|
--- |
|
language: |
|
- en |
|
license: apache-2.0 |
|
tags: |
|
- dialogue policy |
|
- task-oriented dialog |
|
datasets: |
|
- ConvLab/sgd |
|
|
|
--- |
|
|
|
# ddpt-policy-sgd_0.01multiwoz21 |
|
|
|
This is a DDPT model (https://aclanthology.org/2022.coling-1.21/) trained on [Schema-Guided Dialog](https://huggingface.co/datasets/ConvLab/sgd) and afterwards on 1 percent of [MultiWOZ 2.1](https://huggingface.co/datasets/ConvLab/multiwoz21) |
|
|
|
Refer to [ConvLab-3](https://github.com/ConvLab/ConvLab-3) for model description and usage. |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 1e-05 |
|
- train_batch_size: 64 |
|
- seed: 0 |
|
- optimizer: Adam |
|
- num_epochs: 40 |
|
- use checkpoint which performed best on validation set |
|
### Framework versions |
|
- Transformers 4.18.0 |
|
- Pytorch 1.10.2+cu111 |
|
|
