Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
e2029cb
•
1 Parent(s): 3f4135a

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 +242 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30a293547cead44c8b0f0311e7c68ba69cf0d298744492787acea86968e4d492
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 17:15:24 0.0000 0.3652 0.1521 0.2066 0.4886 0.2904 0.1704
3
+ 2 17:18:42 0.0000 0.1580 0.1702 0.2334 0.4867 0.3155 0.1883
4
+ 3 17:22:01 0.0000 0.1081 0.1856 0.2389 0.6439 0.3485 0.2128
5
+ 4 17:25:24 0.0000 0.0813 0.2633 0.2839 0.5947 0.3843 0.2384
6
+ 5 17:28:45 0.0000 0.0596 0.2743 0.2567 0.5587 0.3518 0.2147
7
+ 6 17:32:04 0.0000 0.0432 0.3392 0.2976 0.5777 0.3928 0.2460
8
+ 7 17:35:25 0.0000 0.0343 0.3687 0.2686 0.6098 0.3729 0.2302
9
+ 8 17:38:45 0.0000 0.0232 0.4458 0.2606 0.6420 0.3707 0.2291
10
+ 9 17:42:05 0.0000 0.0156 0.4665 0.2704 0.6326 0.3789 0.2350
11
+ 10 17:45:26 0.0000 0.0102 0.4450 0.2920 0.6061 0.3941 0.2467
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 17:12:09,248 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 17:12:09,249 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 17:12:09,249 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 17:12:09,249 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 17:12:09,249 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 17:12:09,249 Train: 20847 sentences
55
+ 2023-10-15 17:12:09,249 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 17:12:09,249 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 17:12:09,249 Training Params:
58
+ 2023-10-15 17:12:09,249 - learning_rate: "5e-05"
59
+ 2023-10-15 17:12:09,249 - mini_batch_size: "8"
60
+ 2023-10-15 17:12:09,249 - max_epochs: "10"
61
+ 2023-10-15 17:12:09,249 - shuffle: "True"
62
+ 2023-10-15 17:12:09,249 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 17:12:09,249 Plugins:
64
+ 2023-10-15 17:12:09,249 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 17:12:09,249 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 17:12:09,249 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 17:12:09,249 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 17:12:09,249 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 17:12:09,249 Computation:
70
+ 2023-10-15 17:12:09,249 - compute on device: cuda:0
71
+ 2023-10-15 17:12:09,249 - embedding storage: none
72
+ 2023-10-15 17:12:09,250 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 17:12:09,250 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-15 17:12:09,250 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 17:12:09,250 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 17:12:27,851 epoch 1 - iter 260/2606 - loss 1.50266157 - time (sec): 18.60 - samples/sec: 1874.61 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-15 17:12:46,100 epoch 1 - iter 520/2606 - loss 0.98485226 - time (sec): 36.85 - samples/sec: 1923.74 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-15 17:13:05,208 epoch 1 - iter 780/2606 - loss 0.74065196 - time (sec): 55.96 - samples/sec: 1931.59 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-15 17:13:24,603 epoch 1 - iter 1040/2606 - loss 0.61022759 - time (sec): 75.35 - samples/sec: 1930.01 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-15 17:13:43,557 epoch 1 - iter 1300/2606 - loss 0.53605747 - time (sec): 94.31 - samples/sec: 1935.21 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-15 17:14:03,940 epoch 1 - iter 1560/2606 - loss 0.47601742 - time (sec): 114.69 - samples/sec: 1936.25 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-15 17:14:22,308 epoch 1 - iter 1820/2606 - loss 0.43680600 - time (sec): 133.06 - samples/sec: 1943.35 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-15 