feat: Add Expected extensions output mode to Magika CLI #78
Conversation
nhonx
changed the title
nhonx
changed the title
|
Thank you for the PR! I'll need to give some thoughts about it, we have received a similar request and need to think how to integrate it for the long term. Leaving this open for now for visibility; will follow up later on. Thanks! |
|
@reyammer: I think we should think about this as a long-term feature. And we also need to update more correct/expected extensions to https://github.com/google/magika/blob/main/python/magika/config/content_types_config.json |
|
yes, indeed. We are already working on v2, and adding many more extensions / types. your examples are very spot on. And sorry for not following up on this, didn't have time to think about this and we are currently pushing for starting new training round. Will get to this once I have a moment, and I agree we should have a feature in this direction! |
|
Hi @reyammer , should I close this? As I see we have new code adding to |
- Uses the latest release candidate of the ORT library, which provides async inference. - The CLI creates only one session and shares it between the different inference tasks. - Implements the features extraction reference tests. - Implements the %e format for #78.
Add
-e / --expected-extsmode to Magika CLI, which output one or several expected file extensions of input file, in case the input extension is missed or incorrect