New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Develop Tooling Around Gorgonia #81

Open
chewxy opened this Issue Feb 7, 2017 · 2 comments

Comments

Projects
None yet
2 participants
@chewxy
Copy link
Member

chewxy commented Feb 7, 2017

Here are some tooling ideas:

  • A program that detects x := tensor.New(tensor.WithShape(1,2,3)) as something that will definitely fail at runtime (because no Dtype or backing was passed in)
  • Collorary to the above - detect that this is an ineffectual setting: x := tensor.New(tensor.Of(dt), tensor.WithBacking(a)) - the main reason is the backing type will override the tensor.Of option.
  • Better graph output - maybe interactive as graphviz doesn't seem to handle big graphs

@chewxy chewxy added the ideation label Feb 7, 2017

@kabaka0

This comment has been minimized.

Copy link
Collaborator

kabaka0 commented Feb 8, 2017

Perhaps it is worth sticking with the spirit of Go and trying to make sure that code is written to prevent failures at runtime, even without additional tools. I'd certainly expect that of the core functions and types. Maybe by combining default rules (clearly listed in the documentation) and a logging option (whenever defaults override or fill in for user selections).

@chewxy

This comment has been minimized.

Copy link
Member

chewxy commented Feb 8, 2017

I think for the most part that's true - most functions and method return an error, leaving it to the user to handle. The only point of contention I have is the New functions - by convention the New functions don't return an error. Rather they panic. Hence the idea.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment