KoT5-test-add-data-from5ep-continue
This model is a fine-tuned version of hyorea1/KoT5-test-add-data-from5ep-continue on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1881
- Rouge1: 11.7784
- Rouge2: 2.959
- Rougel: 11.6648
- Rougelsum: 11.6892
- Gen Len: 34.7301
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 100
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Gen Len | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
---|---|---|---|---|---|---|---|---|
1.4317 | 0.32 | 800 | 34.7618 | 1.2414 | 11.8765 | 3.2439 | 11.7982 | 11.8203 |
0.9488 | 0.64 | 1600 | 35.1324 | 1.2255 | 11.5076 | 3.0739 | 11.394 | 11.4492 |
1.1868 | 0.97 | 2400 | 34.2368 | 1.1983 | 10.7675 | 2.8679 | 10.7567 | 10.7806 |
1.3349 | 1.29 | 3200 | 34.3772 | 1.2170 | 11.0853 | 2.8116 | 10.9947 | 11.0642 |
1.3918 | 1.61 | 4000 | 34.7368 | 1.1845 | 11.6434 | 2.9694 | 11.5189 | 11.5525 |
1.6205 | 1.93 | 4800 | 33.9897 | 1.1801 | 11.1446 | 2.9624 | 11.0259 | 11.0535 |
1.1958 | 2.25 | 5600 | 34.6926 | 1.1845 | 11.3408 | 2.9759 | 11.2451 | 11.2685 |
1.2391 | 2.58 | 6400 | 34.8382 | 1.1879 | 11.227 | 2.832 | 11.0999 | 11.12 |
1.458 | 2.9 | 7200 | 34.8904 | 1.1869 | 11.4615 | 2.832 | 11.3029 | 11.3413 |
1.0598 | 3.22 | 8000 | 1.1877 | 11.2705 | 2.8787 | 11.1582 | 11.2173 | 34.8993 |
1.3546 | 3.54 | 8800 | 1.1832 | 11.9647 | 2.9161 | 11.848 | 11.8769 | 34.5897 |
1.5696 | 3.86 | 9600 | 1.1859 | 11.352 | 2.8466 | 11.2177 | 11.2336 | 34.6441 |
1.3378 | 4.19 | 10400 | 1.1873 | 11.9282 | 2.959 | 11.8205 | 11.8427 | 34.7125 |
1.063 | 4.51 | 11200 | 1.1877 | 11.8063 | 2.9284 | 11.6855 | 11.7112 | 34.6426 |
1.184 | 4.83 | 12000 | 1.1881 | 11.7784 | 2.959 | 11.6648 | 11.6892 | 34.7301 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.7.1
- Tokenizers 0.13.2
- Downloads last month
- 13
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.