Hugging Face's logo

vuihocrnd
/
cls-comment-phobert-base-v2-v3.2

Text Classification
Transformers
Safetensors
roberta
Generated from Trainer
Model card Files Community
tiennguyenbnbk commited on
Commit
137e791
·
verified ·
1 Parent(s): 771ab9c

Model save

Browse files
Files changed (2) hide show
  1. README.md +46 -44
  2. model.safetensors +1 -1
README.md CHANGED
@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [vinai/phobert-base-v2](https://huggingface.co/vinai/phobert-base-v2) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.6495
22
- - Accuracy: 0.9313
23
- - F1 Score: 0.9168
24
- - Recall: 0.9161
25
- - Precision: 0.9175
26
 
27
  ## Model description
28
 
@@ -51,50 +51,52 @@ The following hyperparameters were used during training:
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_ratio: 0.1
53
  - training_steps: 4000
54
- - label_smoothing_factor: 0.1
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall | Precision |
59
  |:-------------:|:-------:|:----:|:---------------:|:--------:|:--------:|:------:|:---------:|
60
- | 1.8746 | 0.8658 | 100 | 1.7251 | 0.4004 | 0.0817 | 0.1429 | 0.0572 |
61
- | 1.591 | 1.7316 | 200 | 1.3510 | 0.6455 | 0.2863 | 0.3127 | 0.2775 |
62
- | 1.2461 | 2.5974 | 300 | 1.0290 | 0.7813 | 0.5633 | 0.5670 | 0.5618 |
63
- | 0.9936 | 3.4632 | 400 | 0.8636 | 0.8280 | 0.6071 | 0.6261 | 0.7310 |
64
- | 0.846 | 4.3290 | 500 | 0.7818 | 0.8606 | 0.6996 | 0.7001 | 0.7379 |
65
- | 0.7528 | 5.1948 | 600 | 0.7184 | 0.8845 | 0.7881 | 0.7747 | 0.8855 |
66
- | 0.6829 | 6.0606 | 700 | 0.6788 | 0.9065 | 0.8721 | 0.8622 | 0.8877 |
67
- | 0.6318 | 6.9264 | 800 | 0.6686 | 0.9060 | 0.8772 | 0.8843 | 0.8735 |
68
- | 0.5946 | 7.7922 | 900 | 0.6710 | 0.9055 | 0.8830 | 0.8779 | 0.8941 |
69
- | 0.5787 | 8.6580 | 1000 | 0.6430 | 0.9231 | 0.9046 | 0.8969 | 0.9136 |
70
- | 0.5465 | 9.5238 | 1100 | 0.6391 | 0.9234 | 0.9029 | 0.8996 | 0.9075 |
71
- | 0.5351 | 10.3896 | 1200 | 0.6590 | 0.9163 | 0.9010 | 0.9029 | 0.9032 |
72
- | 0.5253 | 11.2554 | 1300 | 0.6566 | 0.9171 | 0.8992 | 0.9017 | 0.9002 |
73
- | 0.5129 | 12.1212 | 1400 | 0.6489 | 0.9215 | 0.8995 | 0.9157 | 0.8853 |
74
- | 0.507 | 12.9870 | 1500 | 0.6600 | 0.9188 | 0.8960 | 0.9084 | 0.8851 |
75
- | 0.498 | 13.8528 | 1600 | 0.6436 | 0.9261 | 0.9038 | 0.8986 | 0.9106 |
76
- | 0.4928 | 14.7186 | 1700 | 0.6421 | 0.9283 | 0.9090 | 0.9139 | 0.9051 |
77
- | 0.488 | 15.5844 | 1800 | 0.6527 | 0.9242 | 0.9070 | 0.9171 | 0.8984 |
78
- | 0.4855 | 16.4502 | 1900 | 0.6503 | 0.9288 | 0.9104 | 0.9118 | 0.9104 |
79
- | 0.4827 | 17.3160 | 2000 | 0.6523 | 0.9264 | 0.9083 | 0.9077 | 0.9101 |
80
- | 0.4806 | 18.1818 | 2100 | 0.6727 | 0.9226 | 0.9070 | 0.9107 | 0.9069 |
81
- | 0.475 | 19.0476 | 2200 | 0.6789 | 0.9188 | 0.8988 | 0.9076 | 0.8925 |
82
- | 0.4769 | 19.9134 | 2300 | 0.6616 | 0.9239 | 0.9047 | 0.9115 | 0.8995 |
83
- | 0.4717 | 20.7792 | 2400 | 0.6539 | 0.9267 | 0.9044 | 0.9051 | 0.9048 |
84
- | 0.4714 | 21.6450 | 2500 | 0.6580 | 0.9286 | 0.9087 | 0.9210 | 0.8978 |
85
- | 0.4674 | 22.5108 | 2600 | 0.6538 | 0.9280 | 0.9102 | 0.9121 | 0.9088 |
86
- | 0.4647 | 23.3766 | 2700 | 0.6711 | 0.9237 | 0.9094 | 0.9186 | 0.9021 |
87
- | 0.4641 | 24.2424 | 2800 | 0.6495 | 0.9313 | 0.9168 | 0.9161 | 0.9175 |
88
- | 0.4661 | 25.1082 | 2900 | 0.6524 | 0.9291 | 0.9107 | 0.9121 | 0.9097 |
89
- | 0.4642 | 25.9740 | 3000 | 0.6616 | 0.9272 | 0.9086 | 0.9158 | 0.9024 |
90
- | 0.4634 | 26.8398 | 3100 | 0.6608 | 0.9272 | 0.9100 | 0.9157 | 0.9048 |
91
- | 0.4621 | 27.7056 | 3200 | 0.6604 | 0.9302 | 0.9127 | 0.9180 | 0.9082 |
92
- | 0.4607 | 28.5714 | 3300 | 0.6679 | 0.9275 | 0.9095 | 0.9158 | 0.9042 |
93
- | 0.4605 | 29.4372 | 3400 | 0.6584 | 0.9294 | 0.9128 | 0.9142 | 0.9121 |
94
- | 0.4594 | 30.3030 | 3500 | 0.6636 | 0.9275 | 0.9103 | 0.9135 | 0.9078 |
95
- | 0.4603 | 31.1688 | 3600 | 0.6612 | 0.9283 | 0.9114 | 0.9141 | 0.9095 |
96
- | 0.4589 | 32.0346 | 3700 | 0.6628 | 0.9283 | 0.9107 | 0.9196 | 0.9029 |
97
- | 0.4594 | 32.9004 | 3800 | 0.6590 | 0.9294 | 0.9121 | 0.9148 | 0.9102 |
 
 
98
 
99
 
100
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [vinai/phobert-base-v2](https://huggingface.co/vinai/phobert-base-v2) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.4661
22
+ - Accuracy: 0.9399
23
+ - F1 Score: 0.9222
24
+ - Recall: 0.9304
25
+ - Precision: 0.9146
26
 
27
  ## Model description
28
 
 
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_ratio: 0.1
53
  - training_steps: 4000
54
+ - label_smoothing_factor: 0.05
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall | Precision |
59
  |:-------------:|:-------:|:----:|:---------------:|:--------:|:--------:|:------:|:---------:|
60
+ | 1.854 | 0.8696 | 100 | 1.6864 | 0.3995 | 0.0816 | 0.1429 | 0.0571 |
61
+ | 1.5604 | 1.7391 | 200 | 1.3090 | 0.6106 | 0.2504 | 0.2819 | 0.2671 |
62
+ | 1.1923 | 2.6087 | 300 | 0.9517 | 0.7864 | 0.5715 | 0.5777 | 0.5674 |
63
+ | 0.9214 | 3.4783 | 400 | 0.7528 | 0.8372 | 0.6124 | 0.6292 | 0.5977 |
64
+ | 0.758 | 4.3478 | 500 | 0.6326 | 0.8668 | 0.6659 | 0.6622 | 0.7570 |
65
+ | 0.6389 | 5.2174 | 600 | 0.5610 | 0.8913 | 0.7804 | 0.7578 | 0.8994 |
66
+ | 0.5577 | 6.0870 | 700 | 0.5189 | 0.9098 | 0.8772 | 0.8752 | 0.8849 |
67
+ | 0.4924 | 6.9565 | 800 | 0.4865 | 0.9158 | 0.8883 | 0.8831 | 0.8952 |
68
+ | 0.4466 | 7.8261 | 900 | 0.4718 | 0.9234 | 0.9011 | 0.9000 | 0.9036 |
69
+ | 0.4074 | 8.6957 | 1000 | 0.4614 | 0.9242 | 0.9037 | 0.9023 | 0.9075 |
70
+ | 0.3986 | 9.5652 | 1100 | 0.4673 | 0.9236 | 0.9049 | 0.9137 | 0.8981 |
71
+ | 0.3673 | 10.4348 | 1200 | 0.4504 | 0.9307 | 0.9134 | 0.9059 | 0.9213 |
72
+ | 0.3579 | 11.3043 | 1300 | 0.4478 | 0.9315 | 0.9145 | 0.9180 | 0.9123 |
73
+ | 0.3408 | 12.1739 | 1400 | 0.4463 | 0.9315 | 0.9126 | 0.9101 | 0.9161 |
74
+ | 0.3316 | 13.0435 | 1500 | 0.4618 | 0.9304 | 0.9114 | 0.9233 | 0.9015 |
75
+ | 0.321 | 13.9130 | 1600 | 0.4430 | 0.9361 | 0.9178 | 0.9203 | 0.9163 |
76
+ | 0.3113 | 14.7826 | 1700 | 0.4418 | 0.9394 | 0.9206 | 0.9233 | 0.9180 |
77
+ | 0.3085 | 15.6522 | 1800 | 0.4470 | 0.9391 | 0.9214 | 0.9226 | 0.9207 |
