Julian Klug

Why is signaly of the calibration marker used as scale value?

2

Thank you for this great work! I was wondering why the y component of the signal from the calibration marker is used to scale both X and Y component of...

When computing predictions with the Cox PH model it would be great to also get standard errors (similar to: https://stat.ethz.ch/R-manual/R-devel/library/survival/html/predict.coxph.html) Is there a workaround to get this currently?

[Documentation] List Torch version requirement (Not working with pytorch==1.12)

1

**Location of the documentation** Main dependencies could be listed as requirements. Further, torch version could be stated in setup.cfg https://github.com/vanderschaarlab/temporai/blob/717e9cabb40b29d8d4b370924c1cc8a83406db07/setup.cfg#L83C5-L83C10 **Documentation problem** By testing two pytorch versions in my setup...

[Documentation] Train/Test split in tutorials

1

**Location of the documentation** Tutorial example: https://temporai.readthedocs.io/en/latest/user_guide/usage/tutorial06_treatments.html#Using-a-temporal-treatment-effects-plugin. tempor.data.dataset module: https://temporai.readthedocs.io/en/latest/api/tempor.data.dataset.html#tempor.data.dataset.BaseDataset.split **Documentation problem** Training and inference is done on the same data in all tutorials. **Suggested fix for documentation** Encourage train...

[Feature Request] Treatment effect estimation with temporal treatment, but single timepoint outcome

1

**Describe your proposed feature** Estimation of individual treatment effect with an intervention carried out over time, and outcome measured at a single time-point is currently not possible with the TemporAI...

Given a sequence, how should nan values be handled for multiscale entropy? Current behavior will raise an unhelpful error as _r_ will be undefined: `AssertionError: r: must be a positive...