update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-sa-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: gpt_16_5_3e-5_lp5_nb5
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# gpt_16_5_3e-5_lp5_nb5
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [skt/kogpt2-base-v2](https://huggingface.co/skt/kogpt2-base-v2) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 3.9078
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 3e-05
|
| 37 |
+
- train_batch_size: 16
|
| 38 |
+
- eval_batch_size: 16
|
| 39 |
+
- seed: 42
|
| 40 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
+
- lr_scheduler_type: linear
|
| 42 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 43 |
+
- num_epochs: 5.0
|
| 44 |
+
|
| 45 |
+
### Training results
|
| 46 |
+
|
| 47 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 49 |
+
| 3.3245 | 0.38 | 1000 | 4.0176 |
|
| 50 |
+
| 3.1222 | 0.76 | 2000 | 3.9845 |
|
| 51 |
+
| 2.9992 | 1.13 | 3000 | 3.9635 |
|
| 52 |
+
| 2.8843 | 1.51 | 4000 | 3.9377 |
|
| 53 |
+
| 2.882 | 1.89 | 5000 | 3.9268 |
|
| 54 |
+
| 2.7411 | 2.27 | 6000 | 3.9208 |
|
| 55 |
+
| 2.7204 | 2.64 | 7000 | 3.9160 |
|
| 56 |
+
| 2.7106 | 3.02 | 8000 | 3.9171 |
|
| 57 |
+
| 2.5857 | 3.4 | 9000 | 3.9162 |
|
| 58 |
+
| 2.5863 | 3.78 | 10000 | 3.9037 |
|
| 59 |
+
| 2.5674 | 4.15 | 11000 | 3.9135 |
|
| 60 |
+
| 2.4901 | 4.53 | 12000 | 3.9125 |
|
| 61 |
+
| 2.505 | 4.91 | 13000 | 3.9078 |
|
| 62 |
+
|
| 63 |
+
|
| 64 |
+
### Framework versions
|
| 65 |
+
|
| 66 |
+
- Transformers 4.25.1
|
| 67 |
+
- Pytorch 1.9.0+cu102
|
| 68 |
+
- Datasets 2.8.0
|
| 69 |
+
- Tokenizers 0.13.2
|