dswah / sgcrfpy

SGCRFpy: Sparse Gaussian Conditional Random Fields in Python
MIT License
29 stars 5 forks source link

fit should allow fitting from current estimate #10

Closed dswah closed 8 years ago

dswah commented 8 years ago

this way if we see that our loss hasnt gone down enough, we can continue fitting.

could be as simple as checking if Lamand Theta are None before fitting. if so, initialize with eye and zero matrices, otherwise, use current ones.