Error in PyPOTS_Imputation.ipynb

[dbuscombe-usgs]

the following code returns an error

testing_mae = calc_mae(
    saits_imputation, 
    physionet2012_dataset['test_X_ori'], 
    physionet2012_dataset['test_X_indicating_mask'],
)

test_X_indicating_mask is not a key in the physionet2012_dataset dictionary. The following keys are available:

dict_keys(['n_classes', 'n_steps', 'n_features', 'scaler', 'train_X', 'train_y', 'train_ICUType', 'val_X', 'val_y', 'val_ICUType', 'test_X', 'test_y', 'test_ICUType', 'val_X_ori', 'test_X_ori'])

How might I create the variable test_X_indicating_mask. Thanks

[WenjieDu]

Hi there 👋,

Thank you so much for your attention to PyPOTS and BrewPOTS! You can follow me on GitHub to receive the latest news of PyPOTS. If you find PyPOTS and BrewPOTS helpful to your work, please star our repositories. 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 was able to create the mask by study other examples

indicating_mask = np.isnan(physionet2012_dataset['test_X']) ^ np.isnan(physionet2012_dataset['test_X'])  # indicating mask for imputation error calculation

# calculate mean absolute error on the ground truth (artificially-missing values)
testing_mae = calc_mae(
    saits_imputation, 
    physionet2012_dataset['test_X_ori'], 
    indicating_mask,
)
print(f"Testing mean absolute error: {testing_mae:.4f}")

However, now I get a new error

AssertionError: `targets` mustn't contain NaN values, but detected NaN in it

[dbuscombe-usgs]

I believe this is fixed using the np.nan_to_num function for the data with the NaNs. Thank you very much for these tutorials, I have been able to mimic this workflow for my own data