Hugging Face's logo

hmbert
/
flair-hipe-2022-topres19th-en

Token Classification
Flair
PyTorch
English
sequence-tagger-model
Model card Files Community
stefan-it commited on
Commit
2aee4f7
·
1 Parent(s): 0fe0efc

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 +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d1e2c3454708e9fb4655cda7449a7f6944ba29197b36107c16bf88290a53de1
3
+ size 443311175
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 00:03:02 0.0000 0.3872 0.0565 0.6416 0.7553 0.6938 0.5424
3
+ 2 00:03:50 0.0000 0.0769 0.0547 0.7633 0.7890 0.7759 0.6448
4
+ 3 00:04:38 0.0000 0.0485 0.0662 0.7341 0.8270 0.7778 0.6512
5
+ 4 00:05:26 0.0000 0.0318 0.0804 0.7958 0.8059 0.8008 0.6821
6
+ 5 00:06:14 0.0000 0.0219 0.0928 0.7717 0.8270 0.7984 0.6689
7
+ 6 00:07:02 0.0000 0.0171 0.0989 0.7598 0.8143 0.7862 0.6632
8
+ 7 00:07:49 0.0000 0.0115 0.1013 0.7817 0.8312 0.8057 0.6840
9
+ 8 00:08:37 0.0000 0.0081 0.1050 0.7866 0.8397 0.8122 0.6958
10
+ 9 00:09:25 0.0000 0.0056 0.1134 0.7773 0.8101 0.7934 0.6737
11
+ 10 00:10:12 0.0000 0.0036 0.1106 0.7720 0.8143 0.7926 0.6725
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 00:02:15,413 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 00:02:15,414 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=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-17 00:02:15,414 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-17 00:02:15,414 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
52
+ - NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
53
+ 2023-10-17 00:02:15,414 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-17 00:02:15,414 Train: 6183 sentences
55
+ 2023-10-17 00:02:15,414 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-17 00:02:15,414 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-17 00:02:15,415 Training Params:
58
+ 2023-10-17 00:02:15,415 - learning_rate: "3e-05"
59
+ 2023-10-17 00:02:15,415 - mini_batch_size: "8"
60
+ 2023-10-17 00:02:15,415 - max_epochs: "10"
61
+ 2023-10-17 00:02:15,415 - shuffle: "True"
62
+ 2023-10-17 00:02:15,415 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 00:02:15,415 Plugins:
64
+ 2023-10-17 00:02:15,415 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-17 00:02:15,415 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 00:02:15,415 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-17 00:02:15,415 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-17 00:02:15,415 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-17 00:02:15,415 Computation:
70
+ 2023-10-17 00:02:15,415 - compute on device: cuda:0
71
+ 2023-10-17 00:02:15,415 - embedding storage: none
72
+ 2023-10-17 00:02:15,415 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 00:02:15,415 Model training base path: "hmbench-topres19th/en-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-17 00:02:15,415 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-17 00:02:15,415 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-17 00:02:19,821 epoch 1 - iter 77/773 - loss 2.32537798 - time (sec): 4.40 - samples/sec: 2923.31 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-17 00:02:24,181 epoch 1 - iter 154/773 - loss 1.39644595 - time (sec): 8.76 - samples/sec: 2910.79 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-17 00:02:28,671 epoch 1 - iter 231/773 - loss 1.01126391 - time (sec): 13.25 - samples/sec: 2847.50 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-17 00:02:33,108 epoch 1 - iter 308/773 - loss 0.80543336 - time (sec): 17.69 - samples/sec: 2821.83 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-17 00:02:37,725 epoch 1 - iter 385/773 - loss 0.66957136 - time (sec): 22.31 - samples/sec: 2803.17 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-17 00:02:42,147 epoch 1 - iter 462/773 - loss 0.58326545 - time (sec): 26.73 - samples/sec: 2778.15 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-17 00:02:46,740 epoch 1 - iter 539/773 - loss 0.51629996 - time (sec): 31.32 - samples/sec: 2754.64 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-17 00:02:51,150 epoch 1 - iter 616/773 - loss 0.46530910 - time (sec): 35.73 - samples/sec: 2756.33 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-17 00:02:55,889 epoch 1 - iter 693/773 - loss 0.42285036 - time (sec): 40.47 - samples/sec: 2748.13 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-17 00:03:00,302 epoch 1 - iter 770/773 - loss 0.38822438 - time (sec): 44.89 - samples/sec: 2760.80 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-17 00:03:00,451 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-17 00:03:00,451 EPOCH 1 done: loss 0.3872 - lr: 0.000030