17:14:40,531 epoch 1 - iter 2080/2606 - loss 0.40770913 - time (sec): 151.28 - samples/sec: 1949.84 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-15 17:14:58,591 epoch 1 - iter 2340/2606 - loss 0.38693475 - time (sec): 169.34 - samples/sec: 1941.41 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-15 17:15:18,047 epoch 1 - iter 2600/2606 - loss 0.36558100 - time (sec): 188.80 - samples/sec: 1942.74 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-15 17:15:18,426 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 17:15:18,426 EPOCH 1 done: loss 0.3652 - lr: 0.000050
88
+ 2023-10-15 17:15:24,131 DEV : loss 0.15214306116104126 - f1-score (micro avg) 0.2904
89
+ 2023-10-15 17:15:24,157 saving best model
90
+ 2023-10-15 17:15:24,554 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 17:15:42,495 epoch 2 - iter 260/2606 - loss 0.21425752 - time (sec): 17.94 - samples/sec: 1856.99 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-15 17:16:01,095 epoch 2 - iter 520/2606 - loss 0.18763868 - time (sec): 36.54 - samples/sec: 1924.91 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-15 17:16:19,351 epoch 2 - iter 780/2606 - loss 0.18203253 - time (sec): 54.80 - samples/sec: 1930.66 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-15 17:16:38,367 epoch 2 - iter 1040/2606 - loss 0.17096932 - time (sec): 73.81 - samples/sec: 1936.80 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-15 17:16:56,653 epoch 2 - iter 1300/2606 - loss 0.16919457 - time (sec): 92.10 - samples/sec: 1939.99 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-15 17:17:16,519 epoch 2 - iter 1560/2606 - loss 0.16758215 - time (sec): 111.96 - samples/sec: 1947.98 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-15 17:17:35,476 epoch 2 - iter 1820/2606 - loss 0.16614397 - time (sec): 130.92 - samples/sec: 1941.10 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-15 17:17:54,052 epoch 2 - iter 2080/2606 - loss 0.16420431 - time (sec): 149.50 - samples/sec: 1942.57 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-15 17:18:13,424 epoch 2 - iter 2340/2606 - loss 0.16007055 - time (sec): 168.87 - samples/sec: 1941.56 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-15 17:18:33,096 epoch 2 - iter 2600/2606 - loss 0.15826879 - time (sec): 188.54 - samples/sec: 1942.06 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-15 17:18:33,596 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 17:18:33,596 EPOCH 2 done: loss 0.1580 - lr: 0.000044
103
+ 2023-10-15 17:18:42,526 DEV : loss 0.1702265739440918 - f1-score (micro avg) 0.3155
104
+ 2023-10-15 17:18:42,552 saving best model
105
+ 2023-10-15 17:18:43,006 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 17:19:00,588 epoch 3 - iter 260/2606 - loss 0.11956974 - time (sec): 17.58 - samples/sec: 1906.64 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-15 17:19:18,723 epoch 3 - iter 520/2606 - loss 0.11388643 - time (sec): 35.72 - samples/sec: 1921.25 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-15 17:19:39,576 epoch 3 - iter 780/2606 - loss 0.10643473 - time (sec): 56.57 - samples/sec: 1944.70 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-15 17:19:59,157 epoch 3 - iter 1040/2606 - loss 0.10454978 - time (sec): 76.15 - samples/sec: 1936.60 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-15 17:20:18,167 epoch 3 - iter 1300/2606 - loss 0.10544035 - time (sec): 95.16 - samples/sec: 1939.11 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-15 17:20:36,446 epoch 3 - iter 1560/2606 - loss 0.10643284 - time (sec): 113.44 - samples/sec: 1946.85 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-15 17:20:55,437 epoch 3 - iter 