Hugging Face's logo

stefan-it
/
hmbench-topres19th-en-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
English
sequence-tagger-model
Model card Files Community
stefan-it commited on
Commit
a7f3202
•
1 Parent(s): 25841f1

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 sha256:0b2e7f54e3988d88c9d3af8bf5c158e5c94cdc3d78abc119eac033ef57d26aa4
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 22:47:37 0.0000 0.3166 0.0581 0.6834 0.7468 0.7137 0.5728
3
+ 2 22:48:25 0.0000 0.0788 0.0493 0.8165 0.7511 0.7824 0.6568
4
+ 3 22:49:13 0.0000 0.0526 0.0622 0.7749 0.7553 0.7650 0.6393
5
+ 4 22:50:01 0.0000 0.0358 0.0944 0.6912 0.8312 0.7548 0.6215
6
+ 5 22:50:49 0.0000 0.0241 0.1080 0.7559 0.8101 0.7821 0.6508
7
+ 6 22:51:37 0.0000 0.0196 0.1095 0.7782 0.7848 0.7815 0.6549
8
+ 7 22:52:24 0.0000 0.0151 0.1099 0.7621 0.7975 0.7794 0.6585
9
+ 8 22:53:13 0.0000 0.0088 0.1129 0.7773 0.8101 0.7934 0.6690
10
+ 9 22:54:01 0.0000 0.0062 0.1103 0.8042 0.8143 0.8092 0.6918
11
+ 10 22:54:49 0.0000 0.0036 0.1120 0.7814 0.8143 0.7975 0.6772
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-16 22:46:49,399 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-16 22:46:49,400 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
+ 2023-10-16 22:46:49,400 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-16 22:46:49,400 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
+ 2023-10-16 22:46:49,400 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-16 22:46:49,400 Train: 6183 sentences
55
+ 2023-10-16 22:46:49,400 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-16 22:46:49,400 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-16 22:46:49,400 Training Params:
58
+ 2023-10-16 22:46:49,400 - learning_rate: "5e-05"
59
+ 2023-10-16 22:46:49,400 - mini_batch_size: "8"
60
+ 2023-10-16 22:46:49,400 - max_epochs: "10"
61
+ 2023-10-16 22:46:49,400 - shuffle: "True"
62
+ 2023-10-16 22:46:49,400 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-16 22:46:49,400 Plugins:
64
+ 2023-10-16 22:46:49,400 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-16 22:46:49,400 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-16 22:46:49,401 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-16 22:46:49,401 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-16 22:46:49,401 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-16 22:46:49,401 Computation:
70
+ 2023-10-16 22:46:49,401 - compute on device: cuda:0
71
+ 2023-10-16 22:46:49,401 - embedding storage: none
72
+ 2023-10-16 22:46:49,401 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-16 22:46:49,401 Model training base path: "hmbench-topres19th/en-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-16 22:46:49,401 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-16 22:46:49,401 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-16 22:46:54,117 epoch 1 - iter 77/773 - loss 1.85204228 - time (sec): 4.72 - samples/sec: 2772.74 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-16 22:46:58,843 epoch 1 - iter 154/773 - loss 1.09260658 - time (sec): 9.44 - samples/sec: 2663.29 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-16 22:47:03,506 epoch 1 - iter 231/773 - loss 0.78200160 - time (sec): 14.10 - samples/sec: 2710.45 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-16 22:47:07,811 epoch 1 - iter 308/773 - loss 0.62435683 - time (sec): 18.41 - samples/sec: 2733.35 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-16 22:47:12,365 epoch 1 - iter 385/773 - loss 0.52479921 - time (sec): 22.96 - samples/sec: 2728.68 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-16 22:47:16,890 epoch 1 - iter 462/773 - loss 0.45944559 - time (sec): 27.49 - samples/sec: 2705.16 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-16 22:47:21,384 epoch 1 - iter 539/773 - loss 0.40938835 - time (sec): 31.98 - samples/sec: 2709.25 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-16 22:47:25,730 epoch 1 - iter 616/773 - loss 0.37131140 - time (sec): 36.33 - samples/sec: 2724.14 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-16 22:47:30,373 epoch 1 - iter 693/773 - loss 0.34140050 - time (sec): 40.97 - samples/sec: 2722.91 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-16 22:47:34,837 epoch 1 - iter 770/773 - loss 0.31723684 - time (sec): 45.43 - samples/sec: 2726.82 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-16 22:47:34,988 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-16 22:47:34,988 EPOCH 1 done: loss 0.3166 - lr: 0.000050
88
+ 2023-10-16 22:47:37,086 DEV : loss 0.05812298133969307 - f1-score (micro avg) 0.7137
89
+ 2023-10-16 22:47:37,102 saving best model
90
+ 2023-10-16 22:47:37,457 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-16 22:47:41,895 epoch 2 - iter 77/773 - loss 0.09219451 - time (sec): 4.44 - samples/sec: 2766.63 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-16 22:47:46,460 epoch 2 - iter 154/773 - loss 0.08570322 - time (sec): 9.00 - samples/sec: 2825.73 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-16 22:47:50,870 epoch 2 - iter 231/773 - loss 0.08707063 - time (sec): 13.41 - samples/sec: 2761.19 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-16 22:47:55,496 epoch 2 - iter 308/773 - loss 0.08465278 - time (sec): 18.04 - samples/sec: 2736.45 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-16 22:47:59,913 epoch 2 - iter 385/773 - loss 0.08603529 - time (sec): 22.45 - samples/sec: 2728.10 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-16 22:48:04,824 epoch 2 - iter 462/773 - loss 0.08248188 - time (sec): 27.37 - samples/sec: 2732.13 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-16 22:48:09,328 epoch 2 - iter 539/773 - loss 0.08143942 - time (sec): 31.87 - samples/sec: 2723.81 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-16 22:48:13,974 epoch 2 - iter 616/773 - loss 0.08180843 - time (sec): 36.52 - samples/sec: 2708.15 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-16 22:48:18,340 epoch 2 - iter 693/773 - loss 0.07966322 - time (sec): 40.88 - samples/sec: 2712.23 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-16 22:48:23,004 epoch 2 - iter 770/773 - loss 0.07901639 - time (sec): 45.55 - samples/sec: 2719.62 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-16 22:48:23,163 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-16 22:48:23,163 EPOCH 2 done: loss 0.0788 - lr: 0.000044
103
+ 2023-10-16 22:48:25,238 DEV : loss 0.04929770901799202 - f1-score (micro avg) 0.7824
104
+ 2023-10-16 22:48:25,251 saving best model
105
+ 2023-10-16 22:48:25,719 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-16 22:48:30,138 epoch 3 - iter 77/773 - loss 0.05283669 - time (sec): 4.42 - samples/sec: 2848.46 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-16 22:48:34,806 epoch 3 - iter 154/773 - loss 0.05183408 - time (sec): 9.09 - samples/sec: 2757.45 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-16 22:48:39,342 epoch 3 - iter 231/773 - loss 0.05153972 - time (sec): 13.62 - samples/sec: 2737.19 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-16 22:48:43,865 epoch 3 - iter 308/773 - loss 0.04919031 - time (sec): 18.14 - samples/sec: 2713.46 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-16 22:48:48,279 epoch 3 - iter 385/773 - loss 0.05182767 - time (sec): 22.56 - samples/sec: 2699.47 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-16 22:48:52,627 epoch 3 - iter 462/773 - loss 0.05174989 - time (sec): 26.91 - samples/sec: 2694.28 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-16 22:48:57,120 epoch 3 - iter 539/773 - loss 0.05137726 - time (sec): 31.40 - samples/sec: 2702.93 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-16 22:49:01,721 epoch 3 - iter 616/773 - loss 0.05160722 - time (sec): 36.00 - samples/sec: 2710.49 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-16 22:49:06,370 epoch 3 - iter 693/773 - loss 0.05267520 - time (sec): 40.65 - samples/sec: 2713.63 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-16 22:49:11,167 epoch 3 - iter 770/773 - loss 0.05271309 - time (sec): 45.45 - samples/sec: 2725.20 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-16 22:49:11,328 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-16 22:49:11,328 EPOCH 3 done: loss 0.0526 - lr: 0.000039
118
+ 2023-10-16 22:49:13,790 DEV : loss 0.06217540055513382 - f1-score (micro avg) 0.765
119
+ 2023-10-16 22:49:13,803 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-16 22:49:18,337 epoch 4 - iter 77/773 - loss 0.03307915 - time (sec): 4.53 - samples/sec: 2597.75 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-16 22:49:23,144 epoch 4 - iter 154/773 - loss 0.03502772 - time (sec): 9.34 - samples/sec: 2708.64 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-16 22:49:27,501 epoch 4 - iter 231/773 - loss 0.03841144 - time (sec): 13.70 - samples/sec: 2704.05 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-16 22:49:31,990 epoch 4 - iter 308/773 - loss 0.03694235 - time (sec): 18.19 - samples/sec: 2696.06 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-16 22:49:36,598 epoch 4 - iter 385/773 - loss 0.03453724 - time (sec): 22.79 - samples/sec: 2720.71 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-16 22:49:41,295 epoch 4 - iter 462/773 - loss 0.03493765 - time (sec): 27.49 - samples/sec: 2732.97 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-16 22:49:45,665 epoch 4 - iter 539/773 - loss 0.03520193 - time (sec): 31.86 - samples/sec: 2712.03 - lr: 0.000035 - momentum: 0.000000
127
+ 2023-10-16 22:49:50,371 epoch 4 - iter 616/773 - loss 0.03666515 - time (sec): 36.57 - samples/sec: 2713.85 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-16 22:49:54,733 epoch 4 - iter 693/773 - loss 0.03600062 - time (sec): 40.93 - samples/sec: 2716.65 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-16 22:49:59,478 epoch 4 - iter 770/773 - loss 0.03587709 - time (sec): 45.67 - samples/sec: 2713.34 - lr: 0.000033 - momentum: 0.000000
130
+ 2023-10-16 22:49:59,640 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-16 22:49:59,640 EPOCH 4 done: loss 0.0358 - lr: 0.000033
132
+ 2023-10-16 22:50:01,833 DEV : loss 0.09437456727027893 - f1-score (micro avg) 0.7548
133
+ 2023-10-16 22:50:01,848 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-16 22:50:06,576 epoch 5 - iter 77/773 - loss 0.02673306 - time (sec): 4.73 - samples/sec: 2600.08 - lr: 0.000033 - momentum: 0.000000
135
+ 2023-10-16 22:50:11,310 epoch 5 - iter 154/773 - loss 0.02324661 - time (sec): 9.46 - samples/sec: 2654.40 - lr: 0.000032 - momentum: 0.000000
136
+ 2023-10-16 22:50:15,933 epoch 5 - iter 231/773 - loss 0.02222597 - time (sec): 14.08 - samples/sec: 2662.75 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-16 22:50:20,768 epoch 5 - iter 308/773 - loss 0.02255704 - time (sec): 18.92 - samples/sec: 2690.90 - lr: 0.000031 - momentum: 0.000000
138
+ 2023-10-16 22:50:25,371 epoch 5 - iter 385/773 - loss 0.02407859 - time (sec): 23.52 - samples/sec: 2700.14 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-16 22:50:29,740 epoch 5 - iter 462/773 - loss 0.02477722 - time (sec): 27.89 - samples/sec: 2727.07 - lr: 0.000030 - momentum: 0.000000
140
+ 2023-10-16 22:50:34,296 epoch 5 - iter 539/773 - loss 0.02515718 - time (sec): 32.45 - samples/sec: 2745.77 - lr: 0.000029 - momentum: 0.000000
141
+ 2023-10-16 22:50:38,637 epoch 5 - iter 616/773 - loss 0.02459067 - time (sec): 36.79 - samples/sec: 2745.62 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-16 22:50:42,904 epoch 5 - iter 693/773 - loss 0.02490799 - time (sec): 41.05 - samples/sec: 2740.82 - lr: 0.000028 - momentum: 0.000000
143
+ 2023-10-16 22:50:47,140 epoch 5 - iter 770/773 - loss 0.02419246 - time (sec): 45.29 - samples/sec: 2737.34 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-16 22:50:47,290 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-16 22:50:47,290 EPOCH 5 done: loss 0.0241 - lr: 0.000028
146
+ 2023-10-16 22:50:49,400 DEV : loss 0.10803093016147614 - f1-score (micro avg) 0.7821
147
+ 2023-10-16 22:50:49,414 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-16 22:50:53,937 epoch 6 - iter 77/773 - loss 0.01084402 - time (sec): 4.52 - samples/sec: 2709.74 - lr: 0.000027 - momentum: 0.000000
149
+ 2023-10-16 22:50:58,588 epoch 6 - iter 154/773 - loss 0.01565568 - time (sec): 9.17 - samples/sec: 2682.37 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-16 22:51:03,088 epoch 6 - iter 231/773 - loss 0.01799942 - time (sec): 13.67 - samples/sec: 2671.24 - lr: 0.000026 - momentum: 0.000000
151
+ 2023-10-16 22:51:07,640 epoch 6 - iter 308/773 - loss 0.01771408 - time (sec): 18.22 - samples/sec: 2707.41 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-16 22:51:11,960 epoch 6 - iter 385/773 - loss 0.01871910 - time (sec): 22.54 - samples/sec: 2739.49 - lr: 0.000025 - momentum: 0.000000
153
+ 2023-10-16 22:51:16,757 epoch 6 - iter 462/773 - loss 0.01928403 - time (sec): 27.34 - samples/sec: 2747.99 - lr: 0.000024 - momentum: 0.000000
154
+ 2023-10-16 22:51:21,399 epoch 6 - iter 539/773 - loss 0.01888480 - time (sec): 31.98 - samples/sec: 2721.82 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-16 22:51:25,908 epoch 6 - iter 616/773 - loss 0.02050580 - time (sec): 36.49 - samples/sec: 2709.59 - lr: 0.000023 - momentum: 0.000000
156
+ 2023-10-16 22:51:30,516 epoch 6 - iter 693/773 - loss 0.01943211 - time (sec): 41.10 - samples/sec: 2701.94 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-16 22:51:35,177 epoch 6 - iter 770/773 - loss 0.01958240 - time (sec): 45.76 - samples/sec: 2708.62 - lr: 0.000022 - momentum: 0.000000
158
+ 2023-10-16 22:51:35,336 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-16 22:51:35,337 EPOCH 6 done: loss 0.0196 - lr: 0.000022
160
+ 2023-10-16 22:51:37,400 DEV : loss 0.1094818264245987 - f1-score (micro avg) 0.7815
161
+ 2023-10-16 22:51:37,415 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-16 22:51:41,880 epoch 7 - iter 77/773 - loss 0.00528052 - time (sec): 4.46 - samples/sec: 2697.30 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-16 22:51:46,322 epoch 7 - iter 154/773 - loss 0.00899972 - time (sec): 8.91 - samples/sec: 2684.24 - lr: 0.000021 - momentum: 0.000000
164
+ 2023-10-16 22:51:50,741 epoch 7 - iter 231/773 - loss 0.01093330 - time (sec): 13.33 - samples/sec: 2729.67 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-16 22:51:55,196 epoch 7 - iter 308/773 - loss 0.01394532 - time (sec): 17.78 - samples/sec: 2733.90 - lr: 0.000020 - momentum: 0.000000
166
+ 2023-10-16 22:51:59,811 epoch 7 - iter 385/773 - loss 0.01538091 - time (sec): 22.39 - samples/sec: 2741.05 - lr: 0.000019 - momentum: 0.000000
167
+ 2023-10-16 22:52:04,408 epoch 7 - iter 462/773 - loss 0.01469576 - time (sec): 26.99 - samples/sec: 2754.87 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-16 22:52:09,001 epoch 7 - iter 539/773 - loss 0.01528889 - time (sec): 31.58 - samples/sec: 2752.79 - lr: 0.000018 - momentum: 0.000000
169
+ 2023-10-16 22:52:13,667 epoch 7 - iter 616/773 - loss 0.01471401 - time (sec): 36.25 - samples/sec: 2731.80 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-16 22:52:18,014 epoch 7 - iter 693/773 - loss 0.01434424 - time (sec): 40.60 - samples/sec: 2741.98 - lr: 0.000017 - momentum: 0.000000
171
+ 2023-10-16 22:52:22,595 epoch 7 - iter 770/773 - loss 0.01514214 - time (sec): 45.18 - samples/sec: 2744.16 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-16 22:52:22,749 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-16 22:52:22,749 EPOCH 7 done: loss 0.0151 - lr: 0.000017
174
+ 2023-10-16 22:52:24,858 DEV : loss 0.10991214215755463 - f1-score (micro avg) 0.7794
175
+ 2023-10-16 22:52:24,871 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-16 22:52:29,258 epoch 8 - iter 77/773 - loss 0.00333898 - time (sec): 4.39 - samples/sec: 2595.49 - lr: 0.000016 - momentum: 0.000000
177
+ 2023-10-16 22:52:34,059 epoch 8 - iter 154/773 - loss 0.00641512 - time (sec): 9.19 - samples/sec: 2693.65 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-16 22:52:39,164 epoch 8 - iter 231/773 - loss 0.00625228 - time (sec): 14.29 - samples/sec: 2647.99 - lr: 0.000015 - momentum: 0.000000
179
+ 2023-10-16 22:52:43,962 epoch 8 - iter 308/773 - loss 0.00659826 - time (sec): 19.09 - samples/sec: 2672.30 - lr: 0.000014 - momentum: 0.000000
180
+ 2023-10-16 22:52:48,443 epoch 8 - iter 385/773 - loss 0.00711413 - time (sec): 23.57 - samples/sec: 2675.76 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-16 22:52:52,773 epoch 8 - iter 462/773 - loss 0.00795173 - time (sec): 27.90 - samples/sec: 2677.65 - lr: 0.000013 - momentum: 0.000000
182
+ 2023-10-16 22:52:57,154 epoch 8 - iter 539/773 - loss 0.00841455 - time (sec): 32.28 - samples/sec: 2698.48 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-16 22:53:01,874 epoch 8 - iter 616/773 - loss 0.00849543 - time (sec): 37.00 - samples/sec: 2695.17 - lr: 0.000012 - momentum: 0.000000
184
+ 2023-10-16 22:53:06,448 epoch 8 - iter 693/773 - loss 0.00897091 - time (sec): 41.58 - samples/sec: 2698.10 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-16 22:53:10,765 epoch 8 - iter 770/773 - loss 0.00880887 - time (sec): 45.89 - samples/sec: 2696.82 - lr: 0.000011 - momentum: 0.000000
186
+ 2023-10-16 22:53:10,931 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-16 22:53:10,931 EPOCH 8 done: loss 0.0088 - lr: 0.000011
188
+ 2023-10-16 22:53:13,015 DEV : loss 0.11286821216344833 - f1-score (micro avg) 0.7934
189
+ 2023-10-16 22:53:13,029 saving best model
190
+ 2023-10-16 22:53:13,480 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-16 22:53:18,090 epoch 9 - iter 77/773 - loss 0.01038545 - time (sec): 4.60 - samples/sec: 2556.72 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-16 22:53:22,785 epoch 9 - iter 154/773 - loss 0.00653013 - time (sec): 9.29 - samples/sec: 2558.56 - lr: 0.000010 - momentum: 0.000000
193
+ 2023-10-16 22:53:27,393 epoch 9 - iter 231/773 - loss 0.00606729 - time (sec): 13.90 - samples/sec: 2674.63 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-16 22:53:31,904 epoch 9 - iter 308/773 - loss 0.00657523 - time (sec): 18.41 - samples/sec: 2669.14 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-16 22:53:36,500 epoch 9 - iter 385/773 - loss 0.00675286 - time (sec): 23.01 - samples/sec: 2703.69 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-16 22:53:40,928 epoch 9 - iter 462/773 - loss 0.00630020 - time (sec): 27.44 - samples/sec: 2707.69 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-16 22:53:45,461 epoch 9 - iter 539/773 - loss 0.00593099 - time (sec): 31.97 - samples/sec: 2721.81 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-16 22:53:49,850 epoch 9 - iter 616/773 - loss 0.00631036 - time (sec): 36.36 - samples/sec: 2723.02 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-16 22:53:54,199 epoch 9 - iter 693/773 - loss 0.00613278 - time (sec): 40.71 - samples/sec: 2731.95 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-16 22:53:58,725 epoch 9 - iter 770/773 - loss 0.00617994 - time (sec): 45.23 - samples/sec: 2740.57 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-16 22:53:58,874 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-16 22:53:58,874 EPOCH 9 done: loss 0.0062 - lr: 0.000006
203
+ 2023-10-16 22:54:00,998 DEV : loss 0.11028449237346649 - f1-score (micro avg) 0.8092
204
+ 2023-10-16 22:54:01,011 saving best model
205
+ 2023-10-16 22:54:01,454 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-16 22:54:05,936 epoch 10 - iter 77/773 - loss 0.00135028 - time (sec): 4.48 - samples/sec: 2727.49 - lr: 0.000005 - momentum: 0.000000
207
+ 2023-10-16 22:54:10,426 epoch 10 - iter 154/773 - loss 0.00310208 - time (sec): 8.97 - samples/sec: 2771.61 - lr: 0.000005 - momentum: 0.000000
208
+ 2023-10-16 22:54:14,892 epoch 10 - iter 231/773 - loss 0.00391362 - time (sec): 13.43 - samples/sec: 2768.89 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-16 22:54:19,512 epoch 10 - iter 308/773 - loss 0.00329852 - time (sec): 18.05 - samples/sec: 2754.06 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-16 22:54:23,936 epoch 10 - iter 385/773 - loss 0.00380931 - time (sec): 22.48 - samples/sec: 2768.94 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-16 22:54:28,526 epoch 10 - iter 462/773 - loss 0.00368967 - time (sec): 27.07 - samples/sec: 2765.33 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-16 22:54:33,179 epoch 10 - iter 539/773 - loss 0.00347435 - time (sec): 31.72 - samples/sec: 2740.59 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-16 22:54:37,718 epoch 10 - iter 616/773 - loss 0.00375376 - time (sec): 36.26 - samples/sec: 2740.49 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-16 22:54:42,218 epoch 10 - iter 693/773 - loss 0.00367638 - time (sec): 40.76 - samples/sec: 2737.65 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-16 22:54:46,786 epoch 10 - iter 770/773 - loss 0.00361332 - time (sec): 45.33 - samples/sec: 2734.62 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-16 22:54:46,935 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-16 22:54:46,935 EPOCH 10 done: loss 0.0036 - lr: 0.000000
218
+ 2023-10-16 22:54:49,087 DEV : loss 0.11204110831022263 - f1-score (micro avg) 0.7975
219
+ 2023-10-16 22:54:49,470 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-16 22:54:49,471 Loading model from best epoch ...
221
+ 2023-10-16 22:54:51,168 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
222
+ 2023-10-16 22:54:57,525
223
+ Results:
224
+ - F-score (micro) 0.8227
225
+ - F-score (macro) 0.7472
226
+ - Accuracy 0.7223
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ LOC 0.8625 0.8552 0.8588 946
232
+ BUILDING 0.6836 0.6541 0.6685 185
233
+ STREET 0.7143 0.7143 0.7143 56
234
+
235
+ micro avg 0.8284 0.8172 0.8227 1187
236
+ macro avg 0.7535 0.7412 0.7472 1187
237
+ weighted avg 0.8276 0.8172 0.8223 1187
238
+
239
+ 2023-10-16 22:54:57,525 ----------------------------------------------------------------------------------------------------