MIT License
Beyond the Anaconda distribution of Python, the following packages need to be installed:
In this project, I built and trained a neural network model with CNN (Convolutional Neural Networks) transfer learning, using 8351 dog images of 133 breeds. CNN is a type of deep neural networks, which is commonly used to analyze image data. Typically, a CNN architecture consists of convolutional layers, activation function, pooling layers, fully connected layers and normalization layers. Transfer learning is a technique that allows a model developed for a task to be reused as the starting point for another task. The trained model can be used by a web or mobile application to process real-world, user-supplied images. Given an image of a dog, the algorithm will predict the breed of the dog. If an image of a human is supplied, the code will identify the most resembling dog breed.
Below are main foleders/files for this project:
Note: The dog image dataset used by this project can be downloaded here: https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/dogImages.zip The human image dataset can be downloaded here: https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/lfw.zip
Project files can be found in this github repo: https://github.com/swang13/dog-breeds-classification More discussions can be found in this blog: https://medium.com/@wangshuocugb2005/dog-breeds-classification-with-cnn-transfer-learning-92217cba3129
Credits must be given to Udacity for the starter codes and data images used by this project.