39 and Canine is a project that uses Google's Inception v3 model with Tensorflow along with Django as a server to create a web interface to allow users to upload a photo of a dog for breed classification (the model has been trained with ~120 breeds). The web interface is made using materializecss and jQuery It is extension of this project and the Django server is based on this project's, which can be found here on github.
- TensorFlow's tutorial for image classification neural network
- Stanford's collection of dog images http://vision.stanford.edu/aditya86/ImageNetDogs/
- Retrain the model using your images. Refer here.
- Fork and clone this repo
- Replace the generated graph and label files in
/classify_image/inception_model/ - Check the views.py file to ensure that the final result model name is correct (it currently reads final_result:0--this is determined by your tensorflow project)
- Deploy the Django project on heroku or a similar platform