yinruiqing

In fact, this project has been merged into [pyannote.audio](https://github.com/pyannote/pyannote-audio). You can find more information [here](https://github.com/pyannote/pyannote-audio/tree/master/tutorials/change-detection). PS: We move into python3 in the latest version.

@jun-danieloh `CondaEnvException: Unable to determine environment` I think you should first check you cuda version and cudatoolkit version.

You can find how pl set distributed sampler [here](https://github.com/PyTorchLightning/pytorch-lightning/blob/master/pytorch_lightning/trainer/data_loading.py#L274). If you dive into the source code of DistributedSampler in [pytorch](https://github.com/pytorch/pytorch/blob/master/torch/utils/data/distributed.py#L95), you can find the random seed depends on the epoch....