Congratulations on completing all the lessons in this course! As the final part of your scholarship challenge, you'll be completing a project to test what you've learned. Here, you'll build an image classifier from scratch that will identify different species of flowers.
The data set contains images of flowers from 102 different species. We've provided a training set and a validation set. You can download the images from here as a zipped archive. Just uncompress it and you should be good to go.
Project Notebook We've built a Jupyter Notebook to guide you through developing and training your model. You can get it by cloning this repo on GitHub:
git clone https://github.com/udacity/pytorch_challenge.git In the repo, you'll find the notebook, a couple assets, and a JSON file that maps the category codes to flower names.
For the most part you'll be writing all the code yourself. Feel free to build and train the image classifier in any way you want.
Test Score As part of the challenge, you'll be assessed on how well your model performs on a test set of flower images. You'll want to save your trained model as a checkpoint. We've prepared another notebook for you that will lead you through loading in your model and running the assessment code. We'll calculate a score using your model on the test data, then record it for use in awarding scholarships. Head to the next page for the model assessment.