|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: bnc_spoken-aochildes-notm-log-rarity-seed |
|
|
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. --> |
|
|
|
|
|
# bnc_spoken-aochildes-notm-log-rarity-seed |
|
|
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 4.1809 |
|
|
|
|
|
## 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: 0.0005 |
|
|
- train_batch_size: 64 |
|
|
- eval_batch_size: 64 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: cosine |
|
|
- lr_scheduler_warmup_steps: 1000 |
|
|
- num_epochs: 6 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 6.3629 | 0.29 | 500 | 5.3438 | |
|
|
| 5.0612 | 0.59 | 1000 | 4.9309 | |
|
|
| 4.7299 | 0.88 | 1500 | 4.6994 | |
|
|
| 4.4646 | 1.18 | 2000 | 4.5689 | |
|
|
| 4.3241 | 1.47 | 2500 | 4.4556 | |
|
|
| 4.2231 | 1.77 | 3000 | 4.3658 | |
|
|
| 4.091 | 2.06 | 3500 | 4.2969 | |
|
|
| 3.9203 | 2.36 | 4000 | 4.2456 | |
|
|
| 3.8988 | 2.65 | 4500 | 4.2005 | |
|
|
| 3.8473 | 2.95 | 5000 | 4.1559 | |
|
|
| 3.655 | 3.24 | 5500 | 4.1528 | |
|
|
| 3.6161 | 3.54 | 6000 | 4.1208 | |
|
|
| 3.587 | 3.83 | 6500 | 4.0910 | |
|
|
| 3.472 | 4.12 | 7000 | 4.1000 | |
|
|
| 3.3442 | 4.42 | 7500 | 4.0986 | |
|
|
| 3.3331 | 4.71 | 8000 | 4.0864 | |
|
|
| 3.3218 | 5.01 | 8500 | 4.0869 | |
|
|
| 3.1604 | 5.3 | 9000 | 4.0981 | |
|
|
| 3.1589 | 5.6 | 9500 | 4.0985 | |
|
|
| 3.1531 | 5.89 | 10000 | 4.0983 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.1 |
|
|
- Pytorch 1.11.0+cu113 |
|
|
- Datasets 2.13.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
