dodiscover
dodiscover copied to clipboard
py-why
[Experimental] Global causal discovery algorithms
LGTM. My only issues are with some of the type hints. @adam2392 can you comment here on why some flavors are left out? For example: - mutual information and asymptotic...
## Problem The classifier-based CI tests are nice because they are non-parametric generally, but slow because for each test, you need to re-fit a classifier. For modular components, we should...
LGTM. One comment. The CMI value that is calculated as follows: `val = hxyz - (hxz + hyz - hz).mean()` Doesn't this have an asymptotic chi-squared distribution under the null...
See: https://www.mdpi.com/1099-4300/24/9/1234 We have existing code for CMI-knn and CMI-classifier based approaches, which can probably be used.
**Is your feature request related to a problem? Please describe.** A clear and concise description of what the problem is. Ex. I'm always frustrated when [...] **Describe the solution you'd...
**Is your feature request related to a problem? Please describe.** The type and domain of the variables in the data should be a first class citizen **Describe the solution you'd...
**Is your feature request related to a problem? Please describe.** Add interventions to causica wrapper. Causica already handles interventions data. Depends on us solving interventions first.
**Is your feature request related to a problem? Please describe.** Add causal constraints to Causica.
**Is your feature request related to a problem? Please describe.** We need an initial score-based algorithm. **Describe the solution you'd like** Use causal-learn as a dependency to build GES **Describe...
**Is your feature request related to a problem? Please describe.** Implement GIN as a native algorithm. **Describe the solution you'd like** We have a[ wrapper around causal-learn for the GIN...