update model card README.md
Browse files
README.md
CHANGED
|
@@ -21,7 +21,7 @@ model-index:
|
|
| 21 |
metrics:
|
| 22 |
- name: F1
|
| 23 |
type: f1
|
| 24 |
-
value: 0.
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
This model is a fine-tuned version of [skt/kogpt2-base-v2](https://huggingface.co/skt/kogpt2-base-v2) on the klue dataset.
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
-
- Loss:
|
| 35 |
-
- F1: 0.
|
| 36 |
|
| 37 |
## Model description
|
| 38 |
|
|
@@ -57,42 +57,15 @@ The following hyperparameters were used during training:
|
|
| 57 |
- seed: 42
|
| 58 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 59 |
- lr_scheduler_type: linear
|
| 60 |
-
- num_epochs:
|
| 61 |
|
| 62 |
### Training results
|
| 63 |
|
| 64 |
-
| Training Loss | Epoch | Step
|
| 65 |
-
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.2997 | 4.0 | 3504 | 0.4651 | 0.3192 |
|
| 70 |
-
| 0.2662 | 5.0 | 4380 | 0.4429 | 0.3323 |
|
| 71 |
-
| 0.2393 | 6.0 | 5256 | 0.4527 | 0.3588 |
|
| 72 |
-
| 0.2161 | 7.0 | 6132 | 0.4748 | 0.3580 |
|
| 73 |
-
| 0.1974 | 8.0 | 7008 | 0.4957 | 0.3369 |
|
| 74 |
-
| 0.1807 | 9.0 | 7884 | 0.4911 | 0.3673 |
|
| 75 |
-
| 0.1684 | 10.0 | 8760 | 0.5136 | 0.3597 |
|
| 76 |
-
| 0.1531 | 11.0 | 9636 | 0.4972 | 0.3681 |
|
| 77 |
-
| 0.1405 | 12.0 | 10512 | 0.5515 | 0.3766 |
|
| 78 |
-
| 0.1274 | 13.0 | 11388 | 0.6285 | 0.3617 |
|
| 79 |
-
| 0.1177 | 14.0 | 12264 | 0.5866 | 0.3857 |
|
| 80 |
-
| 0.1078 | 15.0 | 13140 | 0.5878 | 0.3810 |
|
| 81 |
-
| 0.0979 | 16.0 | 14016 | 0.6423 | 0.3764 |
|
| 82 |
-
| 0.0897 | 17.0 | 14892 | 0.6334 | 0.3850 |
|
| 83 |
-
| 0.0805 | 18.0 | 15768 | 0.7045 | 0.3904 |
|
| 84 |
-
| 0.0731 | 19.0 | 16644 | 0.7177 | 0.3941 |
|
| 85 |
-
| 0.0653 | 20.0 | 17520 | 0.7584 | 0.3826 |
|
| 86 |
-
| 0.0594 | 21.0 | 18396 | 0.8027 | 0.3805 |
|
| 87 |
-
| 0.0556 | 22.0 | 19272 | 0.8706 | 0.3736 |
|
| 88 |
-
| 0.0505 | 23.0 | 20148 | 0.8978 | 0.3979 |
|
| 89 |
-
| 0.0472 | 24.0 | 21024 | 0.9166 | 0.3968 |
|
| 90 |
-
| 0.0432 | 25.0 | 21900 | 0.9277 | 0.3978 |
|
| 91 |
-
| 0.0392 | 26.0 | 22776 | 0.9799 | 0.3973 |
|
| 92 |
-
| 0.0372 | 27.0 | 23652 | 1.0087 | 0.4115 |
|
| 93 |
-
| 0.0344 | 28.0 | 24528 | 1.0642 | 0.3996 |
|
| 94 |
-
| 0.032 | 29.0 | 25404 | 1.0861 | 0.4037 |
|
| 95 |
-
| 0.0307 | 30.0 | 26280 | 1.0877 | 0.4116 |
|
| 96 |
|
| 97 |
|
| 98 |
### Framework versions
|
|
|
|
| 21 |
metrics:
|
| 22 |
- name: F1
|
| 23 |
type: f1
|
| 24 |
+
value: 0.37298165525403665
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
This model is a fine-tuned version of [skt/kogpt2-base-v2](https://huggingface.co/skt/kogpt2-base-v2) on the klue dataset.
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 0.4076
|
| 35 |
+
- F1: 0.3730
|
| 36 |
|
| 37 |
## Model description
|
| 38 |
|
|
|
|
| 57 |
- seed: 42
|
| 58 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 59 |
- lr_scheduler_type: linear
|
| 60 |
+
- num_epochs: 3
|
| 61 |
|
| 62 |
### Training results
|
| 63 |
|
| 64 |
+
| Training Loss | Epoch | Step | Validation Loss | F1 |
|
| 65 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 66 |
+
| 0.6084 | 1.0 | 876 | 0.5353 | 0.2118 |
|
| 67 |
+
| 0.3911 | 2.0 | 1752 | 0.4691 | 0.3041 |
|
| 68 |
+
| 0.2855 | 3.0 | 2628 | 0.4076 | 0.3730 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|