Skip to content

Linting errors connected to numpy 2.2 #756

@jonbrenas

Description

@jonbrenas

A recent PR (#752) raised some linting errors:

Image

I think it might be due to changes in numpy. It looks like pinning numpy to 2.1 solves the issue.

The changes seem to be features and not bugs: e.g., np.random.randint() is being "legacy"ed and replaced by np.random.default_rng().integers().

@alimanfoo, @cclarkson, @ahernank, @leehart, do you think the best solution is to pin the version back (as we have done recently with bokeh, I think) or to update the code to be compliant with the way numpy is going? I think updating the code would be better but more time-consuming.

Metadata

Metadata

Labels

bugSomething isn't working

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions