imgbeddings VS awesome-teachable-machine

Seconding the recommendation of CLIP embeddings, especially compared to image histograms + requiring OpenCV.

I wrote a naive, minimal dependency Python package to calculate image embeddings (https://github.com/minimaxir/imgbeddings) with some lookup demo notebooks and it works well in a pinch, although it's due for an upgrade.

Normally, this is Hacker News reductiveness, but yes, image classification via CLIP is that easy, especially with Hugging Face's API for it: https://huggingface.co/docs/transformers/model_doc/clip

I created a Python package to generate image embeddings from CLIP's vision model (without requiring a ML framework), and a simple linear classifier on those embeddings does the trick: https://github.com/minimaxir/imgbeddings

Not sure if I've seen anything of the sort, seems rather specific. Maybe try a Teachable Machine project? https://teachablemachine.withgoogle.com/

Train a computer to recognize your images, sounds, and poses. Use this resource to gain a better understanding.

[Teachable Machine](https://teachablemachine.withgoogle.com/)

We will create an machine learning model that can classify Arabic and English books. To collect, train, and test data, we will use Teachable Machine from Google.

a lot of places! but for a high schooler, better to focus at what you want to do first. or if you still haven't gotten any idea, try a simple explanation on what ml is without the math on youtube and tinker around a no code machine learning platform like https://teachablemachine.withgoogle.com/

The principle is roughly the same as it is possible with teachablemachine.withgoogle.com.

You mentioned app, so check this out: https://teachablemachine.withgoogle.com

Here is a similar tool and source