Skip to content

A tool allowing you to perform image classification on a relatively small dataset by using the Inception V3 pre-trained model.

Notifications You must be signed in to change notification settings

anishchopra/Simple-Image-Classification

Repository files navigation

Simple-Image-Classification

A tool allowing you to perform image classification on a relatively small dataset by using the Inception V3 pre-trained model.

In the example provided, only 42 images were used for each category, including the training and validation images! This is the advantage of using a pre-trained model.

Requirements

  • NumPy
  • Scikit-Learn
  • Keras
  • TensorFlow

Usage

Use generate_inception_features.py to generate the features computed using the Inception V3 pre-trained model.

Use train_nn.py to train a one-layer neural network classifier, which takes in Inception V3 features as input.

Use run_classifier.py to run your newly-trained classifier on a new image.

See demo.py for example usage and further instructions.

About

A tool allowing you to perform image classification on a relatively small dataset by using the Inception V3 pre-trained model.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages