Bloom is a command line app utilizing PyTorch and Jupyter notebooks to develop an image classifier for identifying different species of flowers using deep learning techniques. which is the final project in Udacity Al programming with Python Nanodegree
The project is broken down into three steps:
-Load and preprocess the image dataset -Train the image classifier on your dataset -Use the trained classifier to predict image content
you can use the app by running the 'train.py' and 'predict.py' files in the terminal
using train.py successfully it will train a new network on a dataset of images and saves the model to a checkpoint while The training loss, validation loss, and validation accuracy are printed out. The user has been given the freedom to choose the appropriate architecture and whether they want to use a GPU.
after running train.py, you can load the checkpoint and predict the desired picture using predict.py as it allows the user to print out the top K classes along with associated probabilities