88
+ 2023-10-17 00:03:02,192 DEV : loss 0.05650660768151283 - f1-score (micro avg) 0.6938
89
+ 2023-10-17 00:03:02,205 saving best model
90
+ 2023-10-17 00:03:02,539 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-17 00:03:07,258 epoch 2 - iter 77/773 - loss 0.08489853 - time (sec): 4.72 - samples/sec: 2802.48 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-17 00:03:11,873 epoch 2 - iter 154/773 - loss 0.08347053 - time (sec): 9.33 - samples/sec: 2766.52 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-17 00:03:16,356 epoch 2 - iter 231/773 - loss 0.08160898 - time (sec): 13.82 - samples/sec: 2757.91 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-17 00:03:20,717 epoch 2 - iter 308/773 - loss 0.08349640 - time (sec): 18.18 - samples/sec: 2752.10 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-17 00:03:25,154 epoch 2 - iter 385/773 - loss 0.08202306 - time (sec): 22.61 - samples/sec: 2720.25 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-17 00:03:29,566 epoch 2 - iter 462/773 - loss 0.08138239 - time (sec): 27.03 - samples/sec: 2739.59 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-17 00:03:34,029 epoch 2 - iter 539/773 - loss 0.08140479 - time (sec): 31.49 - samples/sec: 2754.29 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-17 00:03:38,872 epoch 2 - iter 616/773 - loss 0.07747793 - time (sec): 36.33 - samples/sec: 2743.79 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-17 00:03:43,274 epoch 2 - iter 693/773 - loss 0.07718394 - time (sec): 40.73 - samples/sec: 2736.23 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-17 00:03:47,771 epoch 2 - iter 770/773 - loss 0.07694589 - time (sec): 45.23 - samples/sec: 2741.04 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-17 00:03:47,914 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-17 00:03:47,915 EPOCH 2 done: loss 0.0769 - lr: 0.000027
103
+ 2023-10-17 00:03:50,333 DEV : loss 0.0547206737101078 - f1-score (micro avg) 0.7759
104
+ 2023-10-17 00:03:50,346 saving best model
105
+ 2023-10-17 00:03:50,814 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-17 00:03:55,393 epoch 3 - iter 77/773 - loss 0.04187396 - time (sec): 4.58 - samples/sec: 2791.16 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-17 00:04:00,079 epoch 3 - iter 154/773 - loss 0.05682215 - time (sec): 9.26 - samples/sec: 2789.17 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-17 00:04:04,789 epoch 3 - iter 231/773 - loss 0.05388452 - time (sec): 13.97 - samples/sec: 2803.36 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-17 00:04:09,173 epoch 3 - iter 308/773 - loss 0.05046208 - time (sec): 18.36 - samples/sec: 2767.82 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-17 00:04:13,610 epoch 3 - iter 385/773 - loss 0.04968648 - time (sec): 22.79 - samples/sec: 2757.33 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-17 00:04:18,042 epoch 3 - iter 462/773 - loss 0.04971432 - time (sec): 27.23 - samples/sec: 2743.24 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-17 00:04:22,720 epoch 3 - iter 539/773 - loss 0.04992038 - time (sec): 31.90 - samples/sec: 2749.36 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-17 00:04:27,265 epoch 3 - iter 616/773 - loss 0.04910921 - time (sec): 36.45 - samples/sec: 2740.05 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-17 00:04:31,575 epoch 3 - iter 693/773 - loss 0.04872624 - time (sec): 40.76 - samples/sec: 2733.99 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-17 00:04:36,034 epoch 3 - iter 770/773 - loss 0.04859388 - time (sec): 45.22 - samples/sec: 2739.46 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-17 00:04:36,180 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-17 00:04:36,181 EPOCH 3 done: loss 0.0485 - lr: 0.000023
118
+ 2023-10-17 00:04:38,322 DEV : loss 0.06617607176303864 - f1-score (micro avg) 0.7778
119
+ 2023-10-17 00:04:38,335 saving best model
120
