|
--- |
|
base_model: Karzan/ckb-t5-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: walamakan-t5-base |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# walamakan-t5-base |
|
|
|
This model is a fine-tuned version of [Karzan/ckb-t5-base](https://huggingface.co/Karzan/ckb-t5-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.8637 |
|
- Bleu: 0.0 |
|
- Gen Len: 19.0 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:----:|:-------:| |
|
| No log | 1.0 | 87 | 4.3273 | 0.0 | 19.0 | |
|
| No log | 1.99 | 174 | 4.1466 | 0.0 | 18.9807 | |
|
| No log | 2.99 | 261 | 4.0477 | 0.0 | 18.955 | |
|
| No log | 4.0 | 349 | 3.9853 | 0.0 | 19.0 | |
|
| No log | 5.0 | 436 | 3.9426 | 0.0 | 19.0 | |
|
| 4.321 | 5.99 | 523 | 3.9122 | 0.0 | 19.0 | |
|
| 4.321 | 6.99 | 610 | 3.8894 | 0.0 | 19.0 | |
|
| 4.321 | 8.0 | 698 | 3.8749 | 0.0 | 19.0 | |
|
| 4.321 | 9.0 | 785 | 3.8672 | 0.0 | 19.0 | |
|
| 4.321 | 9.97 | 870 | 3.8637 | 0.0 | 19.0 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.4 |
|
- Tokenizers 0.13.3 |
|
|
