A Machine learning (Deep Learning) project on Cat v/s Dog image classification
using CNN
, VGG-16
and VGG-19
in Python.
Link to the dataset
: https://download.microsoft.com/download/3/E/1/3E1C3F21-ECDB-4869-8368-6DEBA77B919F/kagglecatsanddogs_5340.zip- The Dataset consists of 12500 images of Cat and 12500 images of Dog.
- It is advised to run the notebook in
Google colab
, as colab providesGPU
which lessens the training time of the algorithm. - Also colab is inbuilt with all of the required python pip packages, thereby saving your time to install these packages on your local system.
- To run the notebook in your native pc, just change the paths of image folders accordingly.
- To use the
TestImages
, upload the test images folders to your drive and give the path accordingly. - A brief description of all the 3 algorithms namely
CNN
,VGG-16
andVGG-19
are available in the presentations uploaded. (Both part 1 and part 2). Explanation on methods to approach the problem, result analysis of all the 3 algorithms are provided in these presentations as well. - All the codes are documented for deeper understanding.