ORF: use two-dimensional target feature

Hi!

Thanks for bringing such an amazing package!

I have a question about target features in ORF. If I wanna estimate heterogeneity along two features, X1 and X2, how should I set the test points? If both X1 and X2 are continuous variables ranging from 0 to 1, I suppose I could set the test points to:
X_test = np.array([[0, 0.2, 0.4, 0.6, 0.8, 1], [0, 0.2, 0.4, 0.6, 0.8, 1]]) (the step is 0.2 in this example)

But if X1 is a continuous variable ranging from 0 to 1 and X2 is a binary variable that takes values 0 and 1, then how should I set the test points? The following specification seems to be wrong because X_test should be in shape (n, d_x).
X_test = np.array([[0, 0.2, 0.4, 0.6, 0.8, 1], [0, 1]])

Then how should I specify the test points in the latter case? Thanks in advance!!

Best,

Start by locating the ORF target-feature prediction API and its documentation or examples, then inspect how X_test shape and mixed continuous/binary features are handled. Document the accepted layout with examples for both cases in the issue, and consider a focused test or example showing the expected two-dimensional input. Done means a newcomer can construct valid test points without further clarification.