:15:36,427 EPOCH 1 done: loss 1.2758 - lr: 0.000159
110
+ 2023-10-11 16:15:58,475 DEV : loss 0.26407888531684875 - f1-score (micro avg) 0.2839
111
+ 2023-10-11 16:15:58,513 saving best model
112
+ 2023-10-11 16:15:59,455 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-11 16:16:53,512 epoch 2 - iter 89/893 - loss 0.30599282 - time (sec): 54.06 - samples/sec: 475.97 - lr: 0.000158 - momentum: 0.000000
114
+ 2023-10-11 16:17:48,903 epoch 2 - iter 178/893 - loss 0.30164929 - time (sec): 109.45 - samples/sec: 465.09 - lr: 0.000156 - momentum: 0.000000
115
+ 2023-10-11 16:18:45,107 epoch 2 - iter 267/893 - loss 0.27983412 - time (sec): 165.65 - samples/sec: 468.00 - lr: 0.000155 - momentum: 0.000000
116
+ 2023-10-11 16:19:37,291 epoch 2 - iter 356/893 - loss 0.26596551 - time (sec): 217.83 - samples/sec: 462.78 - lr: 0.000153 - momentum: 0.000000
117
+ 2023-10-11 16:20:32,275 epoch 2 - iter 445/893 - loss 0.24908676 - time (sec): 272.82 - samples/sec: 459.68 - lr: 0.000151 - momentum: 0.000000
118
+ 2023-10-11 16:21:26,899 epoch 2 - iter 534/893 - loss 0.23391028 - time (sec): 327.44 - samples/sec: 457.37 - lr: 0.000149 - momentum: 0.000000
119
+ 2023-10-11 16:22:17,982 epoch 2 - iter 623/893 - loss 0.22443918 - time (sec): 378.52 - samples/sec: 459.29 - lr: 0.000148 - momentum: 0.000000
120
+ 2023-10-11 16:23:07,317 epoch 2 - iter 712/893 - loss 0.21432762 - time (sec): 427.86 - samples/sec: 463.46 - lr: 0.000146 - momentum: 0.000000
121
+ 2023-10-11 16:23:54,170 epoch 2 - iter 801/893 - loss 0.20470001 - time (sec): 474.71 - samples/sec: 467.17 - lr: 0.000144 - momentum: 0.000000
122
+ 2023-10-11 16:24:43,031 epoch 2 - iter 890/893 - loss 0.19693577 - time (sec): 523.57 - samples/sec: 472.97 - lr: 0.000142 - momentum: 0.000000
123
+ 2023-10-11 16:24:44,752 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-11 16:24:44,752 EPOCH 2 done: loss 0.1964 - lr: 0.000142
125
+ 2023-10-11 16:25:05,266 DEV : loss 0.10753299295902252 - f1-score (micro avg) 0.7415
126
+ 2023-10-11 16:25:05,296 saving best model
127
+ 2023-10-11 16:25:08,063 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-11 16:25:56,375 epoch 3 - iter 89/893 - loss 0.09369444 - time (sec): 48.31 - samples/sec: 513.41 - lr: 0.000140 - momentum: 0.000000
129
+ 2023-10-11 16:26:45,041 epoch 3 - iter 178/893 - loss 0.09463543 - time (sec): 96.97 - samples/sec: 514.91 - lr: 0.000139 - momentum: 0.000000
130
+ 2023-10-11 16:27:32,896 epoch 3 - iter 267/893 - loss 0.08742843 - time (sec): 144.83 - samples/sec: 511.93 - lr: 0.000137 - momentum: 0.000000
131
+ 2023-10-11 16:28:21,237 epoch 3 - iter 356/893 - loss 0.08324272 - time (sec): 193.17 - samples/sec: 508.09 - lr: 0.000135 - momentum: 0.000000
132
+ 2023-10-11 16:29:09,547 epoch 3 - iter 445/893 - loss 0.08475619 - time (sec): 241.48 - samples/sec: 508.37 - lr: 0.000133 - momentum: 0.000000
133
+ 2023-10-11 16:29:59,366 epoch 3 - iter 534/893 - loss 0.08485703 - time (sec): 291.30 - samples/sec: 506.26 - lr: 0.000132 - momentum: 0.000000
134
+ 2023-10-11 16:30:48,049 epoch 3 - iter 623/893 - loss 0.08297446 - time (sec): 339.98 - samples/sec: 505.75 - lr: 0.000130 - momentum: 0.000000
135
+ 2023-10-11 16:31:36,824 epoch 3 - iter 712/893 - loss 0.08307353 - time (sec): 388.76 - samples/sec: 507.19 - lr: 0.000128 - momentum: 0.000000
136
+ 2023-10-11 16:32:26,257 epoch 3 - iter 801/893 - loss 0.08179325 - time (sec): 438.19 - samples/sec: 507.93 - lr: 0.000126 - momentum: 0.000000
137
+ 2023-10-11 16:33:14,792 epoch 3 - iter 890/893 - loss 0.08102671 - time (sec): 486.73 - samples/sec: 509.04 - lr: 0.000125 - momentum: 0.000000
138
+ 2023-10-11 16:33:16,459 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-11 16:33:16,460 EPOCH 3 done: loss 0.0812 - lr: 0.000125
140
+ 2023-10-11 16:33:38,017 DEV : loss 0.11330673843622208 - f1-score (micro avg) 0.7545
141
+ 2023-10-11 16:33:38,056 saving best model
142
+ 2023-10-11 16:33:40,713 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-11 16:34:31,047 epoch 4 - iter 89/893 - loss 0.05729569 - time (sec): 50.33 - samples/sec: 489.85 - lr: 0.000123 - momentum: 0.000000
144
+ 2023-10-11 16:35:20,311 epoch 4 - iter 178/893 - loss 0.05027581 - time (sec): 99.59 - samples/sec: 498.88 - lr: 0.000121 - momentum: 0.000000
145
+ 2023-10-11 16:36:10,722 epoch 4 - iter 267/893 - loss 0.05117128 - time (sec): 150.00 - samples/sec: 506.36 - lr: 0.000119 - momentum: 0.000000
146
+ 2023-10-11 16:37:00,183 epoch 4 - iter 356/893 - loss 0.05260726 - time (sec): 199.47 - samples/sec: 504.60 - lr: 0.000117 - momentum: 0.000000
147
+ 2023-10-11 16:37:48,423 epoch 4 - iter 445/893 - loss 0.05204046 - time (sec): 247.71 - samples/sec: 503.98 - lr: 0.000116 - momentum: 0.000000
148
+ 2023-10-11 16:38:37,235 epoch 4 - iter 534/893 - loss 0.05076359 - time (sec): 296.52 - samples/sec: 505.45 - lr: 0.000114 - momentum: 0.000000
149
+ 2023-10-11 16:39:27,083 epoch 4 - iter 623/893 - loss 0.05138865 - time (sec): 346.37 - samples/sec: 509.10 - lr: 0.000112 - momentum: 0.000000
150
+ 2023-10-11 16:40:17,785 epoch 4 - iter 712/893 - loss 0.05173834 - time (sec): 397.07 - samples/sec: 502.80 - lr: 0.000110 - momentum: 0.000000
151
+ 2023-10-11 16:41:07,370 epoch 4 - iter 801/893 - loss 0.05192190 - time (sec): 446.65 - samples/sec: 500.87 - lr: 0.000109 - momentum: 0.000000
152
+ 2023-10-11 16:41:57,636 epoch 4 - iter 890/893 - loss 0.05095217 - time (sec): 496.92 - samples/sec: 499.12 - lr: 0.000107 - momentum: 0.000000
153
+ 2023-10-11 16:41:59,142 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-11 16:41:59,143 EPOCH 4 done: loss 0.0509 - lr: 0.000107
155
+ 2023-10-11 16:42:20,924 DEV : loss 0.13362975418567657 - f1-score (micro avg) 0.7709
156
+ 2023-10-11 16:42:20,958 saving best model
157
+ 2023-10-11 16:42:23,554 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-11 16:43:11,098 epoch 5 - iter 89/893 - loss 0.03653288 - time (sec): 47.54 - samples/sec: 500.14 - lr: 0.000105 - momentum: 0.000000
159
+ 2023-10-11 16:44:00,117 epoch 5 - iter 178/893 - loss 0.03595217 - time (sec): 96.56 - samples/sec: 485.02 - lr: 0.000103 - momentum: 0.000000
160
+ 2023-10-11 16:44:49,513 epoch 5 - iter 267/893 - loss 0.03646653 - time (sec): 145.95 - samples/sec: 502.95 - lr: 0.000101 - momentum: 0.000000
161
+ 2023-10-11 16:45:39,118 epoch 5 - iter 356/893 - loss 0.03750037 - time (sec): 195.56 - samples/sec: 505.26 - lr: 0.000100 - momentum: 0.000000
162
+ 2023-10-11 16:46:28,017 epoch 5 - iter 445/893 - loss 0.03695092 - time (sec): 244.46 - samples/sec: 503.19 - lr: 0.000098 - momentum: 0.000000
163
+ 2023-10-11 16:47:16,434 epoch 5 - iter 534/893 - loss 0.03608794 - time (sec): 292.88 - samples/sec: 499.94 - lr: 0.000096 - momentum: 0.000000
164
+ 2023-10-11 16:48:07,153 epoch 5 - iter 623/893 - loss 0.03592828 - time (sec): 343.59 - samples/sec: 502.61 - lr: 0.000094 - momentum: 0.000000
165
+ 2023-10-11 16:48:56,551 epoch 5 - iter 712/893 - loss 0.03600119 - time (sec): 392.99 - samples/sec: 500.53 - lr: 0.000093 - momentum: 0.000000
166
+ 2023-10-11 16:49:47,849 epoch 5 - iter 801/893 - loss 0.03656926 - time (sec): 444.29 - samples/sec: 500.02 - lr: 0.000091 - momentum: 0.000000
167
+ 2023-10-11 16:50:39,277 epoch 5 - iter 890/893 - loss 0.03689169 - time (sec): 495.72 - samples/sec: 499.79 - lr: 0.000089 - momentum: 0.000000
168
+ 2023-10-11 16:50:40,983 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-11 16:50:40,984 EPOCH 5 done: loss 0.0370 - lr: 0.000089
170
+ 2023-10-11 16:51:02,760 DEV : loss 0.1409212052822113 - f1-score (micro avg) 0.7971
171
+ 2023-10-11 16:51:02,797 saving best model
172
+ 2023-10-11 16:51:05,388 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-11 16:51:56,629 epoch 6 - iter 89/893 - loss 0.03093811 - time (sec): 51.24 - samples/sec: 490.34 - lr: 0.000087 - momentum: 0.000000
174
+ 2023-10-11 16:52:48,003 epoch 6 - iter 178/893 - loss 0.02884945 - time (sec): 102.61 - samples/sec: 483.47 - lr: 0.000085 - momentum: 0.000000
175
+ 2023-10-11 16:53:37,672 epoch 6 - iter 267/893 - loss 0.02709581 - time (sec): 152.28 - samples/sec: 483.70 - lr: 0.000084 - momentum: 0.000000
176
+ 2023-10-11 16:54:28,141 epoch 6 - iter 356/893 - loss 0.02801270 - time (sec): 202.75 - samples/sec: 486.61 - lr: 0.000082 - momentum: 0.000000
177
+ 2023-10-11 16:55:18,447 epoch 6 - iter 445/893 - loss 0.02788463 - time (sec): 253.05 - samples/sec: 486.04 - lr: 0.000080 - momentum: 0.000000
178
+ 2023-10-11 16:56:09,874 epoch 6 - iter 534/893 - loss 0.02959514 - time (sec): 304.48 - samples/sec: 487.32 - lr: 0.000078 - momentum: 0.000000
179
+ 2023-10-11 16:57:00,389 epoch 6 - iter 623/893 - loss 0.02989629 - time (sec): 355.00 - samples/sec: 488.21 - lr: 0.000077 - momentum: 0.000000
180
+ 2023-10-11 16:57:51,530 epoch 6 - iter 712/893 - loss 0.02978443 - time (sec): 406.14 - samples/sec: 487.85 - lr: 0.000075 - momentum: 0.000000
181
