Hugging Face's logo

hmbert
/
flair-hipe-2022-newseye-de

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
dcccb85
•
1 Parent(s): 2aa08e4

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 +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:467f72e144796a1d43624d0c8583661e68ba29d7f7b556588c6b7083144bbe53
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 19:53:53 0.0000 0.3708 0.0987 0.2859 0.3958 0.3320 0.1998
3
+ 2 19:57:15 0.0000 0.1524 0.1584 0.2472 0.5473 0.3406 0.2067
4
+ 3 20:00:40 0.0000 0.1078 0.1524 0.3169 0.5739 0.4084 0.2585
5
+ 4 20:03:59 0.0000 0.0776 0.2422 0.2974 0.5644 0.3895 0.2437
6
+ 5 20:07:18 0.0000 0.0584 0.3234 0.2523 0.5720 0.3501 0.2137
7
+ 6 20:10:38 0.0000 0.0442 0.3603 0.2762 0.6250 0.3831 0.2383
8
+ 7 20:13:58 0.0000 0.0326 0.3720 0.2851 0.5966 0.3858 0.2406
9
+ 8 20:17:17 0.0000 0.0231 0.3844 0.2919 0.5777 0.3878 0.2421
10
+ 9 20:20:35 0.0000 0.0153 0.4430 0.2738 0.5947 0.3749 0.2321
11
+ 10 20:23:53 0.0000 0.0100 0.4735 0.2795 0.6004 0.3815 0.2366
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 19:50:34,450 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 19:50:34,451 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 19:50:34,451 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 19:50:34,451 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 19:50:34,451 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 19:50:34,451 Train: 20847 sentences
55
+ 2023-10-15 19:50:34,451 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 19:50:34,451 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 19:50:34,451 Training Params:
58
+ 2023-10-15 19:50:34,451 - learning_rate: "5e-05"
59
+ 2023-10-15 19:50:34,451 - mini_batch_size: "8"
60
+ 2023-10-15 19:50:34,451 - max_epochs: "10"
61
+ 2023-10-15 19:50:34,451 - shuffle: "True"
62
+ 2023-10-15 19:50:34,451 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 19:50:34,451 Plugins:
64
+ 2023-10-15 19:50:34,451 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 19:50:34,451 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 19:50:34,451 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 19:50:34,451 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 19:50:34,451 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 19:50:34,451 Computation:
70
+ 2023-10-15 19:50:34,451 - compute on device: cuda:0
71
+ 2023-10-15 19:50:34,451 - embedding storage: none
72
+ 2023-10-15 19:50:34,451 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 19:50:34,452 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-15 19:50:34,452 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 19:50:34,452 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 19:50:52,747 epoch 1 - iter 260/2606 - loss 1.59028644 - time (sec): 18.29 - samples/sec: 1909.58 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-15 19:51:12,202 epoch 1 - iter 520/2606 - loss 0.98518790 - time (sec): 37.75 - samples/sec: 1945.90 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-15 19:51:30,947 epoch 1 - iter 780/2606 - loss 0.76092996 - time (sec): 56.49 - samples/sec: 1924.76 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-15 19:51:50,886 epoch 1 - iter 1040/2606 - loss 0.63417270 - time (sec): 76.43 - samples/sec: 1893.42 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-15 19:52:10,166 epoch 1 - iter 1300/2606 - loss 0.55309956 - time (sec): 95.71 - samples/sec: 1915.70 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-15 19:52:29,070 epoch 1 - iter 1560/2606 - loss 0.49758201 - time (sec): 114.62 - samples/sec: 1905.72 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-15 19:52:48,118 epoch 1 - iter 1820/2606 - loss 0.45440578 - time (sec): 133.67 - samples/sec: 1904.65 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-15 19:53:07,758 epoch 1 - iter 2080/2606 - loss 0.41855391 - time (sec): 153.30 - samples/sec: 1900.97 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-15 19:53:26,472 epoch 1 - iter 2340/2606 - loss 0.39560327 - time (sec): 172.02 - samples/sec: 1905.23 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-15 19:53:46,488 epoch 1 - iter 2600/2606 - loss 0.37125077 - time (sec): 192.03 - samples/sec: 1909.56 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-15 19:53:46,912 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 19:53:46,912 EPOCH 1 done: loss 0.3708 - lr: 0.000050
88
+ 2023-10-15 19:53:53,679 DEV : loss 0.09866511821746826 - f1-score (micro avg) 0.332
89
+ 2023-10-15 19:53:53,707 saving best model
90
+ 2023-10-15 19:53:54,083 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 19:54:13,897 epoch 2 - iter 260/2606 - loss 0.16979246 - time (sec): 19.81 - samples/sec: 1913.91 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-15 19:54:33,718 epoch 2 - iter 520/2606 - loss 0.15396119 - time (sec): 39.63 - samples/sec: 1921.04 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-15 19:54:52,579 epoch 2 - iter 780/2606 - loss 0.15278405 - time (sec): 58.49 - samples/sec: 1938.50 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-15 19:55:11,345 epoch 2 - iter 1040/2606 - loss 0.15627393 - time (sec): 77.26 - samples/sec: 1914.12 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-15 19:55:30,228 epoch 2 - iter 1300/2606 - loss 0.15531609 - time (sec): 96.14 - samples/sec: 1926.26 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-15 19:55:48,789 epoch 2 - iter 1560/2606 - loss 0.15197189 - time (sec): 114.70 - samples/sec: 1925.11 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-15 19:56:08,319 epoch 2 - iter 1820/2606 - loss 0.14969900 - time (sec): 134.23 - samples/sec: 1933.14 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-15 19:56:26,763 epoch 2 - iter 2080/2606 - loss 0.15359140 - time (sec): 152.68 - samples/sec: 1928.53 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-15 19:56:46,738 epoch 2 - iter 2340/2606 - loss 0.15309562 - time (sec): 172.65 - samples/sec: 1927.59 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-15 19:57:05,105 epoch 2 - iter 2600/2606 - loss 0.15242289 - time (sec): 191.02 - samples/sec: 1921.72 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-15 19:57:05,409 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 19:57:05,409 EPOCH 2 done: loss 0.1524 - lr: 0.000044
103
+ 2023-10-15 19:57:15,626 DEV : loss 0.1584077775478363 - f1-score (micro avg) 0.3406
104
+ 2023-10-15 19:57:15,657 saving best model
105
+ 2023-10-15 19:57:16,211 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 19:57:35,187 epoch 3 - iter 260/2606 - loss 0.11299931 - time (sec): 18.97 - samples/sec: 1921.95 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-15 19:57:54,309 epoch 3 - iter 520/2606 - loss 0.10767341 - time (sec): 38.09 - samples/sec: 1918.22 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-15 19:58:15,019 epoch 3 - iter 780/2606 - loss 0.10899205 - time (sec): 58.80 - samples/sec: 1875.68 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-15 19:58:35,731 epoch 3 - iter 1040/2606 - loss 0.11063881 - time (sec): 79.52 - samples/sec: 1865.12 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-15 19:58:55,613 epoch 3 - iter 1300/2606 - loss 0.10698292 - time (sec): 99.40 - samples/sec: 1857.48 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-15 19:59:14,663 epoch 3 - iter 1560/2606 - loss 0.10679134 - time (sec): 118.45 - samples/sec: 1858.88 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-15 19:59:33,448 epoch 3 - iter 1820/2606 - loss 0.10893641 - time (sec): 137.23 - samples/sec: 1867.67 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-15 19:59:53,465 epoch 3 - iter 2080/2606 - loss 0.10710962 - time (sec): 157.25 - samples/sec: 1877.55 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-15 20:00:12,483 epoch 3 - iter 2340/2606 - loss 0.10687887 - time (sec): 176.27 - samples/sec: 1883.24 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-15 20:00:30,949 epoch 3 - iter 2600/2606 - loss 0.10778076 - time (sec): 194.74 - samples/sec: 1882.60 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-15 20:00:31,324 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 20:00:31,324 EPOCH 3 done: loss 0.1078 - lr: 0.000039
118
+ 2023-10-15 20:00:40,591 DEV : loss 0.15240149199962616 - f1-score (micro avg) 0.4084
119
+ 2023-10-15 20:00:40,626 saving best model
120
+ 2023-10-15 20:00:41,208 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-15 20:01:00,629 epoch 4 - iter 260/2606 - loss 0.08035620 - time (sec): 19.41 - samples/sec: 1887.95 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-15 20:01:18,964 epoch 4 - iter 520/2606 - loss 0.07715406 - time (sec): 37.75 - samples/sec: 1884.39 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-15 20:01:37,159 epoch 4 - iter 780/2606 - loss 0.08039817 - time (sec): 55.94 - samples/sec: 1918.31 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-15 20:01:56,554 epoch 4 - iter 1040/2606 - loss 0.07833131 - time (sec): 75.34 - samples/sec: 1914.94 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-15 20:02:14,628 epoch 4 - iter 1300/2606 - loss 0.07761709 - time (sec): 93.41 - samples/sec: 1924.78 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-15 20:02:32,876 epoch 4 - iter 1560/2606 - loss 0.07793021 - time (sec): 111.66 - samples/sec: 1930.25 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-15 20:02:52,296 epoch 4 - iter 1820/2606 - loss 0.07911638 - time (sec): 131.08 - samples/sec: 1936.60 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-15 20:03:10,776 epoch 4 - iter 2080/2606 - loss 0.07888385 - time (sec): 149.56 - samples/sec: 1932.16 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-15 20:03:29,996 epoch 4 - iter 2340/2606 - loss 0.07940196 - time (sec): 168.78 - samples/sec: 1941.28 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-15 20:03:50,163 epoch 4 - iter 2600/2606 - loss 0.07772595 - time (sec): 188.95 - samples/sec: 1939.73 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-15 20:03:50,679 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-15 20:03:50,679 EPOCH 4 done: loss 0.0776 - lr: 0.000033
133
+ 2023-10-15 20:03:59,788 DEV : loss 0.24223537743091583 - f1-score (micro avg) 0.3895
134
+ 2023-10-15 20:03:59,816 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-15 20:04:17,856 epoch 5 - iter 260/2606 - loss 0.05320156 - time (sec): 18.04 - samples/sec: 1901.00 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-15 20:04:36,525 epoch 5 - iter 520/2606 - loss 0.06004399 - time (sec): 36.71 - samples/sec: 1892.24 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-15 20:04:55,064 epoch 5 - iter 780/2606 - loss 0.06361788 - time (sec): 55.25 - samples/sec: 1900.04 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-15 20:05:13,980 epoch 5 - iter 1040/2606 - loss 0.06308322 - time (sec): 74.16 - samples/sec: 1916.92 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-15 20:05:33,591 epoch 5 - iter 1300/2606 - loss 0.06120360 - time (sec): 93.77 - samples/sec: 1917.50 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-15 20:05:52,267 