Hugging Face's logo

harkase
/
mbti_4axis_koelectra

Text Classification
Transformers
Safetensors
electra
Generated from Trainer
Model card Files
xet
 Community
harkase commited on
Commit
9e05e1a
·
verified ·
1 Parent(s): 55a08aa

End of training

Browse files
Files changed (4) hide show
  1. README.md +18 -18
  2. config.json +2 -2
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -8,20 +8,20 @@ metrics:
8
  - accuracy
9
  - f1
10
  model-index:
11
- - name: results
12
  results: []
13
  ---
14
 
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
  should probably proofread and complete it, then remove this comment. -->
17
 
18
- # results
19
 
20
  This model is a fine-tuned version of [monologg/koelectra-base-v3-discriminator](https://huggingface.co/monologg/koelectra-base-v3-discriminator) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.6599
23
- - Accuracy: 0.6028
24
- - F1: 0.5504
25
 
26
  ## Model description
27
 
@@ -33,35 +33,35 @@ More information needed
33
 
34
  ## Training and evaluation data
35
 
36
- - [github/dev-jaemin/Korean-MBTI-Conversation-Dataset](https://github.com/dev-jaemin/Korean-MBTI-Conversation-Dataset)
37
- - use data in qna_cleaned.tsv, multiple_qna_cleaned.tsv
38
- - refine [answer, a_mbti]
39
- - concat both refiend data
40
- - Training and evaluation data split 8:2 ratio
41
  ## Training procedure
42
 
43
  ### Training hyperparameters
44
 
45
  The following hyperparameters were used during training:
46
- - learning_rate: 4e-05
47
  - train_batch_size: 32
48
  - eval_batch_size: 32
49
  - seed: 42
50
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
 
52
  - num_epochs: 5
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
57
  |:-------------:|:------:|:----:|:---------------:|:--------:|:------:|
58
- | 0.6604 | 0.0669 | 200 | 0.6573 | 0.6132 | 0.5419 |
59
- | 0.6629 | 0.1339 | 400 | 0.6563 | 0.6200 | 0.5321 |
60
- | 0.6549 | 0.2008 | 600 | 0.6542 | 0.6237 | 0.5145 |
61
- | 0.6587 | 0.2677 | 800 | 0.6599 | 0.6028 | 0.5504 |
62
- | 0.6631 | 0.3347 | 1000 | 0.6599 | 0.6028 | 0.5503 |
63
- | 0.6584 | 0.4016 | 1200 | 0.6594 | 0.6105 | 0.3872 |
64
- | 0.6588 | 0.4685 | 1400 | 0.6590 | 0.6105 | 0.3872 |
 
 
65
 
66
 
67
  ### Framework versions
 
8
  - accuracy
9
  - f1
10
  model-index:
11
+ - name: mbti_4axis_koelectra
12
  results: []
13
  ---
14
 
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
  should probably proofread and complete it, then remove this comment. -->
17
 
18
+ # mbti_4axis_koelectra
19
 
20
  This model is a fine-tuned version of [monologg/koelectra-base-v3-discriminator](https://huggingface.co/monologg/koelectra-base-v3-discriminator) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.6614
23
+ - Accuracy: 0.6027
24
+ - F1: 0.6878
25
 
26
  ## Model description
27
 
 
33
 
34
  ## Training and evaluation data
35
 
36
+ More information needed
37
+
 
 
 
38
  ## Training procedure
39
 
40
  ### Training hyperparameters
41
 
42
  The following hyperparameters were used during training:
43
+ - learning_rate: 3e-05
44
  - train_batch_size: 32
45
  - eval_batch_size: 32
46
  - seed: 42
47
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_ratio: 0.1
50
  - num_epochs: 5
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
55
  |:-------------:|:------:|:----:|:---------------:|:--------:|:------:|
56
+ | 0.6586 | 0.3347 | 1000 | 0.6545 | 0.5338 | 0.5518 |
57
+ | 0.6605 | 0.6693 | 2000 | 0.6594 | 0.5354 | 0.5261 |
58
+ | 0.6615 | 1.0040 | 3000 | 0.6617 | 0.5166 | 0.3746 |
59
+ | 0.663 | 1.3387 | 4000 | 0.6614 | 0.5308 | 0.6315 |
60
+ | 0.6615 | 1.6734 | 5000 | 0.6617 | 0.5227 | 0.6865 |
61
+ | 0.6592 | 2.0080 | 6000 | 0.6614 | 0.6027 | 0.6878 |
62
+ | 0.6484 | 2.3427 | 7000 | 0.6488 | 0.5657 | 0.5889 |
63
+ | 0.666 | 2.6774 | 8000 | 0.6614 | 0.4390 | 0.4533 |
64
+ | 0.6466 | 3.0120 | 9000 | 0.6483 | 0.5402 | 0.5783 |
65
 
66
 
67
  ### Framework versions
config.json CHANGED
@@ -24,9 +24,9 @@
24
  "intermediate_size": 3072,
25
  "label2id": {
26
  "IE": 0,
 
27
  "SN": 1,
28
- "TF": 2,
29
- "JP": 3
30
  },
31
  "layer_norm_eps": 1e-12,
32
  "max_position_embeddings": 512,
 
24
  "intermediate_size": 3072,
25
  "label2id": {
26
  "IE": 0,
27
+ "JP": 3,
28
  "SN": 1,
29
+ "TF": 2
 
30
  },
31
  "layer_norm_eps": 1e-12,
32
  "max_position_embeddings": 512,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f1a9175fe1b9ea66f81fdf1c4af35ba1295ce330cbb6279150b15164d20aafd0
3
  size 451721824
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec5137cf34017c6a6b4a7763268336cfb16a75e3e001458fca6fd1eb7c153e0f
3
  size 451721824
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:72483dff6de8d981085c7c018954068b3ad824b171f1b7a69a9a9751bbe43c37
3
  size 5713
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a98d146f2c9d42d9028a38e621b0eb0e5b6e3c9018c9b564631c824594d3573a
3
  size 5713