A simple implementation for image classifier web application with Keras and Flask. The application allows to upload image file and determines what animal (Cat or Dog) is located on it with using convolutional neural network builded in Keras.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
- Clone this repository with Git Large File Storage(LFS)
git lfs clone https://github.com/mitkir/keras-flask-image-classifier - Open project's directory
cd keras-flask-image-classifier - Install all necessary dependencies
pip install -r requirements.txt - Run application
python application.py - Open
http://127.0.0.1:5000/on your browser - Click the file select button and select test image for classifier.
The application requres some external libraries to run (flask-1.1.1, pillow-6.2.1, tensorflow-2.0.0, keras-2.3.1, numpy-1.17.3, h5py-2.10.0). All dependencies you can find at requirement.txt and install it.
For building image classifier model we used Keras VGG16 with transfer learning (ImageNet) + data augentation and train it with Kaggle dataset. Please, visit https://www.kaggle.com/mitkir/cat-dog-classifier-using-vgg16-tl-da for detail explanation.
