Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
677a215
•
1 Parent(s): e18a70f

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. test.tsv +0 -0
  5. training.log +240 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0d062c937b682ecca27f225487e0a3da7c62435e272aeae82a7821245d9f278
3
+ size 443323527
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 15:13:30 0.0000 0.4199 0.2364 0.1757 0.7443 0.2843 0.1667
3
+ 2 15:17:51 0.0000 0.1695 0.1622 0.2839 0.5909 0.3835 0.2385
4
+ 3 15:22:10 0.0000 0.1159 0.3360 0.2306 0.6307 0.3377 0.2044
5
+ 4 15:26:31 0.0000 0.0828 0.2833 0.2595 0.4280 0.3231 0.1940
6
+ 5 15:30:54 0.0000 0.0588 0.3179 0.2913 0.5682 0.3851 0.2392
7
+ 6 15:35:16 0.0000 0.0413 0.2984 0.3144 0.5549 0.4014 0.2524
8
+ 7 15:39:42 0.0000 0.0311 0.4335 0.2777 0.5701 0.3734 0.2303
9
+ 8 15:44:06 0.0000 0.0223 0.4504 0.2674 0.6402 0.3772 0.2336
10
+ 9 15:48:29 0.0000 0.0152 0.4913 0.2599 0.6364 0.3690 0.2272
11
+ 10 15:52:52 0.0000 0.0111 0.4786 0.2720 0.6155 0.3772 0.2333
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 15:09:10,345 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 15:09:10,346 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 768)
7
+ (position_embeddings): Embedding(512, 768)
8
+ (token_type_embeddings): Embedding(2, 768)
9
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=768, out_features=768, bias=True)
18
+ (key): Linear(in_features=768, out_features=768, bias=True)
19
+ (value): Linear(in_features=768, out_features=768, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=768, out_features=768, bias=True)
24
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
34
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=768, out_features=768, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=768, out_features=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 15:09:10,347 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
52
+ - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
53
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 15:09:10,347 Train: 20847 sentences
55
+ 2023-10-15 15:09:10,347 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 15:09:10,347 Training Params:
58
+ 2023-10-15 15:09:10,347 - learning_rate: "3e-05"
59
+ 2023-10-15 15:09:10,347 - mini_batch_size: "4"
60
+ 2023-10-15 15:09:10,347 - max_epochs: "10"
61
+ 2023-10-15 15:09:10,347 - shuffle: "True"
62
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 15:09:10,347 Plugins:
64
+ 2023-10-15 15:09:10,347 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 15:09:10,347 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 15:09:10,347 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 15:09:10,347 Computation:
70
+ 2023-10-15 15:09:10,347 - compute on device: cuda:0
71
+ 2023-10-15 15:09:10,347 - embedding storage: none
72
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 15:09:10,347 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 15:09:10,347 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 15:09:36,345 epoch 1 - iter 521/5212 - loss 1.70001218 - time (sec): 26.00 - samples/sec: 1455.46 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-15 15:10:02,999 epoch 1 - iter 1042/5212 - loss 1.06814573 - time (sec): 52.65 - samples/sec: 1418.65 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-15 15:10:29,184 epoch 1 - iter 1563/5212 - loss 0.80933221 - time (sec): 78.84 - samples/sec: 1453.59 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-15 15:10:54,029 epoch 1 - iter 2084/5212 - loss 0.69069653 - time (sec): 103.68 - samples/sec: 1449.58 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-15 15:11:18,988 epoch 1 - iter 2605/5212 - loss 0.60440945 - time (sec): 128.64 - samples/sec: 1454.63 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-15 15:11:43,958 epoch 1 - iter 3126/5212 - loss 0.54666260 - time (sec): 153.61 - samples/sec: 1454.40 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-15 15:12:09,304 epoch 1 - iter 3647/5212 - loss 0.50222284 - time (sec): 178.96 - samples/sec: 1447.75 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-15 15:12:34,296 epoch 1 - iter 4168/5212 - loss 0.47277877 - time (sec): 203.95 - samples/sec: 1441.16 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-15 15:12:59,391 epoch 1 - iter 4689/5212 - loss 0.44260534 - time (sec): 229.04 - samples/sec: 1444.12 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-15 15:13:25,052 epoch 1 - iter 5210/5212 - loss 0.41994283 - time (sec): 254.70 - samples/sec: 1442.37 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-15 15:13:25,147 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 15:13:25,148 EPOCH 1 done: loss 0.4199 - lr: 0.000030
