qns: Saving each experiment run
Closed this issue · 2 comments
Is there something that I can serialize and deserialize for aiding the next learning experiment run?
It looks like I can serialize and deserialize policy
Hey @alanpoon, at the moment all of this kind of functionality must be done manually. Also (and apologies for this) not every type in the framework currently implements Serialize and Deserialize from the serde crate; the features are a bit of a misnomer at the moment.
In the meantime, I would suggest cloning this locally, implementing these traits and then you can use, say, bincode to actually save these objects.
Sorry for that, I agree that having these built in would be great. My plan is to build a separate crate that implements all the automation logic, something akin to ray.