# TripPy mindspore

**Repository Path**: lifancong/trip-py-mindspore

## Basic Information

- **Project Name**: TripPy mindspore
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2021-11-11
- **Last Updated**: 2021-11-11

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# TripPy mindspore 复现
# TripPy based on mindspore
原论文地址: https://arxiv.org/pdf/2005.02877.pdf

Source paper: https://arxiv.org/pdf/2005.02877.pdf

How to run (运行指南):
``` bash
# install pytorch, transformers and mindspore
pip install pytorch
pip install transformers
......

# training the TripPy model
python run_dst.py

# evaluate the checkpoint
python eval.py --pretrained=${your checkpoint}


# compute the joint goal acc
python metric_bert_dst.py

```

-----
The caches are created by the TripPy original code, we only provide MultiWOZ2.1 caches.
ms_bert_base.ckpt is the pretraining model of mindspore bert, you could runing the following command to convert pytoch checkpoint to mindspore checkpoint.
``` bash
python convert_params.py
```
if you want to train on GPU, just change context.set_context(mode=context.GRAPH_MODE, device_target='Ascend') in the run_dst.py to 'GPU'.