+ 2023-10-17 00:04:38,793 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-17 00:04:43,088 epoch 4 - iter 77/773 - loss 0.03283229 - time (sec): 4.29 - samples/sec: 2716.07 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-17 00:04:47,633 epoch 4 - iter 154/773 - loss 0.02889967 - time (sec): 8.83 - samples/sec: 2678.88 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-17 00:04:52,242 epoch 4 - iter 231/773 - loss 0.03012282 - time (sec): 13.44 - samples/sec: 2702.53 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-17 00:04:56,714 epoch 4 - iter 308/773 - loss 0.02887586 - time (sec): 17.92 - samples/sec: 2711.46 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-17 00:05:01,312 epoch 4 - iter 385/773 - loss 0.03075624 - time (sec): 22.51 - samples/sec: 2701.08 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-17 00:05:05,714 epoch 4 - iter 462/773 - loss 0.03144270 - time (sec): 26.92 - samples/sec: 2704.90 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-17 00:05:10,256 epoch 4 - iter 539/773 - loss 0.03187058 - time (sec): 31.46 - samples/sec: 2716.26 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-17 00:05:14,768 epoch 4 - iter 616/773 - loss 0.03220682 - time (sec): 35.97 - samples/sec: 2715.62 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-17 00:05:19,176 epoch 4 - iter 693/773 - loss 0.03218910 - time (sec): 40.38 - samples/sec: 2735.10 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-17 00:05:23,884 epoch 4 - iter 770/773 - loss 0.03176330 - time (sec): 45.09 - samples/sec: 2743.13 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-17 00:05:24,069 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-17 00:05:24,069 EPOCH 4 done: loss 0.0318 - lr: 0.000020
133
+ 2023-10-17 00:05:26,248 DEV : loss 0.08043687045574188 - f1-score (micro avg) 0.8008
134
+ 2023-10-17 00:05:26,261 saving best model
135
+ 2023-10-17 00:05:26,692 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-17 00:05:31,172 epoch 5 - iter 77/773 - loss 0.02544472 - time (sec): 4.48 - samples/sec: 2766.54 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-17 00:05:35,661 epoch 5 - iter 154/773 - loss 0.02257079 - time (sec): 8.97 - samples/sec: 2720.45 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-17 00:05:40,059 epoch 5 - iter 231/773 - loss 0.02161194 - time (sec): 13.36 - samples/sec: 2755.55 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-17 00:05:44,332 epoch 5 - iter 308/773 - loss 0.02181905 - time (sec): 17.64 - samples/sec: 2788.07 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-17 00:05:49,054 epoch 5 - iter 385/773 - loss 0.02217563 - time (sec): 22.36 - samples/sec: 2775.13 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-17 00:05:53,652 epoch 5 - iter 462/773 - loss 0.02354034 - time (sec): 26.96 - samples/sec: 2746.15 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-17 00:05:58,250 epoch 5 - iter 539/773 - loss 0.02344632 - time (sec): 31.55 - samples/sec: 2774.95 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-17 00:06:02,708 epoch 5 - iter 616/773 - loss 0.02222092 - time (sec): 36.01 - samples/sec: 2760.67 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-17 00:06:07,215 epoch 5 - iter 693/773 - loss 0.02187424 - time (sec): 40.52 - samples/sec: 2757.91 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-17 00:06:11,843 epoch 5 - iter 770/773 - loss 0.02196420 - time (sec): 45.15 - samples/sec: 2746.18 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-17 00:06:11,993 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-17 00:06:11,994 EPOCH 5 done: loss 0.0219 - lr: 0.000017
148
+ 2023-10-17 00:06:14,045 DEV : loss 0.09281734377145767 - f1-score (micro avg) 0.7984
149
+ 2023-10-17 00:06:14,058 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-17 00:06:18,476 epoch 6 - iter 77/773 - loss 0.01754628 - time (sec): 4.42 - samples/sec: 2851.07 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-17 00:06:23,103 epoch 6 - iter 154/773 - loss 0.01478913 - time (sec): 9.04 - samples/sec: 2805.88 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-17 00:06:27,628 epoch 6 - iter 231/773 - loss 0.01586836 - time (sec): 13.57 - samples/sec: 2737.88 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-17 00:06:32,019 epoch 6 - iter 308/773 - loss 0.01810026 - time (sec): 17.96 - samples/sec: 2764.63 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-17 00:06:36,634 epoch 6 - iter 385/773 - loss 0.01805116 - time (sec): 22.58 - samples/sec: 2764.74 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-17 00:06:41,164 epoch 6 - iter 462/773 - loss 0.01877742 - time (sec): 27.10 - samples/sec: 2739.87 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-17 00:06:45,844 epoch 6 - iter 539/773 - loss 0.01719607 - time (sec): 31.79 - samples/sec: 2734.19 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-17 00:06:50,351 epoch 6 - iter 616/773 - loss 0.01753376 - time (sec): 36.29 - samples/sec: 2692.58 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-17 00:06:55,337 epoch 6 - iter 693/773 - loss 0.01746324 - time (sec): 41.28 - samples/sec: 2680.67 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-17 00:06:59,848 epoch 6 - iter 770/773 - loss 0.01698989 - time (sec): 45.79 - samples/sec: 2707.29 - lr: 0.000013 - momentum: 0.000000