epoch 5 - iter 1560/2606 - loss 0.06016703 - time (sec): 112.45 - samples/sec: 1913.29 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-15 20:06:11,407 epoch 5 - iter 1820/2606 - loss 0.06035800 - time (sec): 131.59 - samples/sec: 1920.61 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-15 20:06:30,790 epoch 5 - iter 2080/2606 - loss 0.05910496 - time (sec): 150.97 - samples/sec: 1928.30 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-15 20:06:50,194 epoch 5 - iter 2340/2606 - loss 0.05814657 - time (sec): 170.38 - samples/sec: 1929.74 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-15 20:07:09,957 epoch 5 - iter 2600/2606 - loss 0.05838664 - time (sec): 190.14 - samples/sec: 1927.76 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-15 20:07:10,400 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-15 20:07:10,401 EPOCH 5 done: loss 0.0584 - lr: 0.000028
147
+ 2023-10-15 20:07:18,725 DEV : loss 0.3234298527240753 - f1-score (micro avg) 0.3501
148
+ 2023-10-15 20:07:18,755 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-15 20:07:38,043 epoch 6 - iter 260/2606 - loss 0.04968909 - time (sec): 19.29 - samples/sec: 1943.44 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-15 20:07:58,295 epoch 6 - iter 520/2606 - loss 0.04591580 - time (sec): 39.54 - samples/sec: 1911.03 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-15 20:08:16,820 epoch 6 - iter 780/2606 - loss 0.04497115 - time (sec): 58.06 - samples/sec: 1917.34 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-15 20:08:36,677 epoch 6 - iter 1040/2606 - loss 0.04222920 - time (sec): 77.92 - samples/sec: 1932.37 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-15 20:08:55,221 epoch 6 - iter 1300/2606 - loss 0.04176238 - time (sec): 96.46 - samples/sec: 1935.86 - lr: 0.000025 - momentum: 0.000000
154
+ 2023-10-15 20:09:14,166 epoch 6 - iter 1560/2606 - loss 0.04201919 - time (sec): 115.41 - samples/sec: 1930.35 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-15 20:09:32,322 epoch 6 - iter 1820/2606 - loss 0.04211214 - time (sec): 133.57 - samples/sec: 1928.39 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-15 20:09:51,338 epoch 6 - iter 2080/2606 - loss 0.04276177 - time (sec): 152.58 - samples/sec: 1919.57 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-15 20:10:09,800 epoch 6 - iter 2340/2606 - loss 0.04345221 - time (sec): 171.04 - samples/sec: 1915.62 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-15 20:10:29,580 epoch 6 - iter 2600/2606 - loss 0.04434131 - time (sec): 190.82 - samples/sec: 1918.41 - lr: 0.000022 - momentum: 0.000000
159
+ 2023-10-15 20:10:30,220 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-15 20:10:30,220 EPOCH 6 done: loss 0.0442 - lr: 0.000022
161
+ 2023-10-15 20:10:38,461 DEV : loss 0.360347181558609 - f1-score (micro avg) 0.3831
162
+ 2023-10-15 20:10:38,489 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-15 20:10:57,550 epoch 7 - iter 260/2606 - loss 0.03605703 - time (sec): 19.06 - samples/sec: 1967.84 - lr: 0.000022 - momentum: 0.000000
164
+ 2023-10-15 20:11:17,462 epoch 7 - iter 520/2606 - loss 0.03233024 - time (sec): 38.97 - samples/sec: 1942.69 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-15 20:11:36,061 epoch 7 - iter 780/2606 - loss 0.03422854 - time (sec): 57.57 - samples/sec: 1930.25 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-15 20:11:55,421 epoch 7 - iter 1040/2606 - loss 0.03509936 - time (sec): 76.93 - samples/sec: 1887.52 - lr: 0.000020 - momentum: 0.000000
167
