This repo contains the model for the notebook Image Classification using BigTransfer (BiT).
Full credits go to Sayan Nath
Reproduced by Rushi Chaudhari
BigTransfer (also known as BiT) is a state-of-the-art transfer learning method for image classification.
The Flower Dataset is A large set of images of flowers
The following hyperparameters were used during training:
RESIZE_TO = 384 CROP_TO = 224 BATCH_SIZE = 64 STEPS_PER_EPOCH = 10 AUTO = tf.data.AUTOTUNE # optimise the pipeline performance NUM_CLASSES = 5 # number of classes SCHEDULE_LENGTH = ( 500 # we will train on lower resolution images and will still attain good results ) SCHEDULE_BOUNDARIES = [ 200, 300, 400, ]
The hyperparamteres like
SCHEDULE_BOUNDARIES are determined based on empirical results. The method has been explained in the original paper and in their Google AI Blog Post.
SCHEDULE_LENGTH is aslo determined whether to use MixUp Augmentation or not. You can also find an easy MixUp Implementation in Keras Coding Examples.
- Downloads last month
Inference API does not yet support keras models for this pipeline type.