Model save

(Jamali) #1

Is there any way to save the trained model and then load it back for inference?:slightly_smiling_face:

0 Likes

(Vova Manannikov) #2

As far as I know, there’s no “official” way of doing this in S4TF, yet (correct me if I’m wrong, @bradlarson).
Here’s couple examples of loading model weights:


Saving would be opposite direction, using _Raw.saveV2().
So it boils down to using _Raw.saveV2() on each param tensor to save it and _Raw.restoreV2() to load back.

2 Likes