88
+ 2023-10-15 15:13:30,935 DEV : loss 0.2363644540309906 - f1-score (micro avg) 0.2843
89
+ 2023-10-15 15:13:30,961 saving best model
90
+ 2023-10-15 15:13:31,321 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 15:13:56,373 epoch 2 - iter 521/5212 - loss 0.17672925 - time (sec): 25.05 - samples/sec: 1403.82 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-15 15:14:21,203 epoch 2 - iter 1042/5212 - loss 0.16983337 - time (sec): 49.88 - samples/sec: 1429.70 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-15 15:14:46,722 epoch 2 - iter 1563/5212 - loss 0.18368285 - time (sec): 75.40 - samples/sec: 1445.90 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-15 15:15:11,777 epoch 2 - iter 2084/5212 - loss 0.17905217 - time (sec): 100.45 - samples/sec: 1443.45 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-15 15:15:36,664 epoch 2 - iter 2605/5212 - loss 0.17232351 - time (sec): 125.34 - samples/sec: 1455.15 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-15 15:16:02,030 epoch 2 - iter 3126/5212 - loss 0.17048500 - time (sec): 150.71 - samples/sec: 1460.32 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-15 15:16:26,925 epoch 2 - iter 3647/5212 - loss 0.17267205 - time (sec): 175.60 - samples/sec: 1461.13 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-15 15:16:52,295 epoch 2 - iter 4168/5212 - loss 0.17077349 - time (sec): 200.97 - samples/sec: 1466.42 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-15 15:17:17,453 epoch 2 - iter 4689/5212 - loss 0.17105767 - time (sec): 226.13 - samples/sec: 1464.23 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-15 15:17:42,380 epoch 2 - iter 5210/5212 - loss 0.16953223 - time (sec): 251.06 - samples/sec: 1462.79 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-15 15:17:42,472 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 15:17:42,472 EPOCH 2 done: loss 0.1695 - lr: 0.000027
103
+ 2023-10-15 15:17:51,427 DEV : loss 0.16215363144874573 - f1-score (micro avg) 0.3835
104
+ 2023-10-15 15:17:51,453 saving best model
105
+ 2023-10-15 15:17:51,895 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 15:18:17,980 epoch 3 - iter 521/5212 - loss 0.10469666 - time (sec): 26.08 - samples/sec: 1494.63 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-15 15:18:43,907 epoch 3 - iter 1042/5212 - loss 0.11123288 - time (sec): 52.01 - samples/sec: 1491.14 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-15 15:19:08,918 epoch 3 - iter 1563/5212 - loss 0.11762357 - time (sec): 77.02 - samples/sec: 1491.41 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-15 15:19:33,852 epoch 3 - iter 2084/5212 - loss 0.12270683 - time (sec): 101.95 - samples/sec: 1475.35 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-15 15:19:59,052 epoch 3 - iter 2605/5212 - loss 0.12115573 - time (sec): 127.16 - samples/sec: 1483.64 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-15 15:20:23,967 epoch 3 - iter 3126/5212 - loss 0.12016284 - time (sec): 152.07 - samples/sec: 1467.39 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-15 15:20:49,519 epoch 3 - iter 3647/5212 - loss 0.12039286 - time (sec): 177.62 - samples/sec: 1453.89 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-15 15:21:13,549 epoch 3 - iter 4168/5212 - loss 0.11951394 - time (sec): 201.65 - samples/sec: 1453.79 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-15 15:21:37,889 epoch 3 - iter 4689/5212 - loss 0.11769963 - time (sec): 225.99 - samples/sec: 1464.75 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-15 15:22:01,759 epoch 3 - iter 5210/5212 - loss 0.11590928 - time (sec): 249.86 - samples/sec: 1470.09 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-15 15:22:01,847 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 15:22:01,847 EPOCH 3 done: loss 0.1159 - lr: 0.000023
118
+ 2023-10-15 15:22:10,885 DEV : loss 0.3360123932361603 - f1-score (micro avg) 0.3377
119
