|
|
--- |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: bert-dp-4 |
|
|
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. --> |
|
|
|
|
|
# bert-dp-4 |
|
|
|
|
|
This model is a fine-tuned version of [](https://huggingface.co/) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 6.7082 |
|
|
|
|
|
## 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: 40 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 6.3445 | 1.89 | 1000 | 5.9292 | |
|
|
| 5.8327 | 3.78 | 2000 | 5.8495 | |
|
|
| 6.5089 | 5.67 | 3000 | 6.7228 | |
|
|
| 6.7256 | 7.56 | 4000 | 6.7197 | |
|
|
| 6.7194 | 9.45 | 5000 | 6.7135 | |
|
|
| 6.7174 | 11.34 | 6000 | 6.7132 | |
|
|
| 6.7114 | 13.23 | 7000 | 6.7119 | |
|
|
| 6.7166 | 15.12 | 8000 | 6.7129 | |
|
|
| 6.7109 | 17.01 | 9000 | 6.7107 | |
|
|
| 6.7112 | 18.9 | 10000 | 6.7134 | |
|
|
| 6.7125 | 20.79 | 11000 | 6.7090 | |
|
|
| 6.7099 | 22.68 | 12000 | 6.7085 | |
|
|
| 6.7084 | 24.57 | 13000 | 6.7069 | |
|
|
| 6.7066 | 26.47 | 14000 | 6.7063 | |
|
|
| 6.7083 | 28.36 | 15000 | 6.7037 | |
|
|
| 6.7062 | 30.25 | 16000 | 6.7044 | |
|
|
| 6.705 | 32.14 | 17000 | 6.7022 | |
|
|
| 6.7041 | 34.03 | 18000 | 6.7058 | |
|
|
| 6.7031 | 35.92 | 19000 | 6.7055 | |
|
|
| 6.7031 | 37.81 | 20000 | 6.7067 | |
|
|
| 6.7039 | 39.7 | 21000 | 6.7082 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.1 |
|
|
- Pytorch 1.11.0+cu113 |
|
|
- Datasets 2.13.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
