A neural network build on Food101 TensorFlow Dataset(TFDS) to classify various food images into their 101 categories. Convolutional neural network architecture has been used. (CV) Using the concept of Transfer Learning, build an architecture with feature_extraction model (EfficientNetB0 Architecture). Created efficient data pipelines using tf.data API methods like batch() and prefetch(), for faster training and efficient training of data. Used the concept of mixed precision training for faster GPU computations. Fine Tuned the whole model and got 86% accuracy. Made by using a convolutional neural network (CNN)