+ 2023-10-11 16:58:41,657 epoch 6 - iter 801/893 - loss 0.02942788 - time (sec): 456.26 - samples/sec: 488.38 - lr: 0.000073 - momentum: 0.000000
182
+ 2023-10-11 16:59:32,962 epoch 6 - iter 890/893 - loss 0.02909718 - time (sec): 507.57 - samples/sec: 488.53 - lr: 0.000071 - momentum: 0.000000
183
+ 2023-10-11 16:59:34,505 ----------------------------------------------------------------------------------------------------
184
+ 2023-10-11 16:59:34,505 EPOCH 6 done: loss 0.0292 - lr: 0.000071
185
+ 2023-10-11 16:59:56,468 DEV : loss 0.16406482458114624 - f1-score (micro avg) 0.7936
186
+ 2023-10-11 16:59:56,500 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-11 17:00:44,851 epoch 7 - iter 89/893 - loss 0.01961977 - time (sec): 48.35 - samples/sec: 500.18 - lr: 0.000069 - momentum: 0.000000
188
+ 2023-10-11 17:01:34,295 epoch 7 - iter 178/893 - loss 0.02220619 - time (sec): 97.79 - samples/sec: 498.70 - lr: 0.000068 - momentum: 0.000000
189
+ 2023-10-11 17:02:22,358 epoch 7 - iter 267/893 - loss 0.02475739 - time (sec): 145.86 - samples/sec: 499.63 - lr: 0.000066 - momentum: 0.000000
190
+ 2023-10-11 17:03:11,495 epoch 7 - iter 356/893 - loss 0.02271444 - time (sec): 194.99 - samples/sec: 505.50 - lr: 0.000064 - momentum: 0.000000
191
+ 2023-10-11 17:04:00,355 epoch 7 - iter 445/893 - loss 0.02294087 - time (sec): 243.85 - samples/sec: 508.07 - lr: 0.000062 - momentum: 0.000000
192
+ 2023-10-11 17:04:49,325 epoch 7 - iter 534/893 - loss 0.02184486 - time (sec): 292.82 - samples/sec: 507.52 - lr: 0.000061 - momentum: 0.000000
193
+ 2023-10-11 17:05:38,345 epoch 7 - iter 623/893 - loss 0.02120379 - time (sec): 341.84 - samples/sec: 506.40 - lr: 0.000059 - momentum: 0.000000
194
+ 2023-10-11 17:06:29,178 epoch 7 - iter 712/893 - loss 0.02183594 - time (sec): 392.68 - samples/sec: 505.93 - lr: 0.000057 - momentum: 0.000000
195
+ 2023-10-11 17:07:20,003 epoch 7 - iter 801/893 - loss 0.02155659 - time (sec): 443.50 - samples/sec: 504.30 - lr: 0.000055 - momentum: 0.000000
196
+ 2023-10-11 17:08:08,594 epoch 7 - iter 890/893 - loss 0.02139749 - time (sec): 492.09 - samples/sec: 503.21 - lr: 0.000053 - momentum: 0.000000
197
+ 2023-10-11 17:08:10,436 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-11 17:08:10,436 EPOCH 7 done: loss 0.0213 - lr: 0.000053
199
+ 2023-10-11 17:08:31,770 DEV : loss 0.17885611951351166 - f1-score (micro avg) 0.8011
200
+ 2023-10-11 17:08:31,801 saving best model
201
+ 2023-10-11 17:08:34,366 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-11 17:09:24,407 epoch 8 - iter 89/893 - loss 0.01715620 - time (sec): 50.04 - samples/sec: 493.87 - lr: 0.000052 - momentum: 0.000000
203
+ 2023-10-11 17:10:16,482 epoch 8 - iter 178/893 - loss 0.02006197 - time (sec): 102.11 - samples/sec: 494.07 - lr: 0.000050 - momentum: 0.000000
204
+ 2023-10-11 17:11:06,885 epoch 8 - iter 267/893 - loss 0.01738780 - time (sec): 152.52 - samples/sec: 494.04 - lr: 0.000048 - momentum: 0.000000
205
+ 2023-10-11 17:11:57,243 epoch 8 - iter 356/893 - loss 0.01811409 - time (sec): 202.87 - samples/sec: 494.75 - lr: 0.000046 - momentum: 0.000000
206
+ 2023-10-11 17:12:46,260 epoch 8 - iter 445/893 - loss 0.01897415 - time (sec): 251.89 - samples/sec: 493.42 - lr: 0.000045 - momentum: 0.000000
207
+ 2023-10-11 17:13:35,565 epoch 8 - iter 534/893 - loss 0.01805358 - time (sec): 301.20 - samples/sec: 493.49 - lr: 0.000043 - momentum: 0.000000
208
+ 2023-10-11 17:14:26,683 epoch 8 - iter 623/893 - loss 0.01824633 - time (sec): 352.31 - samples/sec: 494.18 - lr: 0.000041 - momentum: 0.000000
209
+ 2023-10-11 17:15:14,708 epoch 8 - iter 712/893 - loss 0.01811871 - time (sec): 400.34 - samples/sec: 491.30 - lr: 0.000039 - momentum: 0.000000
210
+ 2023-10-11 17:16:04,826 epoch 8 - iter 801/893 - loss 0.01785414 - time (sec): 450.46 - samples/sec: 494.17 - lr: 0.000037 - momentum: 0.000000
211
+ 2023-10-11 17:16:55,097 epoch 8 - iter 890/893 - loss 0.01770565 - time (sec): 500.73 - samples/sec: 495.42 - lr: 0.000036 - momentum: 0.000000
212
+ 2023-10-11 17:16:56,588 ----------------------------------------------------------------------------------------------------
213
+ 2023-10-11 17:16:56,588 EPOCH 8 done: loss 0.0177 - lr: 0.000036
214
+ 2023-10-11 17:17:18,690 DEV : loss 0.19161181151866913 - f1-score (micro avg) 0.7955
215
+ 2023-10-11 17:17:18,720 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-11 17:18:06,228 epoch 9 - iter 89/893 - loss 0.01298334 - time (sec): 47.51 - samples/sec: 497.93 - lr: 0.000034 - momentum: 0.000000
217
+ 2023-10-11 17:18:55,387 epoch 9 - iter 178/893 - loss 0.01537760 - time (sec): 96.67 - samples/sec: 511.42 - lr: 0.000032 - momentum: 0.000000
218
+ 2023-10-11 17:19:45,056 epoch 9 - iter 267/893 - loss 0.01528293 - time (sec): 146.33 - samples/sec: 517.74 - lr: 0.000030 - momentum: 0.000000
219
+ 2023-10-11 17:20:34,247 epoch 9 - iter 356/893 - loss 0.01453369 - time (sec): 195.53 - samples/sec: 514.55 - lr: 0.000029 - momentum: 0.000000
220
+ 2023-10-11 17:21:22,221 epoch 9 - iter 445/893 - loss 0.01412119 - time (sec): 243.50 - samples/sec: 512.37 - lr: 0.000027 - momentum: 0.000000
221
+ 2023-10-11 17:22:10,404 epoch 9 - iter 534/893 - loss 0.01411537 - time (sec): 291.68 - samples/sec: 513.33 - lr: 0.000025 - momentum: 0.000000
222
+ 2023-10-11 17:22:58,710 epoch 9 - iter 623/893 - loss 0.01468822 - time (sec): 339.99 - samples/sec: 511.65 - lr: 0.000023 - momentum: 0.000000
223
+ 2023-10-11 17:23:46,894 epoch 9 - iter 712/893 - loss 0.01427033 - time (sec): 388.17 - samples/sec: 510.66 - lr: 0.000022 - momentum: 0.000000
224
+ 2023-10-11 17:24:35,059 epoch 9 - iter 801/893 - loss 0.01423814 - time (sec): 436.34 - samples/sec: 509.86 - lr: 0.000020 - momentum: 0.000000
225
+ 2023-10-11 17:25:24,567 epoch 9 - iter 890/893 - loss 0.01466708 - time (sec): 485.85 - samples/sec: 510.22 - lr: 0.000018 - momentum: 0.000000
226
+ 2023-10-11 17:25:26,215 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-11 17:25:26,216 EPOCH 9 done: loss 0.0146 - lr: 0.000018
