Deep-Learning-Data-Augmentation-Using-Beans-Dataset

Aim: Demo of data augmentation: a technique to increase the diversity of your training set by applying random (but realistic) transformations, such as image rotation

Libraries Used: Matplotlib, Numpy, Tensorflow-keras, Tensorflow-datasets

Learnings:

• Used Keras preprocessing layers for data augmentation - Got 68% accuracy
• Used Custom data augmentation - Got 70% accuracy