1820/2606 - loss 0.10632511 - time (sec): 132.43 - samples/sec: 1944.58 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-15 17:21:14,086 epoch 3 - iter 2080/2606 - loss 0.10739551 - time (sec): 151.08 - samples/sec: 1936.58 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-15 17:21:32,654 epoch 3 - iter 2340/2606 - loss 0.10835971 - time (sec): 169.65 - samples/sec: 1935.80 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-15 17:21:52,442 epoch 3 - iter 2600/2606 - loss 0.10794681 - time (sec): 189.43 - samples/sec: 1935.52 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-15 17:21:52,809 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 17:21:52,809 EPOCH 3 done: loss 0.1081 - lr: 0.000039
118
+ 2023-10-15 17:22:01,270 DEV : loss 0.18563616275787354 - f1-score (micro avg) 0.3485
119
+ 2023-10-15 17:22:01,316 saving best model
120
+ 2023-10-15 17:22:03,535 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-15 17:22:22,669 epoch 4 - iter 260/2606 - loss 0.07658858 - time (sec): 19.13 - samples/sec: 1936.28 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-15 17:22:42,190 epoch 4 - iter 520/2606 - loss 0.08210827 - time (sec): 38.65 - samples/sec: 1943.33 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-15 17:23:01,175 epoch 4 - iter 780/2606 - loss 0.08206491 - time (sec): 57.64 - samples/sec: 1925.85 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-15 17:23:20,345 epoch 4 - iter 1040/2606 - loss 0.08053312 - time (sec): 76.81 - samples/sec: 1916.75 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-15 17:23:39,237 epoch 4 - iter 1300/2606 - loss 0.08024248 - time (sec): 95.70 - samples/sec: 1916.94 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-15 17:23:58,673 epoch 4 - iter 1560/2606 - loss 0.08189957 - time (sec): 115.13 - samples/sec: 1923.89 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-15 17:24:18,116 epoch 4 - iter 1820/2606 - loss 0.08169910 - time (sec): 134.58 - samples/sec: 1922.71 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-15 17:24:37,989 epoch 4 - iter 2080/2606 - loss 0.08278118 - time (sec): 154.45 - samples/sec: 1904.84 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-15 17:24:56,294 epoch 4 - iter 2340/2606 - loss 0.08182834 - time (sec): 172.76 - samples/sec: 1905.38 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-15 17:25:15,512 epoch 4 - iter 2600/2606 - loss 0.08143557 - time (sec): 191.97 - samples/sec: 1910.29 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-15 17:25:15,865 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-15 17:25:15,865 EPOCH 4 done: loss 0.0813 - lr: 0.000033
133
+ 2023-10-15 17:25:24,202 DEV : loss 0.2632576823234558 - f1-score (micro avg) 0.3843
134
+ 2023-10-15 17:25:24,237 saving best model
135
+ 2023-10-15 17:25:24,727 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-15 17:25:45,899 epoch 5 - iter 260/2606 - loss 0.05364757 - time (sec): 21.17 - samples/sec: 1882.52 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-15 17:26:04,238 epoch 5 - iter 520/2606 - loss 0.05447531 - time (sec): 39.51 - samples/sec: 1922.77 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-15 17:26:23,473 epoch 5 - iter 780/2606 - loss 0.05594421 - time (sec): 58.74 - samples/sec: 1892.98 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-15 17:26:43,477 epoch 5 - iter 1040/2606 - loss 0.05342593 - time (sec): 78.75 - samples/sec: 1903.69 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-15 17:27:02,033 epoch 5 - iter 1300/2606 - loss 0.05823903 - time (sec): 97.30 - samples/sec: 1920.69 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-15 17:27:20,738 epoch 5 - iter 1560/2606 - loss 0.05721454 - time (sec): 116.01 - samples/sec: 1913.24 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-15 17:27:39,173 epoch 5 - iter 1820/2606 - loss 0.05790703 - time (sec): 134.44 - samples/sec: 1906.47 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-15 17:27:58,531 epoch 5 - iter 2080/2606 - loss 0.05849359 - time (sec): 153.80 - samples/sec: 1908.55 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-15 17:28:17,095 epoch 5 - iter 2340/2606 - loss 0.05875878 - time (sec): 172.37 - samples/sec: 1914.33 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-15 17:28:36,319 epoch 5 - iter 2600/2606 - loss 0.05959147 - time (sec): 191.59 - samples/sec: 1913.86 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-15 17:28:36,730 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-15 17:28:36,730 EPOCH 5 done: loss 0.0596 - lr: 0.000028
148
+ 2023-10-15 17:28:44,981 DEV : loss 0.2743144631385803 - f1-score (micro avg) 0.3518
149
+ 2023-10-15 17:28:45,008 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-15 17:29:03,660 epoch 6 - iter 260/2606 - loss 0.03453990 - time (sec): 18.65 - samples/sec: 1920.29 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-15 17:29:22,853 epoch 6 - iter 520/2606 - loss 0.04058561 - time (sec): 37.84 - samples/sec: 1962.50 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-15 17:29:42,468 epoch 6 - iter 780/2606 - loss 0.04199312 - time (sec): 57.46 - samples/sec: 1951.09 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-15 17:30:01,376 epoch 6 - iter 1040/2606 - loss 0.04376216 - time (sec): 76.37 - samples/sec: 1942.05 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-15 17:30:19,228 epoch 6 - iter 1300/2606 - loss 0.04610481 - time (sec): 94.22 - samples/sec: 1931.70 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-15 17:30:40,254 epoch 6 - iter 1560/2606 - loss 0.04504652 - time (sec): 115.24 - samples/sec: 1915.81 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-15 17:30:58,796 epoch 6 - iter 1820/2606 - loss 0.04499603 - time (sec): 133.79 - samples/sec: 1915.50 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-15 17:31:18,595 epoch 6 - iter 2080/2606 - loss 0.04487564 - time (sec): 153.59 - samples/sec: 1912.99 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-15 17:31:37,219 epoch 6 - iter 2340/2606 - loss 0.04366420 - time (sec): 172.21 - samples/sec: 1917.65 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-15 17:31:55,917 epoch 6 - iter 2600/2606 - loss 0.04322832 - time (sec): 190.91 - samples/sec: 1918.66 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-15 17:31:56,494 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-15 17:31:56,494 EPOCH 6 done: loss 0.0432 - lr: 0.000022
162
+ 2023-10-15 17:32:04,745 DEV : loss 0.3391795754432678 - f1-score (micro avg) 0.3928
163
+ 2023-10-15 17:32:04,773 saving best model
164
+ 2023-10-15 17:32:05,262 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-15 17:32:23,978 epoch 7 - iter 260/2606 - loss 0.03643105 - time (sec): 18.71 - samples/sec: 1804.68 - lr: 0.000022 - momentum: 0.000000
166
+ 2023-10-15 17:32:43,242 epoch 7 - iter 520/2606 - loss 0.03438457 - time (sec): 37.98 - samples/sec: 1866.92 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-15 17:33:01,573 epoch 7 - iter 780/2606 - loss 0.03467784 - time (sec): 56.31 - samples/sec: 1857.31 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-15 17:33:20,953 epoch 7 - iter 1040/2606 - loss 0.03271919 - time (sec): 75.69 - samples/sec: 1881.72 - lr: 0.000020 - momentum: 0.000000
169