78
+ | 0.304 | 16.5217 | 1900 | 0.4500 | 0.9370 | 0.9171 | 0.9136 | 0.9217 |
79
+ | 0.2967 | 17.3913 | 2000 | 0.4605 | 0.9345 | 0.9149 | 0.9135 | 0.9175 |
80
+ | 0.2956 | 18.2609 | 2100 | 0.4595 | 0.9348 | 0.9145 | 0.9238 | 0.9061 |
81
+ | 0.2874 | 19.1304 | 2200 | 0.4620 | 0.9378 | 0.9185 | 0.9200 | 0.9179 |
82
+ | 0.2891 | 20.0 | 2300 | 0.4602 | 0.9361 | 0.9167 | 0.9182 | 0.9166 |
83
+ | 0.2862 | 20.8696 | 2400 | 0.4600 | 0.9337 | 0.9149 | 0.9169 | 0.9133 |
84
+ | 0.2851 | 21.7391 | 2500 | 0.4556 | 0.9372 | 0.9184 | 0.9282 | 0.9095 |
85
+ | 0.2798 | 22.6087 | 2600 | 0.4586 | 0.9405 | 0.9223 | 0.9296 | 0.9156 |
86
+ | 0.2787 | 23.4783 | 2700 | 0.4547 | 0.9408 | 0.9250 | 0.9280 | 0.9223 |
87
+ | 0.2806 | 24.3478 | 2800 | 0.4590 | 0.9380 | 0.9188 | 0.9259 | 0.9124 |
88
+ | 0.2768 | 25.2174 | 2900 | 0.4618 | 0.9361 | 0.9188 | 0.9180 | 0.9204 |
89
+ | 0.2773 | 26.0870 | 3000 | 0.4579 | 0.9380 | 0.9203 | 0.9231 | 0.9177 |
90
+ | 0.2724 | 26.9565 | 3100 | 0.4632 | 0.9408 | 0.9253 | 0.9343 | 0.9169 |
91
+ | 0.2716 | 27.8261 | 3200 | 0.4744 | 0.9364 | 0.9205 | 0.9338 | 0.9088 |
92
+ | 0.2705 | 28.6957 | 3300 | 0.4600 | 0.9402 | 0.9218 | 0.9282 | 0.9159 |
93
+ | 0.2682 | 29.5652 | 3400 | 0.4689 | 0.9380 | 0.9196 | 0.9256 | 0.9142 |
94
+ | 0.2718 | 30.4348 | 3500 | 0.4682 | 0.9413 | 0.9226 | 0.9288 | 0.9173 |
95
+ | 0.2694 | 31.3043 | 3600 | 0.4660 | 0.9386 | 0.9201 | 0.9289 | 0.9119 |
96
+ | 0.2678 | 32.1739 | 3700 | 0.4613 | 0.9405 | 0.9216 | 0.9239 | 0.9195 |
97
+ | 0.2679 | 33.0435 | 3800 | 0.4631 | 0.9408 | 0.9224 | 0.9280 | 0.9171 |
98
+ | 0.2681 | 33.9130 | 3900 | 0.4643 | 0.9402 | 0.9223 | 0.9299 | 0.9152 |
99
+ | 0.2685 | 34.7826 | 4000 | 0.4661 | 0.9399 | 0.9222 | 0.9304 | 0.9146 |
100
 
101
 
102
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d14336722075e6f7bd11bedbe7c39046ed2caef32eec5f749e0a29211f22d557
3
  size 540038764
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bc8489e1d1ab86bcea7632285d12e26be8f89dbb627a88310bbb96d546f1f03
3
  size 540038764