Python Application that can train an image classifier on a dataset, then predict new images using the trained model
In this first part of the project, I worked through a Jupyter notebook to implement an image classifier with PyTorch.
After I built and trained the deep neural network on the flower data set, I converted it into an application that others can use. My application is a pair of Python scripts that run from the command line.
The project includes next to the Jupyter Notebook, two files train.py and predict.py:
- train.py will train a new network on a dataset and save the model as a checkpoint.
- predict.py uses a trained network to predict the class for an input image-
- I also created a files just for the functions and classes relating to the model: dataloader.py, fuctions_train.py, functions_predict.py
Clone the repository and open the file in Jupyter Notebook respect. the python files in a text editor and run them fro the command line
$ git clone https://github.com/JannisMueller/Image-Classifier-Project.git