+ 2023-10-15 17:33:40,723 epoch 7 - iter 1300/2606 - loss 0.03405536 - time (sec): 95.46 - samples/sec: 1891.86 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-15 17:33:59,097 epoch 7 - iter 1560/2606 - loss 0.03411291 - time (sec): 113.83 - samples/sec: 1903.15 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-15 17:34:18,022 epoch 7 - iter 1820/2606 - loss 0.03486039 - time (sec): 132.76 - samples/sec: 1920.98 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-15 17:34:36,899 epoch 7 - iter 2080/2606 - loss 0.03442730 - time (sec): 151.64 - samples/sec: 1927.00 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-15 17:34:57,116 epoch 7 - iter 2340/2606 - loss 0.03334037 - time (sec): 171.85 - samples/sec: 1912.03 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-15 17:35:16,378 epoch 7 - iter 2600/2606 - loss 0.03422553 - time (sec): 191.11 - samples/sec: 1918.88 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-15 17:35:16,884 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-15 17:35:16,884 EPOCH 7 done: loss 0.0343 - lr: 0.000017
177
+ 2023-10-15 17:35:25,272 DEV : loss 0.3687325119972229 - f1-score (micro avg) 0.3729
178
+ 2023-10-15 17:35:25,317 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-15 17:35:45,736 epoch 8 - iter 260/2606 - loss 0.02369143 - time (sec): 20.42 - samples/sec: 1926.18 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-15 17:36:05,692 epoch 8 - iter 520/2606 - loss 0.02091381 - time (sec): 40.37 - samples/sec: 1915.24 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-15 17:36:25,235 epoch 8 - iter 780/2606 - loss 0.02276536 - time (sec): 59.92 - samples/sec: 1915.08 - lr: 0.000015 - momentum: 0.000000
182
+ 2023-10-15 17:36:43,567 epoch 8 - iter 1040/2606 - loss 0.02383309 - time (sec): 78.25 - samples/sec: 1913.80 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-15 17:37:02,312 epoch 8 - iter 1300/2606 - loss 0.02346249 - time (sec): 96.99 - samples/sec: 1907.47 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-15 17:37:21,147 epoch 8 - iter 1560/2606 - loss 0.02319510 - time (sec): 115.83 - samples/sec: 1908.88 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-15 17:37:40,866 epoch 8 - iter 1820/2606 - loss 0.02286996 - time (sec): 135.55 - samples/sec: 1909.35 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-15 17:37:59,177 epoch 8 - iter 2080/2606 - loss 0.02346536 - time (sec): 153.86 - samples/sec: 1906.92 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-15 17:38:18,146 epoch 8 - iter 2340/2606 - loss 0.02312509 - time (sec): 172.83 - samples/sec: 1909.04 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-15 17:38:36,335 epoch 8 - iter 2600/2606 - loss 0.02326902 - time (sec): 191.02 - samples/sec: 1916.84 - lr: 0.000011 - momentum: 0.000000
189
+ 2023-10-15 17:38:36,880 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-15 17:38:36,880 EPOCH 8 done: loss 0.0232 - lr: 0.000011
191
+ 2023-10-15 17:38:45,922 DEV : loss 0.4457707703113556 - f1-score (micro avg) 0.3707
192
+ 2023-10-15 17:38:45,950 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-15 17:39:04,477 epoch 9 - iter 260/2606 - loss 0.01712180 - time (sec): 18.53 - samples/sec: 1918.33 - lr: 0.000011 - momentum: 0.000000
194
+ 2023-10-15 17:39:23,052 epoch 9 - iter 520/2606 - loss 0.01713874 - time (sec): 37.10 - samples/sec: 1927.72 - lr: 0.000010 - momentum: 0.000000
195
+ 2023-10-15 17:39:41,699 epoch 9 - iter 780/2606 - loss 0.01649312 - time (sec): 55.75 - samples/sec: 1939.85 - lr: 0.000009 - momentum: 0.000000
196
+ 2023-10-15 17:40:00,707 epoch 9 - iter 1040/2606 - loss 0.01656224 - time (sec): 74.76 - samples/sec: 1931.44 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-15 17:40:19,694 epoch 9 - iter 1300/2606 - loss 0.01693325 - time (sec): 93.74 - samples/sec: 1918.88 - lr: 0.000008 - momentum: 0.000000
198
