|
--- |
|
license: apache-2.0 |
|
base_model: google/flan-t5-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: flan-t5-base-eng-hwp |
|
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. --> |
|
|
|
# flan-t5-base-eng-hwp |
|
|
|
This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.5058 |
|
- Bleu: 4.9532 |
|
- Gen Len: 18.8709 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 12 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:| |
|
| No log | 1.0 | 420 | 1.6158 | 3.6321 | 18.892 | |
|
| 2.1318 | 2.0 | 840 | 1.4711 | 4.19 | 18.8743 | |
|
| 1.5146 | 3.0 | 1260 | 1.4193 | 4.3921 | 18.8608 | |
|
| 1.2936 | 4.0 | 1680 | 1.3936 | 4.5268 | 18.8363 | |
|
| 1.1403 | 5.0 | 2100 | 1.4030 | 4.6813 | 18.8608 | |
|
| 1.0284 | 6.0 | 2520 | 1.4078 | 4.8234 | 18.8684 | |
|
| 1.0284 | 7.0 | 2940 | 1.4192 | 4.8126 | 18.8709 | |
|
| 0.9196 | 8.0 | 3360 | 1.4303 | 4.8599 | 18.87 | |
|
| 0.8459 | 9.0 | 3780 | 1.4506 | 4.7802 | 18.8599 | |
|
| 0.7884 | 10.0 | 4200 | 1.4757 | 4.8895 | 18.8785 | |
|
| 0.739 | 11.0 | 4620 | 1.4945 | 4.9126 | 18.8759 | |
|
| 0.7083 | 12.0 | 5040 | 1.5058 | 4.9532 | 18.8709 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
