AI cloud leader, H2O.AI has launched H2O Hydrogen Torch, a deep learning training engine that lets companies of any size or industry create advanced images, videos, and natural language processing (NLP) models without touching any code.

Users are able to quickly create models using a simple, no-code interface for a range of image, video, and NLP processing use cases, including finding relevant information in text, identifying or classifying objects.

“Accelerated by COVID-19, video streams, speech, audio podcasts, email, and natural language text have become the fastest-growing data for our customers in every industry. Transforming and fine-tuning pre-built deep learning models to deliver high accuracy requires a no-code AI Engine to democratize AI for these use cases,” said Sri Ambati, CEO and founder, “H2O Hydrogen Torch does exactly that by bringing best practices from Grandmasters to tackle problems ranging from improving in-store customer experiences, identifying fashion trends, and discovering vaccines, to saving lives with video-enabled drones fighting fires with AI on the edge. With H2O Hydrogen Torch as a core AI Engine of the H2O AI Cloud, our customers can train models in deep learning and better serve their customers and challenge tech giants.”

In addition to providing opportunities for transforming healthcare (identifying diseases on the basis of medical images), insurance (analyzing claims and damage reports), and manufacturing (predicting maintenance by analyzing images, video, and other sensor data), deep learning models create opportunities to transform other sectors as well.

Hydrogen Torch can be used for object detection, semantic segmentation, classification, regression, and metric learning for images and videos.

Textual classification and regression, token classification, span prediction, sequence-to-sequence analysis, and metric learning can all be trained using Hydrogen Torch for text, or NLP-based use cases.

Some NLP uses include predicting customer satisfaction from the transcription of phone calls to summarizing an entire written work, such as a medical record, in a few sentences using sequence-to-sequence analysis.

Models can then be packaged automatically so they can be easily deployed to Python environments or directly to H2O MLOps using a consumable format.