Skip to content

Latest commit

 

History

History
20 lines (13 loc) · 831 Bytes

File metadata and controls

20 lines (13 loc) · 831 Bytes

Binary-Classification-using-CNN

Problem Question

Create A Binary classifier which categorizes input images as Cats or Dogs

Dataset

Dataset can be downloaded from this link: Cat-Dog-Dat

The dataset provided consists of training data and testing data for both the categories.

Solution

  • This code realizes a CNN for binary classification using tensorflow backened keras.
  • The accuracy obtained was around 82%, and it was the only metric score considered.
  • The algorithm was trained on well classified and labelled image data consisting of 10,000 images.

PS- Change the directory used in the code before running

To use pictures of your own, input the new directory in the code