+ 2023-10-15 20:12:15,268 epoch 7 - iter 1300/2606 - loss 0.03481716 - time (sec): 96.78 - samples/sec: 1892.73 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-15 20:12:33,527 epoch 7 - iter 1560/2606 - loss 0.03453842 - time (sec): 115.04 - samples/sec: 1897.80 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-15 20:12:53,305 epoch 7 - iter 1820/2606 - loss 0.03453028 - time (sec): 134.81 - samples/sec: 1910.27 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-15 20:13:11,312 epoch 7 - iter 2080/2606 - loss 0.03351666 - time (sec): 152.82 - samples/sec: 1910.11 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-15 20:13:30,700 epoch 7 - iter 2340/2606 - loss 0.03294077 - time (sec): 172.21 - samples/sec: 1917.56 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-15 20:13:49,306 epoch 7 - iter 2600/2606 - loss 0.03257111 - time (sec): 190.82 - samples/sec: 1920.47 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-15 20:13:49,775 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-15 20:13:49,775 EPOCH 7 done: loss 0.0326 - lr: 0.000017
175
+ 2023-10-15 20:13:57,989 DEV : loss 0.37199294567108154 - f1-score (micro avg) 0.3858
176
+ 2023-10-15 20:13:58,018 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-15 20:14:16,046 epoch 8 - iter 260/2606 - loss 0.02342308 - time (sec): 18.03 - samples/sec: 1889.83 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-15 20:14:35,427 epoch 8 - iter 520/2606 - loss 0.02364228 - time (sec): 37.41 - samples/sec: 1945.44 - lr: 0.000016 - momentum: 0.000000
179
+ 2023-10-15 20:14:54,448 epoch 8 - iter 780/2606 - loss 0.02356941 - time (sec): 56.43 - samples/sec: 1926.14 - lr: 0.000015 - momentum: 0.000000
180
+ 2023-10-15 20:15:13,454 epoch 8 - iter 1040/2606 - loss 0.02491977 - time (sec): 75.44 - samples/sec: 1918.27 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-15 20:15:32,549 epoch 8 - iter 1300/2606 - loss 0.02428659 - time (sec): 94.53 - samples/sec: 1932.58 - lr: 0.000014 - momentum: 0.000000
182
+ 2023-10-15 20:15:52,160 epoch 8 - iter 1560/2606 - loss 0.02366101 - time (sec): 114.14 - samples/sec: 1931.16 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-15 20:16:10,784 epoch 8 - iter 1820/2606 - loss 0.02343244 - time (sec): 132.77 - samples/sec: 1938.37 - lr: 0.000013 - momentum: 0.000000
184
+ 2023-10-15 20:16:30,328 epoch 8 - iter 2080/2606 - loss 0.02351442 - time (sec): 152.31 - samples/sec: 1925.81 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-15 20:16:49,792 epoch 8 - iter 2340/2606 - loss 0.02300749 - time (sec): 171.77 - samples/sec: 1923.52 - lr: 0.000012 - momentum: 0.000000
186
+ 2023-10-15 20:17:08,499 epoch 8 - iter 2600/2606 - loss 0.02311972 - time (sec): 190.48 - samples/sec: 1924.21 - lr: 0.000011 - momentum: 0.000000
187
+ 2023-10-15 20:17:08,968 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-15 20:17:08,968 EPOCH 8 done: loss 0.0231 - lr: 0.000011
189
+ 2023-10-15 20:17:17,209 DEV : loss 0.38435670733451843 - f1-score (micro avg) 0.3878
190
+ 2023-10-15 20:17:17,237 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-15 20:17:36,880 epoch 9 - iter 260/2606 - loss 0.01429002 - time (sec): 19.64 - samples/sec: 2021.11 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-15 20:17:56,464 epoch 9 - iter 520/2606 - loss 0.01483258 - time (sec): 39.23 - samples/sec: 1987.77 - lr: 0.000010 - momentum: 0.000000
193
