This project requires the following python libraries to be installed:
- NumPy
- Pandas
- tensorflow
- json
- matplotlip
I will train an image classifier to recognize different species of flowers. Imagine using something like this in a phone app that tells you the name of the flower your camera is looking at. I will be using this dataset from Oxford of 102 flower categories.
I will use tensorflow_datasets to load the Oxford Flowers 102 dataset.
This dataset has 3 splits: train, test, and validation.
The validation and testing sets are used to measure the model's performance on data it hasn't seen yet, but you'll still need to normalize and resize the images to the appropriate size.