How do you ACTUALLY see the image classification result of your model?

so i made a working classification model on co-laboratory with a given dataset.

I want to input an image of my choice(e.g, a pic i downloaded from google per se) and see what result the model spit out.

I looked it up how to do this, but i couldn't find one.

has anyone done this? if it hard to explain, could you tell me a link of video/post explaining how to do this?

1 answer

  • answered 2021-06-23 15:08 Mushfirat Mohaimin

    1. Navigate to the file explorer in colab and upload your file from device.
    2. Copy the image path and open it using PIL.Image.open or matplotlib.pyplot.imread or whatever you wish, then convert it to a numpy array.
    3. Then resize the image array and reshape it according to your model.
    4. After that, you can use model.predict() to get a classification of your image.