228
+ 2023-10-11 17:25:46,751 DEV : loss 0.19483081996440887 - f1-score (micro avg) 0.7944
229
+ 2023-10-11 17:25:46,781 ----------------------------------------------------------------------------------------------------
230
+ 2023-10-11 17:26:35,157 epoch 10 - iter 89/893 - loss 0.01073313 - time (sec): 48.37 - samples/sec: 520.32 - lr: 0.000016 - momentum: 0.000000
231
+ 2023-10-11 17:27:25,428 epoch 10 - iter 178/893 - loss 0.01013166 - time (sec): 98.65 - samples/sec: 519.71 - lr: 0.000014 - momentum: 0.000000
232
+ 2023-10-11 17:28:12,901 epoch 10 - iter 267/893 - loss 0.01112463 - time (sec): 146.12 - samples/sec: 510.31 - lr: 0.000013 - momentum: 0.000000
233
+ 2023-10-11 17:29:01,976 epoch 10 - iter 356/893 - loss 0.01098189 - time (sec): 195.19 - samples/sec: 508.58 - lr: 0.000011 - momentum: 0.000000
234
+ 2023-10-11 17:29:49,819 epoch 10 - iter 445/893 - loss 0.01066311 - time (sec): 243.04 - samples/sec: 503.61 - lr: 0.000009 - momentum: 0.000000
235
+ 2023-10-11 17:30:40,164 epoch 10 - iter 534/893 - loss 0.01119657 - time (sec): 293.38 - samples/sec: 508.28 - lr: 0.000007 - momentum: 0.000000
236
+ 2023-10-11 17:31:28,238 epoch 10 - iter 623/893 - loss 0.01162212 - time (sec): 341.45 - samples/sec: 506.25 - lr: 0.000006 - momentum: 0.000000
237
+ 2023-10-11 17:32:17,725 epoch 10 - iter 712/893 - loss 0.01182235 - time (sec): 390.94 - samples/sec: 506.41 - lr: 0.000004 - momentum: 0.000000
238
+ 2023-10-11 17:33:08,012 epoch 10 - iter 801/893 - loss 0.01144574 - time (sec): 441.23 - samples/sec: 506.35 - lr: 0.000002 - momentum: 0.000000
239
+ 2023-10-11 17:33:58,609 epoch 10 - iter 890/893 - loss 0.01158836 - time (sec): 491.83 - samples/sec: 504.65 - lr: 0.000000 - momentum: 0.000000
240
+ 2023-10-11 17:34:00,008 ----------------------------------------------------------------------------------------------------
241
+ 2023-10-11 17:34:00,008 EPOCH 10 done: loss 0.0116 - lr: 0.000000
242
+ 2023-10-11 17:34:20,889 DEV : loss 0.19962604343891144 - f1-score (micro avg) 0.7897
243
+ 2023-10-11 17:34:21,786 ----------------------------------------------------------------------------------------------------
244
+ 2023-10-11 17:34:21,788 Loading model from best epoch ...
245
+ 2023-10-11 17:34:26,562 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
246
+ 2023-10-11 17:35:36,469
247
+ Results:
248
+ - F-score (micro) 0.6917
249
+ - F-score (macro) 0.5961
250
+ - Accuracy 0.545
251
+
252
+ By class:
253
+ precision recall f1-score support
254
+
255
+ LOC 0.7070 0.6986 0.7028 1095
256
+ PER 0.7745 0.7806 0.7776 1012
257
+ ORG 0.4204 0.5770 0.4864 357
258
+ HumanProd 0.3276 0.5758 0.4176 33
259
+
260
+ micro avg 0.6717 0.7129 0.6917 2497
261
+ macro avg 0.5574 0.6580 0.5961 2497
262
+ weighted avg 0.6884 0.7129 0.6984 2497
263
+
264
+ 2023-10-11 17:35:36,470 ----------------------------------------------------------------------------------------------------