Skip to content

Conversation

@cc7768
Copy link
Member

@cc7768 cc7768 commented Mar 7, 2017

Add derivatives for complete polynomials up to order 5.

This is mostly done. I've added some tests for the derivative functions, but found one last bug that needs to be addressed. In particular the biggest one is the way that I'm computing the derivatives doesn't work if any of the elements of the point are 0.0 because they sometimes get raised to the -1. This should only take 5-10 minutes to fix and will do it in next few days.

Also note that I've commented out the cache=True argument for JIT compiling for the time being. See QuantEcon/QuantEcon.py#285 for more details.

@sglyon @albop

@sglyon
Copy link
Member

sglyon commented Mar 7, 2017

Hey @albop do you think this is a good candidate for more of your temptia + numba @generated magic?

@albop
Copy link
Member

albop commented Mar 8, 2017

So does the cache=True thing explain why the tests don't pass ?
I'd like to look a bit more in detail to see how hard it would be to implement complete products of arbitrary orthogonal polynomial bases. This would go into linear_basis.
In the meantime, this could be merged in master. Tell me when you think it's ready.

@sglyon
Copy link
Member

sglyon commented Mar 8, 2017

@albop if you do figure that out let me know. Is like to see an efficient implementation of that!

@cc7768
Copy link
Member Author

cc7768 commented Mar 10, 2017

Hi @albop. Let me know if you have any comments otherwise, I think this is ready to merge.

@sglyon sglyon merged commit 736fef7 into master Mar 15, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants