Entering edit mode
7 weeks ago
Hello, I'm Interested in making a ensemble machine learning model for survival analysis using Random forest, Gradient boosting and Cox proportional hazard model by averaging the survival curve to obtain the survival probability. But while modelling i encountered the issue where the cox model's output is different from the other two models. I want suggestion regarding how can i transform the hazard ratio output from a cox model into a survival function if possible. Any suggestions regarding alternative models and the exclusion of the cox model would also be appreciated. I'm new to this field please feel free to point out if there any mistakes in my approach, Thankyou.
Additional context: I used CoxPHFitter from lifelines to fit the model
