bruno messias

Results 76 comments of bruno messias

Maybe this can help you @GideonBear , https://github.com/pyastrx/pyastrx/tree/main/pyastrx/inference . Also you use this after installing pyastrx ``` mypyq -f test.py ```

"test_tree.py" is running without errors. But it seems more appropriate as an example that can be added following the standard of examples here: https://github.com/fury-gl/fury/tree/master/docs/examples.

Yes, like this using mypy ```python # test.py def test(a: int, b: float)->int: print(a) return a+b ``` ``` mypyq -f test.py ``` ![image](https://user-images.githubusercontent.com/6979335/196307891-c4695dbd-f399-4d7d-86c4-88ecd956e4b1.png)

Note: Pyre is faster than mypy. However, thre is no cli interface for that in pyastrx. https://github.com/pyastrx/pyastrx/blob/fc7c1017b7f8671a5a9445054c772f39a1ab21fa/pyastrx/inference/pyre.py#L10

Hi @gshank, Could I please receive feedback on this PR? One aspect I'm unable to understand is why the codecov is reducing the percentage of coverage. This PR did not...

> I think the original "fix" (setting the materialization in the config_call_dict) was wrong, but I think this approach just doubles down on that wrong approach. This is a tricky...

A big issue is that removing the default table behavior could break numerous pipelines, as views are not supported in Python models.

One option I am considering, given that @aranke only wanted to change the default materialization for Python (so I assume there is no point in considering future default changes), and...

I've updated the PR with the proposed solution @gshank. Could you review the changes to ensure we're on the same page?

Ah, I forgot to mention one thing. The default is set before the update_node_config. As a result, the following test created by @aranke , which was passing in my previous...