Hugging Face's logo

eunyounglee
/
hana-persona-emotion-bert-simple-finetuning-1

Text Classification
Transformers
Safetensors
bert
Generated from Trainer
Model card Files
xet
Community
eunyounglee commited on
Commit
46e0cc8
·
verified ·
1 Parent(s): 5f07125

End of training

Browse files
Files changed (1) hide show
  1. README.md +13 -48
README.md CHANGED
@@ -18,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [klue/bert-base](https://huggingface.co/klue/bert-base) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.2703
22
- - Accuracy: 0.9687
23
- - F1: 0.9672
24
 
25
  ## Model description
26
 
@@ -51,51 +51,16 @@ The following hyperparameters were used during training:
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
53
  |:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|
54
- | 0.1955 | 0.22 | 500 | 0.1092 | 0.9696 | 0.9561 |
55
- | 0.1284 | 0.44 | 1000 | 0.1606 | 0.9685 | 0.9531 |
56
- | 0.109 | 0.66 | 1500 | 0.1018 | 0.9709 | 0.9640 |
57
- | 0.109 | 0.88 | 2000 | 0.1085 | 0.9711 | 0.9596 |
58
- | 0.0951 | 1.09 | 2500 | 0.1014 | 0.9727 | 0.9673 |
59
- | 0.0828 | 1.31 | 3000 | 0.1000 | 0.9724 | 0.9687 |
60
- | 0.0785 | 1.53 | 3500 | 0.0969 | 0.9724 | 0.9672 |
61
- | 0.079 | 1.75 | 4000 | 0.1241 | 0.9730 | 0.9672 |
62
- | 0.0934 | 1.97 | 4500 | 0.1276 | 0.9733 | 0.9673 |
63
- | 0.0524 | 2.19 | 5000 | 0.1455 | 0.9702 | 0.9675 |
64
- | 0.0544 | 2.41 | 5500 | 0.1427 | 0.9692 | 0.9666 |
65
- | 0.0565 | 2.63 | 6000 | 0.1322 | 0.9674 | 0.9662 |
66
- | 0.0594 | 2.85 | 6500 | 0.1333 | 0.9658 | 0.9654 |
67
- | 0.0491 | 3.06 | 7000 | 0.1487 | 0.9710 | 0.9687 |
68
- | 0.0284 | 3.28 | 7500 | 0.1677 | 0.9723 | 0.9678 |
69
- | 0.0319 | 3.5 | 8000 | 0.1719 | 0.9727 | 0.9681 |
70
- | 0.0318 | 3.72 | 8500 | 0.1831 | 0.9704 | 0.9680 |
71
- | 0.0266 | 3.94 | 9000 | 0.1806 | 0.9697 | 0.9668 |
72
- | 0.0225 | 4.16 | 9500 | 0.1989 | 0.9704 | 0.9668 |
73
- | 0.0161 | 4.38 | 10000 | 0.2000 | 0.9681 | 0.9669 |
74
- | 0.0219 | 4.6 | 10500 | 0.2081 | 0.9705 | 0.9672 |
75
- | 0.0252 | 4.82 | 11000 | 0.1979 | 0.9718 | 0.9657 |
76
- | 0.0215 | 5.04 | 11500 | 0.1961 | 0.9708 | 0.9673 |
77
- | 0.0088 | 5.25 | 12000 | 0.2344 | 0.9676 | 0.9657 |
78
- | 0.0115 | 5.47 | 12500 | 0.2307 | 0.9724 | 0.9681 |
79
- | 0.0138 | 5.69 | 13000 | 0.2238 | 0.9675 | 0.9669 |
80
- | 0.016 | 5.91 | 13500 | 0.2258 | 0.9684 | 0.9669 |
81
- | 0.0078 | 6.13 | 14000 | 0.2354 | 0.9698 | 0.9677 |
82
- | 0.0062 | 6.35 | 14500 | 0.2387 | 0.9705 | 0.9670 |
83
- | 0.0103 | 6.57 | 15000 | 0.2485 | 0.9691 | 0.9669 |
84
- | 0.0065 | 6.79 | 15500 | 0.2527 | 0.9702 | 0.9675 |
85
- | 0.0145 | 7.01 | 16000 | 0.2278 | 0.9703 | 0.9678 |
86
- | 0.0054 | 7.22 | 16500 | 0.2329 | 0.9701 | 0.9681 |
87
- | 0.0047 | 7.44 | 17000 | 0.2438 | 0.9699 | 0.9672 |
88
- | 0.0071 | 7.66 | 17500 | 0.2501 | 0.9713 | 0.9681 |
89
- | 0.0058 | 7.88 | 18000 | 0.2470 | 0.9686 | 0.9672 |
90
- | 0.0045 | 8.1 | 18500 | 0.2436 | 0.9701 | 0.9672 |
91
- | 0.0034 | 8.32 | 19000 | 0.2662 | 0.9673 | 0.9661 |
92
- | 0.0035 | 8.54 | 19500 | 0.2645 | 0.9677 | 0.9664 |
93
- | 0.002 | 8.76 | 20000 | 0.2652 | 0.9696 | 0.9681 |
94
- | 0.0045 | 8.98 | 20500 | 0.2605 | 0.9702 | 0.9680 |
95
- | 0.0021 | 9.19 | 21000 | 0.2703 | 0.9691 | 0.9675 |
96
- | 0.0015 | 9.41 | 21500 | 0.2711 | 0.9697 | 0.9670 |
97
- | 0.0025 | 9.63 | 22000 | 0.2712 | 0.9691 | 0.9671 |
98
- | 0.0046 | 9.85 | 22500 | 0.2703 | 0.9687 | 0.9672 |
99
 
100
 
101
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [klue/bert-base](https://huggingface.co/klue/bert-base) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.5295
22
+ - Accuracy: 0.8043
23
+ - F1: 0.8041
24
 
25
  ## Model description
26
 
 
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
53
  |:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|
54
+ | 0.4983 | 1.0 | 2284 | 0.4753 | 0.8052 | 0.8038 |
55
+ | 0.3728 | 2.0 | 4568 | 0.4923 | 0.8098 | 0.8078 |
56
+ | 0.2827 | 3.0 | 6852 | 0.5351 | 0.8070 | 0.8074 |
57
+ | 0.1906 | 4.0 | 9136 | 0.7375 | 0.8106 | 0.8099 |
58
+ | 0.155 | 5.0 | 11420 | 0.8906 | 0.8041 | 0.8040 |
59
+ | 0.1087 | 6.0 | 13704 | 1.1059 | 0.8050 | 0.8050 |
60
+ | 0.0732 | 7.0 | 15988 | 1.2882 | 0.8049 | 0.8053 |
61
+ | 0.0632 | 8.0 | 18272 | 1.3800 | 0.8066 | 0.8062 |
62
+ | 0.0369 | 9.0 | 20556 | 1.4939 | 0.8035 | 0.8028 |
63
+ | 0.0159 | 10.0 | 22840 | 1.5295 | 0.8043 | 0.8041 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
64
 
65
 
66
  ### Framework versions