Continuing the discussion from Lesson 2 official topic:
I just finished Lesson Two and decided to get my hands a lil dirty deploying an AIr Jordan classifier. https://huggingface.co/spaces/ferdizz47/aj_classifier
Love the idea! I tried it with a couple of images but it gave me an error:
I tried your code and .pkl file in Google Colab and it works fine so I think the error is somewhere in your gradio code:
Thanks for trying out the classifier! I always find it difficult identifying other air Jordan models like AJ4, AJ11, AJ13, AJ5 LOL. I’ve noticed that it currently only works with certain image file formats like JPEG and PNG, which are among the most common types. I’m considering tweaking it a bit to automatically convert images to JPEG using the Python Imaging Library (PIL) for broader compatibility.
Nice project. Can you add some examples as well to the demo? So users can easily try out.
I just modified and added some examples. Thank you
Keep it up! Where did you get the training data? duckduckgo or some API?
I used duckduckgo