This project considers the 200 Bird Species image dataset in order to develop a classification model using Convolutional Neural Network (CNN) and Transfer Learning with PyTorch.
The dataset is from Kaggle under the name 400 Bird Species computed on Jetson Nano. As the name indicates, the dataset contains images of 400 species of birds from different regions around the world. The images are in JPEG format with size 3x224x224. The dataset has 27,000+ images for training, 1,000 images for validation and1,000 images for testing.
The objective of the project is to train a model with CNN and transfer learning in order to get a good accuracy, a minimum loss conducting to a precise prediction of the different bird species images.
The training dataset is used to train the model, the validation dataset to evaluate the model while training and the test dataset to test the model using external data
made by- Ayush Mishra, Zayed Syed Haque and Gaurav Saha