Differences of args in ImputationDataset between tsl-0.1.1 and the latest

[mrluin]

Hello,

Thanks for your great contribution in neural spatiotemporal data processing community!

I found that some works used the old version of tsl package, i.e. v-0.1.1, and there are some difference on args compared with the latest version. For example, in tsl-0.1.1 there are training_mask and eval_mask args in class ImputationDataset, but in the latest version training_mask is discarded, so could you please tell me is there something changes or some mechanism allow us can discard training_mask?

Looking forward to your reply, thanks!

[marshka]

Hi,

thanks for your interest in our work! we renamed training_mask as input_mask to clarify that this mask is associated with the input inside the model. Besides that, you can also control how many additional points are whitened at random during training with the parameter whiten_prob in the Imputer.

Actually, we should also rename training_mask in the MissingValues datasets and improve the documentation on this part, thank you for pointing this out! Feel also free to make pull requests should you have improvements to propose.

Best,

Ivan