160
+ 2023-10-17 00:07:00,006 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-17 00:07:00,006 EPOCH 6 done: loss 0.0171 - lr: 0.000013
162
+ 2023-10-17 00:07:02,032 DEV : loss 0.0988919660449028 - f1-score (micro avg) 0.7862
163
+ 2023-10-17 00:07:02,046 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-17 00:07:06,337 epoch 7 - iter 77/773 - loss 0.01237299 - time (sec): 4.29 - samples/sec: 2694.78 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-17 00:07:10,716 epoch 7 - iter 154/773 - loss 0.01304177 - time (sec): 8.67 - samples/sec: 2680.44 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-17 00:07:15,149 epoch 7 - iter 231/773 - loss 0.01266367 - time (sec): 13.10 - samples/sec: 2681.15 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-17 00:07:19,827 epoch 7 - iter 308/773 - loss 0.01099661 - time (sec): 17.78 - samples/sec: 2701.77 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-17 00:07:24,501 epoch 7 - iter 385/773 - loss 0.01065385 - time (sec): 22.45 - samples/sec: 2710.31 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-17 00:07:29,332 epoch 7 - iter 462/773 - loss 0.01100561 - time (sec): 27.29 - samples/sec: 2701.80 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-17 00:07:33,792 epoch 7 - iter 539/773 - loss 0.01140856 - time (sec): 31.74 - samples/sec: 2737.64 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-17 00:07:38,159 epoch 7 - iter 616/773 - loss 0.01149250 - time (sec): 36.11 - samples/sec: 2752.33 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-17 00:07:42,551 epoch 7 - iter 693/773 - loss 0.01180606 - time (sec): 40.50 - samples/sec: 2746.82 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-17 00:07:47,066 epoch 7 - iter 770/773 - loss 0.01156637 - time (sec): 45.02 - samples/sec: 2751.41 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-17 00:07:47,241 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-17 00:07:47,241 EPOCH 7 done: loss 0.0115 - lr: 0.000010
176
+ 2023-10-17 00:07:49,330 DEV : loss 0.10134067386388779 - f1-score (micro avg) 0.8057
177
+ 2023-10-17 00:07:49,343 saving best model
178
+ 2023-10-17 00:07:49,797 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-17 00:07:54,435 epoch 8 - iter 77/773 - loss 0.01014825 - time (sec): 4.64 - samples/sec: 2668.24 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-17 00:07:59,271 epoch 8 - iter 154/773 - loss 0.00874093 - time (sec): 9.47 - samples/sec: 2726.47 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-17 00:08:03,768 epoch 8 - iter 231/773 - loss 0.00992027 - time (sec): 13.97 - samples/sec: 2711.53 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-17 00:08:08,281 epoch 8 - iter 308/773 - loss 0.00892334 - time (sec): 18.48 - samples/sec: 2743.99 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-17 00:08:12,957 epoch 8 - iter 385/773 - loss 0.00881416 - time (sec): 23.16 - samples/sec: 2741.93 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-17 00:08:17,660 epoch 8 - iter 462/773 - loss 0.00832352 - time (sec): 27.86 - samples/sec: 2744.83 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-17 00:08:22,079 epoch 8 - iter 539/773 - loss 0.00829162 - time (sec): 32.28 - samples/sec: 2727.50 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-17 00:08:26,396 epoch 8 - iter 616/773 - loss 0.00850288 - time (sec): 36.60 - samples/sec: 2717.51 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-17 00:08:30,865 epoch 8 - iter 693/773 - loss 0.00806884 - time (sec): 41.07 - samples/sec: 2731.36 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-17 00:08:35,232 epoch 8 - iter 770/773 - loss 0.00809371 - time (sec): 45.43 - samples/sec: 2726.00 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-17 00:08:35,396 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-17 00:08:35,397 EPOCH 8 done: loss 0.0081 - lr: 0.000007
191
+ 2023-10-17 00:08:37,446 DEV : loss 0.10502836853265762 - f1-score (micro avg) 0.8122
192
+ 2023-10-17 00:08:37,459 saving best model
193
+ 2023-10-17 00:08:37,910 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-17 00:08:42,371 epoch 9 - iter 77/773 - loss 0.01014486 - time (sec): 4.46 - samples/sec: 2743.83 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-17 00:08:46,899 epoch 9 - iter 154/773 - loss 0.00742422 - time (sec): 8.99 - samples/sec: 2812.20 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-17 00:08:51,665 epoch 9 - iter 231/773 - loss 0.00589126 - time (sec): 13.75 - samples/sec: 2791.71 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-17 00:08:56,063 epoch 9 - iter 308/773 - loss 0.00726357 - time (sec): 18.15 - samples/sec: 2778.16 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-17 00:09:00,644 epoch 9 - iter 385/773 - loss 0.00604410 - time (sec): 22.73 - samples/sec: 2758.51 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-17 00:09:05,214 epoch 9 - iter 462/773 - loss 0.00579631 - time (sec): 27.30 - samples/sec: 2744.61 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-17 00:09:09,519 epoch 9 - iter 539/773 - loss 0.00544690 - time (sec): 31.61 - samples/sec: 2733.80 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-17 00:09:13,933 epoch 9 - iter 616/773 - loss 0.00539373 - time (sec): 36.02 - samples/sec: 2753.50 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-17 00:09:18,546 epoch 9 - iter 693/773 - loss 0.00528883 - time (sec): 40.63 - samples/sec: 2748.24 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-17 00:09:22,935 epoch 9 - iter 770/773 - loss 0.00566848 - time (sec): 45.02 - samples/sec: 2748.09 - lr: 0.000003 - momentum: 0.000000
204
+ 2023-10-17 00:09:23,109 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-17 00:09:23,109 EPOCH 9 done: loss 0.0056 - lr: 0.000003
206
+ 2023-10-17 00:09:25,144 DEV : loss 0.11343234777450562 - f1-score (micro avg) 0.7934
207
+ 2023-10-17 00:09:25,157 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-17 00:09:29,896 epoch 10 - iter 77/773 - loss 0.00231576 - time (sec): 4.74 - samples/sec: 2654.62 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-17 00:09:34,465 epoch 10 - iter 154/773 - loss 0.00249641 - time (sec): 9.31 - samples/sec: 2684.47 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-17 00:09:38,825 epoch 10 - iter 231/773 - loss 0.00374059 - time (sec): 13.67 - samples/sec: 2660.85 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-17 00:09:43,507 epoch 10 - iter 308/773 - loss 0.00401961 - time (sec): 18.35 - samples/sec: 2682.09 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-17 00:09:48,107 epoch 10 - iter 385/773 - loss 0.00388607 - time (sec): 22.95 - samples/sec: 2728.17 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-17 00:09:52,698 epoch 10 - iter 462/773 - loss 0.00364499 - time (sec): 27.54 - samples/sec: 2749.47 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-17 00:09:57,479 epoch 10 - iter 539/773 - loss 0.00343129 - time (sec): 32.32 - samples/sec: 2723.02 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-17 00:10:01,847 epoch 10 - iter 616/773 - loss 0.00340294 - time (sec): 36.69 - samples/sec: 2714.54 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-17 00:10:06,269 epoch 10 - iter 693/773 - loss 0.00353649 - time (sec): 41.11 - samples/sec: 2721.73 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-17 00:10:10,747 epoch 10 - iter 770/773 - loss 0.00363040 - time (sec): 45.59 - samples/sec: 2716.64 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-17 00:10:10,902 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-17 00:10:10,903 EPOCH 10 done: loss 0.0036 - lr: 0.000000
220
+ 2023-10-17 00:10:12,932 DEV : loss 0.1106431633234024 - f1-score (micro avg) 0.7926
221
+ 2023-10-17 00:10:13,311 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-17 00:10:13,312 Loading model from best epoch ...
223
+ 2023-10-17 00:10:14,890 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
224
+ 2023-10-17 00:10:21,007
225
+ Results:
226
+ - F-score (micro) 0.7928
227
+ - F-score (macro) 0.6941
228
+ - Accuracy 0.6803
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ LOC 0.8269 0.8584 0.8423 946
234
+ BUILDING 0.5848 0.5405 0.5618 185
235
+ STREET 0.6610 0.6964 0.6783 56
236
+
237
+ micro avg 0.7847 0.8012 0.7928 1187
238
+ macro avg 0.6909 0.6984 0.6941 1187
239
+ weighted avg 0.7813 0.8012 0.7909 1187
240
+
241
+ 2023-10-17 00:10:21,007 ----------------------------------------------------------------------------------------------------