+ 2023-10-15 20:18:15,715 epoch 9 - iter 780/2606 - loss 0.01447465 - time (sec): 58.48 - samples/sec: 1973.65 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-15 20:18:34,051 epoch 9 - iter 1040/2606 - loss 0.01472942 - time (sec): 76.81 - samples/sec: 1971.06 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-15 20:18:52,832 epoch 9 - iter 1300/2606 - loss 0.01528011 - time (sec): 95.59 - samples/sec: 1965.67 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-15 20:19:10,916 epoch 9 - iter 1560/2606 - loss 0.01518566 - time (sec): 113.68 - samples/sec: 1940.69 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-15 20:19:30,000 epoch 9 - iter 1820/2606 - loss 0.01575329 - time (sec): 132.76 - samples/sec: 1940.91 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-15 20:19:48,598 epoch 9 - iter 2080/2606 - loss 0.01545732 - time (sec): 151.36 - samples/sec: 1943.70 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-15 20:20:07,608 epoch 9 - iter 2340/2606 - loss 0.01535373 - time (sec): 170.37 - samples/sec: 1939.88 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-15 20:20:26,963 epoch 9 - iter 2600/2606 - loss 0.01534271 - time (sec): 189.72 - samples/sec: 1933.83 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-15 20:20:27,294 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-15 20:20:27,295 EPOCH 9 done: loss 0.0153 - lr: 0.000006
203
+ 2023-10-15 20:20:35,652 DEV : loss 0.44299206137657166 - f1-score (micro avg) 0.3749
204
+ 2023-10-15 20:20:35,697 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-15 20:20:53,742 epoch 10 - iter 260/2606 - loss 0.01007017 - time (sec): 18.04 - samples/sec: 1966.73 - lr: 0.000005 - momentum: 0.000000
206
+ 2023-10-15 20:21:11,849 epoch 10 - iter 520/2606 - loss 0.00872718 - time (sec): 36.15 - samples/sec: 1916.23 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-15 20:21:30,620 epoch 10 - iter 780/2606 - loss 0.00877023 - time (sec): 54.92 - samples/sec: 1938.58 - lr: 0.000004 - momentum: 0.000000
208
+ 2023-10-15 20:21:48,942 epoch 10 - iter 1040/2606 - loss 0.00900546 - time (sec): 73.24 - samples/sec: 1944.56 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-15 20:22:07,349 epoch 10 - iter 1300/2606 - loss 0.00925451 - time (sec): 91.65 - samples/sec: 1947.96 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-15 20:22:26,240 epoch 10 - iter 1560/2606 - loss 0.00955119 - time (sec): 110.54 - samples/sec: 1945.38 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-15 20:22:45,017 epoch 10 - iter 1820/2606 - loss 0.00958418 - time (sec): 129.32 - samples/sec: 1949.64 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-15 20:23:04,810 epoch 10 - iter 2080/2606 - loss 0.00978376 - time (sec): 149.11 - samples/sec: 1952.87 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 20:23:23,622 epoch 10 - iter 2340/2606 - loss 0.00995251 - time (sec): 167.92 - samples/sec: 1946.93 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-15 20:23:43,809 epoch 10 - iter 2600/2606 - loss 0.01001595 - time (sec): 188.11 - samples/sec: 1948.65 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-15 20:23:44,199 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-15 20:23:44,200 EPOCH 10 done: loss 0.0100 - lr: 0.000000
217
+ 2023-10-15 20:23:53,258 DEV : loss 0.4734904170036316 - f1-score (micro avg) 0.3815
218
+ 2023-10-15 20:23:53,688 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-15 20:23:53,689 Loading model from best epoch ...
220
+ 2023-10-15 20:23:55,224 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
221
+ 2023-10-15 20:24:11,965
222
+ Results:
223
+ - F-score (micro) 0.446
224
+ - F-score (macro) 0.2767
225
+ - Accuracy 0.2916
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.4970 0.6063 0.5462 1214
231
+ PER 0.4008 0.3651 0.3821 808
232
+ ORG 0.2091 0.1558 0.1786 353
233
+ HumanProd 0.0000 0.0000 0.0000 15
234
+
235
+ micro avg 0.4379 0.4544 0.4460 2390
236
+ macro avg 0.2767 0.2818 0.2767 2390
237
+ weighted avg 0.4188 0.4544 0.4330 2390
238
+
239
+ 2023-10-15 20:24:11,966 ----------------------------------------------------------------------------------------------------