AutoML or automated machine learning has witnessed an exponential rise when it comes to being adopted by the organisations. AutoML can be said as the next wave of machine learning which helps to bridge the gap between talent landscape and complex workflows. Previously in one of our articles, we had discussed the functionality of AutoML and how it solves data science problems.
Building complex deep learning models from scratch can be said as one of the most time-consuming and challenging tasks in an organisation. It consists of several layers and requires a considerable amount of time as well as an expert to build a robust model. And, that is why almost all the organisations nowadays have been using AutoML in one way or another to complete a task within a shorter period.
E-commerce giant Amazon has been doing a lot when it comes to emerging technologies. A few days ago the e-comm giant unveiled its all-new quantum computing service known as Braket. And, recently, the developers at Amazon have announced the launch of a new open-source library known as AutoGluon.
AutoGluon is an open-source library for developers in order to build machine learning models and applications easily. The library enables easy-to-use and easy-to-extend AutoML with a focus on deep learning and real-world applications of spanning image, text, or tabular data.
AutoGluon automates machine learning tasks enabling a developer to achieve strong predictive performance in machine learning applications easily. With the help of this toolkit, it is easy to train and deploy high-accuracy deep learning models with just a few lines of codes. The idea behind introducing the library is to make deep learning and machine learning available to everyone. In a report, Jonas Mueller, AWS Applied Scientist said, “We developed AutoGluon to truly democratise machine learning, and make the power of deep learning available to all developers.”
Some of the essential features of this library are mentioned below:
- AutoGluon helps in quickly prototyping deep learning solutions for image, text or tabular data with just a few lines of codes.
- This library leverages automatics hyperparameter tuning, model selection, architecture search, and data processing.
- It automatically utilises the state-of-the-art deep learning techniques without the need of any prior experts.
- Not only this library improves the existing bespoke models and data pipelines but also the toolkit can be customised as per the developer’s use case.
AutoGluon In Image Classification
For image classification, AutoGluon can automatically train models with different hyperparameter configuration and returns the model which has achieved the highest level of accuracy with just one simple call function:
AutoGluon’s `fit </api/autogluon.task.html#autogluon.task.ImageClassification.fit>`
AutoGluon in Text Classification
For classifying the snippets of texts such as sentences or short paragraphs, this toolkit provides a simple call function such as fit() which helps in automatically producing high-quality text classification models. The above call function can train the accurate neural networks on the provided text dataset as well as a
AutoGluon in Tabular Prediction
AutoGluon can produce highly-accurate models to predict the values in one column of a data table based on the rest of the columns’ values by using a simple call such as fit(). This library can be used with tabular data for both classification and regression problems.
There has always been a trust issue of having a black box when it comes to the outcomes provided by deep learning models. For instance, sectors like healthcare and military where the complex algorithms need to make crucial decisions related to human lives — AutoGluon automates such decisions by leveraging the available compute resources to create a model with high accuracy within its specific time.
This toolkit requires Python version 3.6 or more. Amazon’s AutoGluon can serve as an alternative to popular libraries like Keras and Theano. Currently, the library supports only Linux installation, while other versions will be available soon.
Enjoyed this story? Join our Telegram group. And be part of an engaging community.
Provide your comments below
What's Your Reaction?
A Technical Journalist who loves writing about Machine Learning and Artificial Intelligence. A lover of music, writing and learning something out of the box. Contact: firstname.lastname@example.org