In this article, I will brief you through the methods of model optimization using TensorFlow.
When I used to look at my model sizes once training gets over I always gets frustrated because of the number it shows up. But no more worries as the machine learning world has developed immensely over the last decade. In this article, I will brief you through the methods of model optimization using TensorFlow. Tensorflow is an open-sourced deep learning framework created by Google and it should be in the list of every machine learning researchers and aspirants. Arguably every new model is based out of either Tensorflow or Pytorch in this decade. Initially, there were other players in the market like MXNet, Caffe etc but now their share is very meagre when compared to Tensorflow and Pytorch. With weight clustering added to the already present pruning and quantization, there are three major ways by which one can reduce the size of their model without much reduction in metrics.
This video on TensorFlow and Keras tutorial will help you understand Deep Learning frameworks, what is TensorFlow, TensorFlow features and applications, how TensorFlow works, TensorFlow 1.0 vs TensorFlow 2.0, TensorFlow architecture with a demo. Then we will move into understanding what is Keras, models offered in Keras, what are neural networks and they work.
Deep learning is one of the most interesting and promising areas of artificial intelligence (AI) and machine learning currently. With great advances in technology and algorithms in recent years, deep learning has opened the door to a new era of AI applications.
The Association of Data Scientists has announced a hands-on workshop on deep learning model deployment and management.
In this article, we explore gradient descent - the grandfather of all optimization techniques and it’s variations. We implement them from scratch with Python.
In this video we will learn What is Optimizer in Deep Learning? (Keras & TensorFlow)