Jeremy Jordan

cannot reuse deleted experiment names

13

Thank you for submitting an issue. Please refer to our [issue policy](https://www.github.com/mlflow/mlflow/blob/master/ISSUE_POLICY.md) for information on what types of issues we address. For help with debugging your code, please refer to...

fiddler integration

1

I'm leaving this PR up as an example so that you can get an idea of what changes might be required to integrate with a monitoring tool like [Fiddler](https://www.fiddler.ai/).

Moves `TorchRepresentationDataset` and `TorchMusicFactoryDataset` to be defined outside of `to_pytorch_dataset` so that `pickle` can import the class definitions in a new process. Fixes #74

PyTorch's [Dataloader](https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader) has an argument for `num_workers` which can fetch items from your dataset in parallel using multiprocessing, but this requires your dataset to be able to be pickled so...