+ 2023-10-15 17:40:39,415 epoch 9 - iter 1560/2606 - loss 0.01641210 - time (sec): 113.46 - samples/sec: 1928.31 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-15 17:40:57,945 epoch 9 - iter 1820/2606 - loss 0.01612644 - time (sec): 131.99 - samples/sec: 1933.37 - lr: 0.000007 - momentum: 0.000000
200
+ 2023-10-15 17:41:16,629 epoch 9 - iter 2080/2606 - loss 0.01614170 - time (sec): 150.68 - samples/sec: 1929.74 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-15 17:41:36,069 epoch 9 - iter 2340/2606 - loss 0.01580469 - time (sec): 170.12 - samples/sec: 1933.42 - lr: 0.000006 - momentum: 0.000000
202
+ 2023-10-15 17:41:55,657 epoch 9 - iter 2600/2606 - loss 0.01557134 - time (sec): 189.71 - samples/sec: 1933.00 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-15 17:41:56,016 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-15 17:41:56,016 EPOCH 9 done: loss 0.0156 - lr: 0.000006
205
+ 2023-10-15 17:42:05,154 DEV : loss 0.46648621559143066 - f1-score (micro avg) 0.3789
206
+ 2023-10-15 17:42:05,182 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-15 17:42:23,987 epoch 10 - iter 260/2606 - loss 0.01270944 - time (sec): 18.80 - samples/sec: 1908.91 - lr: 0.000005 - momentum: 0.000000
208
+ 2023-10-15 17:42:43,038 epoch 10 - iter 520/2606 - loss 0.01205548 - time (sec): 37.85 - samples/sec: 1934.15 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-15 17:43:01,852 epoch 10 - iter 780/2606 - loss 0.01025038 - time (sec): 56.67 - samples/sec: 1948.42 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-15 17:43:21,081 epoch 10 - iter 1040/2606 - loss 0.00978047 - time (sec): 75.90 - samples/sec: 1931.66 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-15 17:43:41,232 epoch 10 - iter 1300/2606 - loss 0.00973465 - time (sec): 96.05 - samples/sec: 1930.85 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-15 17:44:00,420 epoch 10 - iter 1560/2606 - loss 0.00965036 - time (sec): 115.24 - samples/sec: 1923.94 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-15 17:44:19,182 epoch 10 - iter 1820/2606 - loss 0.00972466 - time (sec): 134.00 - samples/sec: 1927.01 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-15 17:44:37,598 epoch 10 - iter 2080/2606 - loss 0.01031159 - time (sec): 152.41 - samples/sec: 1925.30 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-15 17:44:56,824 epoch 10 - iter 2340/2606 - loss 0.01023345 - time (sec): 171.64 - samples/sec: 1917.64 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-15 17:45:16,466 epoch 10 - iter 2600/2606 - loss 0.01014657 - time (sec): 191.28 - samples/sec: 1917.45 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-15 17:45:16,848 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-15 17:45:16,848 EPOCH 10 done: loss 0.0102 - lr: 0.000000
219
+ 2023-10-15 17:45:26,009 DEV : loss 0.44499194622039795 - f1-score (micro avg) 0.3941
220
+ 2023-10-15 17:45:26,057 saving best model
221
+ 2023-10-15 17:45:27,099 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-15 17:45:27,100 Loading model from best epoch ...
223
+ 2023-10-15 17:45:28,580 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
224
+ 2023-10-15 17:45:44,396
225
+ Results:
226
+ - F-score (micro) 0.466
227
+ - F-score (macro) 0.3223
228
+ - Accuracy 0.3071
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ LOC 0.4869 0.5519 0.5174 1214
234
+ PER 0.4213 0.5037 0.4589 808
235
+ ORG 0.3062 0.3201 0.3130 353
236
+ HumanProd 0.0000 0.0000 0.0000 15
237
+
238
+ micro avg 0.4380 0.4979 0.4660 2390
239
+ macro avg 0.3036 0.3439 0.3223 2390
240
+ weighted avg 0.4350 0.4979 0.4642 2390
241
+
242
+ 2023-10-15 17:45:44,396 ----------------------------------------------------------------------------------------------------