+ 2023-10-15 15:22:10,912 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 15:22:36,517 epoch 4 - iter 521/5212 - loss 0.09249251 - time (sec): 25.60 - samples/sec: 1518.66 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-15 15:23:01,241 epoch 4 - iter 1042/5212 - loss 0.09120710 - time (sec): 50.33 - samples/sec: 1442.37 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-15 15:23:26,282 epoch 4 - iter 1563/5212 - loss 0.08893166 - time (sec): 75.37 - samples/sec: 1456.98 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-15 15:23:50,937 epoch 4 - iter 2084/5212 - loss 0.08371995 - time (sec): 100.02 - samples/sec: 1452.06 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-15 15:24:15,942 epoch 4 - iter 2605/5212 - loss 0.08118124 - time (sec): 125.03 - samples/sec: 1456.63 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-15 15:24:41,451 epoch 4 - iter 3126/5212 - loss 0.08066818 - time (sec): 150.54 - samples/sec: 1459.68 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-15 15:25:06,504 epoch 4 - iter 3647/5212 - loss 0.08362064 - time (sec): 175.59 - samples/sec: 1457.66 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-15 15:25:31,596 epoch 4 - iter 4168/5212 - loss 0.08316214 - time (sec): 200.68 - samples/sec: 1455.02 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-15 15:25:57,067 epoch 4 - iter 4689/5212 - loss 0.08345938 - time (sec): 226.15 - samples/sec: 1457.65 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-15 15:26:22,539 epoch 4 - iter 5210/5212 - loss 0.08274423 - time (sec): 251.63 - samples/sec: 1459.97 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-15 15:26:22,628 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 15:26:22,628 EPOCH 4 done: loss 0.0828 - lr: 0.000020
132
+ 2023-10-15 15:26:31,008 DEV : loss 0.28332096338272095 - f1-score (micro avg) 0.3231
133
+ 2023-10-15 15:26:31,039 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-15 15:26:57,145 epoch 5 - iter 521/5212 - loss 0.05268117 - time (sec): 26.11 - samples/sec: 1408.81 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-15 15:27:23,828 epoch 5 - iter 1042/5212 - loss 0.05092813 - time (sec): 52.79 - samples/sec: 1481.48 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-15 15:27:49,128 epoch 5 - iter 1563/5212 - loss 0.05327142 - time (sec): 78.09 - samples/sec: 1456.97 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-15 15:28:15,064 epoch 5 - iter 2084/5212 - loss 0.05794021 - time (sec): 104.02 - samples/sec: 1466.54 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-15 15:28:40,167 epoch 5 - iter 2605/5212 - loss 0.05639404 - time (sec): 129.13 - samples/sec: 1456.04 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-15 15:29:05,571 epoch 5 - iter 3126/5212 - loss 0.05672885 - time (sec): 154.53 - samples/sec: 1464.34 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-15 15:29:30,847 epoch 5 - iter 3647/5212 - loss 0.05616822 - time (sec): 179.81 - samples/sec: 1451.42 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-15 15:29:55,838 epoch 5 - iter 4168/5212 - loss 0.05709755 - time (sec): 204.80 - samples/sec: 1446.81 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-15 15:30:21,191 epoch 5 - iter 4689/5212 - loss 0.05865289 - time (sec): 230.15 - samples/sec: 1451.81 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-15 15:30:45,876 epoch 5 - iter 5210/5212 - loss 0.05877178 - time (sec): 254.84 - samples/sec: 1441.30 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-15 15:30:45,974 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-15 15:30:45,974 EPOCH 5 done: loss 0.0588 - lr: 0.000017
146
+ 2023-10-15 15:30:54,204 DEV : loss 0.3178791105747223 - f1-score (micro avg) 0.3851
147
+ 2023-10-15 15:30:54,233 saving best model
148
+ 2023-10-15 15:30:54,606 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-15 15:31:20,370 epoch 6 - iter 521/5212 - loss 0.03111639 - time (sec): 25.76 - samples/sec: 1497.34 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-15 15:31:45,536 epoch 6 - iter 1042/5212 - loss 0.04174411 - time (sec): 50.93 - samples/sec: 1462.79 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-15 15:32:10,834 epoch 6 - iter 1563/5212 - loss 0.03928689 - time (sec): 76.23 - samples/sec: 1467.34 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-15 15:32:36,518 epoch 6 - iter 2084/5212 - loss 0.04056787 - time (sec): 101.91 - samples/sec: 1437.64 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-15 15:33:01,811 epoch 6 - iter 2605/5212 - loss 0.03954421 - time (sec): 127.20 - samples/sec: 1446.65 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-15 15:33:27,381 epoch 6 - iter 3126/5212 - loss 0.04043613 - time (sec): 152.77 - samples/sec: 1450.98 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-15 15:33:53,366 epoch 6 - iter 3647/5212 - loss 0.04034187 - time (sec): 178.76 - samples/sec: 1465.42 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-15 15:34:18,104 epoch 6 - iter 4168/5212 - loss 0.04068735 - time (sec): 203.50 - samples/sec: 1453.71 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-15 15:34:42,838 epoch 6 - iter 4689/5212 - loss 0.04106378 - time (sec): 228.23 - samples/sec: 1454.69 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-15 15:35:07,496 epoch 6 - iter 5210/5212 - loss 0.04133139 - time (sec): 252.89 - samples/sec: 1452.72 - lr: 0.000013 - momentum: 0.000000
