🚀 feat(inspect): Allow user to upload images, train model and display training progress

[dwesolow]

📝 Description

This PR introduces a few features:

• allow user to upload images (temporary feature, to be removed when we support stream capture);
• list uploaded images;
• allow user to train a model when there is enough number of normal images;
• display proper message while training.

This is just starting point, we will improve that once we know that training is stable.

✨ Changes

Select what type of change your PR is:

• 🚀 New feature (non-breaking change which adds functionality)
• 🐞 Bug fix (non-breaking change which fixes an issue)
• 🔄 Refactor (non-breaking change which refactors the code base)
• ⚡ Performance improvements
• 🎨 Style changes (code style/formatting)
• 🧪 Tests (adding/modifying tests)
• 📚 Documentation update
• 📦 Build system changes
• 🚧 CI/CD configuration
• 🔧 Chore (general maintenance)
• 🔒 Security update
• 💥 Breaking change (fix or feature that would cause existing functionality to not work as expected)

✅ Checklist

Before you submit your pull request, please make sure you have completed the following steps:

• 📚 I have made the necessary updates to the documentation (if applicable).
• 🧪 I have written tests that support my changes and prove that my fix is effective or my feature works (if applicable).
• 🏷️ My PR title follows conventional commit format.

For more information about code review checklists, see the Code Review Checklist.

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[MarkRedeman]

Looks good, thanks @dwesolow !
I have some minor comments but feel free to handle them in follow up PRs

[MarkRedeman]

np: I think this is a bit more readable, (it makes it easier to see that we render the media items by default, and potentially add some undefined placeholders)

Suggested change

	const mediaItemsToRender =
	mediaItems.length >= REQUIRED_NUMBER_OF_NORMAL_IMAGES_TO_TRIGGER_TRAINING
	? mediaItems
	: Array.from({ length: REQUIRED_NUMBER_OF_NORMAL_IMAGES_TO_TRIGGER_TRAINING }).map((_, index) =>
	index <= mediaItems.length - 1 ? mediaItems[index] : undefined
	);
	const mediaItemsToRender = [
	...mediaItems,
	...Array.from({ length: Math.max(0, REQUIRED_NUMBER_OF_NORMAL_IMAGES_TO_TRIGGER_TRAINING - mediaItems.length) })
	]

[dwesolow]

Nice, will add it, thanks!

[MarkRedeman]

@ashwinvaidya17 can you comment on what other models should be available? The id that is used here will be used in the model training service.

[MarkRedeman]

I'd maybe also add a if (jobProgres.status === 'cancelled') here instead of relying that type inference guarantees that the status must be cancelled.
Having the explicit if statement here makes it a bit easier to read when reviewing outside of an IDE (I don't have to make sure that in the if statements above we handled all except the cancelled status)

[MarkRedeman]

Can we use the '/api/jobs' endpoint for this? Perhaps we can make it so that if the jobs endpoint returns 1 or more jobs for this project, then we render all of those jobs' progresses.

[dwesolow]

Yeah I was wondering implementing this but I wasn't sure but if u also agree then I'll implement that ;).

[MarkRedeman]

Let's go with this on the short term.
Long term we have to see how long training takes and how frequently we expect new models to be trained (and if they are trained in bulk), if they are then I think a job management panel like we have in Geti would be a good approach.

[MarkRedeman]

It might be better to change this function to an async function and await each mutation. Otherwise if the user selects 1000 images then we would immediately try to upload all of them.

Though given that we will likely remove this feature once the source pipeline is integrated I guess it won't be worth it to optimize this?

[dwesolow]

Yeah, I wanted to keep it really simple since it's a temporary feature.