Is it a Bird?

This project trains a deep learning model to classify whether an image contains a bird or not.

Dataset

We download images of birds and forests from the web.
Each class contains 200 images, stored in separate directories:

• bird/
• forest/

Each image is automatically labeled based on its directory name.

We split the dataset as follows:

• 80% training
• 20% validation

Model

We use ResNet18, a convolutional neural network pretrained on ImageNet, and apply transfer learning.

learn = vision_learner(dls, resnet18, metrics=error_rate)
learn.fine_tune(3)