Getting errors when I run the example on the README

[dbuscombe-usgs]

Hi, on a fresh install, I tried to test the library by running the example code on the README

from pypots.data import load_specific_dataset
data = load_specific_dataset('physionet_2012')  # PyPOTS will automatically download and extract it.
X = data['X']

However, data does not contain X

KeyError: 'X'

I get errors when I load my own data and fit the saits model on it, so I'm looking for an example dataset and workflow, so I can troubleshoot. Thanks!

[github-actions]

Hi there 👋,

Thank you so much for your attention to PyPOTS! You can follow me on GitHub
to receive the latest news of PyPOTS. If you find PyPOTS helpful to your work, please star⭐️ this repository.
Your star is your recognition, which can help more people notice PyPOTS and grow PyPOTS community.
It matters and is definitely a kind of contribution to the community.

I have received your message and will respond ASAP. Thank you for your patience! 😃

Best,
Wenjie

[dbuscombe-usgs]

I got the model to run using the

X = data['train_X']

and skipping the data scaling. Since you already have a test split in the data dictionary, it would be nice to update the example in the README so it includes a test of the model on the test split.

You say:

Here I use the whole dataset as the training set because ground truth is not visible to the model, you can also split it into train/val/test sets

Like, how exactly do you give it a validation set to train with? Finally, what is the purpose of computing the model error on the train subset? Surely that tells you nothing about how well the model would perform during actual application?

Thanks

[dbuscombe-usgs]

I found a more complete and up to date example here https://github.com/WenjieDu/BrewPOTS/blob/main/PyPOTS_Imputation.ipynb, via the website. Perhaps just scrap the README example, and point new users to this tutorial? Thanks