159
+ 2023-10-15 15:35:07,587 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-15 15:35:07,587 EPOCH 6 done: loss 0.0413 - lr: 0.000013
161
+ 2023-10-15 15:35:16,109 DEV : loss 0.298368901014328 - f1-score (micro avg) 0.4014
162
+ 2023-10-15 15:35:16,143 saving best model
163
+ 2023-10-15 15:35:16,629 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-15 15:35:41,719 epoch 7 - iter 521/5212 - loss 0.02768974 - time (sec): 25.09 - samples/sec: 1345.24 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-15 15:36:06,868 epoch 7 - iter 1042/5212 - loss 0.03254404 - time (sec): 50.24 - samples/sec: 1382.74 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-15 15:36:33,160 epoch 7 - iter 1563/5212 - loss 0.03323169 - time (sec): 76.53 - samples/sec: 1414.01 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-15 15:36:58,282 epoch 7 - iter 2084/5212 - loss 0.03438222 - time (sec): 101.65 - samples/sec: 1407.75 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-15 15:37:23,204 epoch 7 - iter 2605/5212 - loss 0.03394163 - time (sec): 126.57 - samples/sec: 1399.38 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-15 15:37:48,331 epoch 7 - iter 3126/5212 - loss 0.03392244 - time (sec): 151.70 - samples/sec: 1411.36 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-15 15:38:14,730 epoch 7 - iter 3647/5212 - loss 0.03283761 - time (sec): 178.10 - samples/sec: 1410.94 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-15 15:38:41,408 epoch 7 - iter 4168/5212 - loss 0.03197315 - time (sec): 204.78 - samples/sec: 1418.09 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-15 15:39:07,470 epoch 7 - iter 4689/5212 - loss 0.03103844 - time (sec): 230.84 - samples/sec: 1426.21 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-15 15:39:34,028 epoch 7 - iter 5210/5212 - loss 0.03110437 - time (sec): 257.40 - samples/sec: 1426.17 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-15 15:39:34,151 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-15 15:39:34,152 EPOCH 7 done: loss 0.0311 - lr: 0.000010
176
+ 2023-10-15 15:39:42,426 DEV : loss 0.43352431058883667 - f1-score (micro avg) 0.3734
177
+ 2023-10-15 15:39:42,457 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-15 15:40:08,079 epoch 8 - iter 521/5212 - loss 0.01599032 - time (sec): 25.62 - samples/sec: 1493.11 - lr: 0.000010 - momentum: 0.000000
179
+ 2023-10-15 15:40:33,449 epoch 8 - iter 1042/5212 - loss 0.01861222 - time (sec): 50.99 - samples/sec: 1472.02 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-15 15:40:58,531 epoch 8 - iter 1563/5212 - loss 0.02102450 - time (sec): 76.07 - samples/sec: 1430.47 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-15 15:41:23,941 epoch 8 - iter 2084/5212 - loss 0.02070202 - time (sec): 101.48 - samples/sec: 1429.28 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-15 15:41:49,399 epoch 8 - iter 2605/5212 - loss 0.02112981 - time (sec): 126.94 - samples/sec: 1410.72 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-15 15:42:14,841 epoch 8 - iter 3126/5212 - loss 0.02245286 - time (sec): 152.38 - samples/sec: 1417.79 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-15 15:42:40,962 epoch 8 - iter 3647/5212 - loss 0.02181557 - time (sec): 178.50 - samples/sec: 1437.19 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-15 15:43:06,269 epoch 8 - iter 4168/5212 - loss 0.02303020 - time (sec): 203.81 - samples/sec: 1445.16 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-15 15:43:31,410 epoch 8 - iter 4689/5212 - loss 0.02258626 - time (sec): 228.95 - samples/sec: 1441.42 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-15 15:43:57,192 epoch 8 - iter 5210/5212 - loss 0.02227638 - time (sec): 254.73 - samples/sec: 1441.65 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-15 15:43:57,287 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-15 15:43:57,287 EPOCH 8 done: loss 0.0223 - lr: 0.000007
190
+ 2023-10-15 15:44:06,531 DEV : loss 0.450359582901001 - f1-score (micro avg) 0.3772
191
+ 2023-10-15 15:44:06,563 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-15 15:44:31,209 epoch 9 - iter 521/5212 - loss 0.01562105 - time (sec): 24.64 - samples/sec: 1407.65 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-15 15:44:55,932 epoch 9 - iter 1042/5212 - loss 0.01453439 - time (sec): 49.37 - samples/sec: 1388.63 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-15 15:45:21,936 epoch 9 - iter 1563/5212 - loss 0.01536925 - time (sec): 75.37 - samples/sec: 1437.22 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-15 15:45:47,163 epoch 9 - iter 2084/5212 - loss 0.01490582 - time (sec): 100.60 - samples/sec: 1446.45 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-15 15:46:12,767 epoch 9 - iter 2605/5212 - loss 0.01487153 - time (sec): 126.20 - samples/sec: 1454.60 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-15 15:46:38,173 epoch 9 - iter 3126/5212 - loss 0.01533863 - time (sec): 151.61 - samples/sec: 1451.73 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-15 15:47:03,532 epoch 9 - iter 3647/5212 - loss 0.01597661 - time (sec): 176.97 - samples/sec: 1455.60 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-15 15:47:28,771 epoch 9 - iter 4168/5212 - loss 0.01580615 - time (sec): 202.21 - samples/sec: 1456.29 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-15 15:47:54,354 epoch 9 - iter 4689/5212 - loss 0.01547589 - time (sec): 227.79 - samples/sec: 1454.52 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-15 15:48:19,499 epoch 9 - iter 5210/5212 - loss 0.01521545 - time (sec): 252.93 - samples/sec: 1452.41 - lr: 0.000003 - momentum: 0.000000
202
+ 2023-10-15 15:48:19,584 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-15 15:48:19,584 EPOCH 9 done: loss 0.0152 - lr: 0.000003
204
+ 2023-10-15 15:48:28,971 DEV : loss 0.4913356900215149 - f1-score (micro avg) 0.369
205
+ 2023-10-15 15:48:29,002 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-15 15:48:55,241 epoch 10 - iter 521/5212 - loss 0.00783394 - time (sec): 26.24 - samples/sec: 1417.76 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-15 15:49:20,769 epoch 10 - iter 1042/5212 - loss 0.01120969 - time (sec): 51.77 - samples/sec: 1427.38 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-15 15:49:46,662 epoch 10 - iter 1563/5212 - loss 0.00991988 - time (sec): 77.66 - samples/sec: 1460.85 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-15 15:50:11,819 epoch 10 - iter 2084/5212 - loss 0.00947631 - time (sec): 102.82 - samples/sec: 1461.11 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-15 15:50:36,250 epoch 10 - iter 2605/5212 - loss 0.01010491 - time (sec): 127.25 - samples/sec: 1450.05 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-15 15:51:01,852 epoch 10 - iter 3126/5212 - loss 0.01027554 - time (sec): 152.85 - samples/sec: 1444.89 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-15 15:51:27,567 epoch 10 - iter 3647/5212 - loss 0.01132153 - time (sec): 178.56 - samples/sec: 1448.36 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 15:51:52,655 epoch 10 - iter 4168/5212 - loss 0.01087304 - time (sec): 203.65 - samples/sec: 1443.38 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-15 15:52:17,989 epoch 10 - iter 4689/5212 - loss 0.01102544 - time (sec): 228.99 - samples/sec: 1443.50 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-15 15:52:42,744 epoch 10 - iter 5210/5212 - loss 0.01114682 - time (sec): 253.74 - samples/sec: 1447.96 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-15 15:52:42,831 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-15 15:52:42,832 EPOCH 10 done: loss 0.0111 - lr: 0.000000
218
+ 2023-10-15 15:52:52,269 DEV : loss 0.47859835624694824 - f1-score (micro avg) 0.3772
219
+ 2023-10-15 15:52:52,722 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-15 15:52:52,724 Loading model from best epoch ...
221
+ 2023-10-15 15:52:54,374 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
222
+ 2023-10-15 15:53:11,590
223
+ Results:
224
+ - F-score (micro) 0.4759
225
+ - F-score (macro) 0.3146
226
+ - Accuracy 0.3163
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ LOC 0.5790 0.5947 0.5868 1214
232
+ PER 0.4096 0.4121 0.4109 808
233
+ ORG 0.2527 0.2691 0.2606 353
234
+ HumanProd 0.0000 0.0000 0.0000 15
235
+
236
+ micro avg 0.4707 0.4812 0.4759 2390
237
+ macro avg 0.3103 0.3190 0.3146 2390
238
+ weighted avg 0.4699 0.4812 0.4754 2390
239
+
240
+ 2023-10-15 15:53:11,590 ----------------------------------------------------------------------------------------------------