yvkrishna / ai-webapp Goto Github PK

Fix the code of image classifer

Create nodejs routs for the improvement of the backend. Also need to connect the tensorflow models to the backend

Have to find out any way such that the same image classification can be used for object detection but need to add boxing to it.

Need to complete the front end (ui) using bootstrap and make it completly responsive

image classification has only few images and with that we were able to achieve upto 68% for training and 48% for validating accuracy...

So try to change the path in image_classifer.py and add as many as pictures as the dataset as possible