Juan Quiroz Aguilera

Having max_steps as a parameter to the fit method would also be useful with datasets with large number of variables, where convergence is extremely slow. I would also add precision...

Traditional ROC curves are not used in survival analysis. For time-dependent version, see https://scikit-survival.readthedocs.io/en/stable/user_guide/evaluating-survival-models.html#Time-dependent-Area-under-the-ROC

Just curious, any progress on this issue?

@yikuanli can you share your solution here?

I applied the fix to my private fork and the error remained (python 3.9.4, pandas 1.1.5). This indexing is what causes the issue: https://github.com/CamDavidsonPilon/lifelines/blob/aa019b18f28909f3658496d6f819f41077e6e060/lifelines/plotting.py#L514

I used the solution as in here https://github.com/CamDavidsonPilon/lifelines/commit/9b36d87410c2fea79f5780bb4d58e8c8153be8f2, changing the indexing and the rename (adding "columns=") to get it to work with pandas 1.1.5 (due to project restrictions). ```python if...

I'll play around with the code and see if there's an easy way to implement it.

The details are available in the paper: https://mental.jmir.org/2018/3/e10153/ Data was collected from a Samsung R720 Gear S2 Sport BT. The raw data is under folder raw_data. Coding of participants to...

@Lnzry good point! You can identify the features by tracing this function: https://github.com/juancq/emotion-recognition-smartwatch/blob/0c9c7c644f2e59da3908867fde1df2785123eead/extract_windows.py#L73-L110 The above also calls this function: https://github.com/juancq/emotion-recognition-smartwatch/blob/0c9c7c644f2e59da3908867fde1df2785123eead/extract_windows.py#L30-L70

Correction for multiple testing is not without its problems: https://www.jstor.org/stable/20065622. Ultimately, it should be up to the user discretion to set this variable.