Qi Zhu

For evaluating policy with user simulator conveniently, one should assign random seed for each simulated dialogue.

Since we assume the user will install this repo first, no need to use ``PYTHONPATH=../../../.. python xx.py``, just ``python xx.py`` will be ok.

[Feature] Benchmark Evaluation Script

1

**Describe the feature** Provide unified, fast evaluation script for Benchmark using unified datasets #11 and standard dataloaders #14 Evaluation metrics: - [x] NLU: Acc, dialog act tuple P/R/F1 - [x]...

[Feature] add datasets in unified format

4

**Describe the feature** Transform the following datasets into the [Unified format](https://github.com/ConvLab/ConvLab-3/blob/master/data/unified_datasets/README.md) under `data/unified_datasets`. Suggestions of adding/deleting datasets are welcome! - [x] MultiWOZ 2.1 - [ ] MultiWOZ 2.2 - [...