what kind of model in pycox works for sequential patterns

[CFZhai]

I have sequential data where I aim to predict the survival rate or event rate for each timestep t within a sequence. Each timestep contains multiple features, making it a many-to-many sequential prediction task.

For instance, in my scenario, I am analyzing sequential online user learning logs. Each user performs daily learning tasks on day t. I track a user's learning progress from day 0 up to day t, with t varying for different users. My objective is to use a survival model to predict the likelihood of a user churning at each day t+1 using historical data day 0 to day t.
The input sequence length varies, which is different from the deephit and other models. Any one can help me ? thank you!