H2O.ai democratizes advanced deep learning for data scientists and developers with H2O Hydrogen Torch
The development of applications that combine sophisticated technologies such as AI and ML is progressing, this time combining multiple deployment outcomes into a single general-purpose, no-code platform. Line of business users can easily move from studying data records to natural language processing, images, and video outputs this way.
Since many software developers can only focus on one of these outputs at a time, this kind of adaptability for non-IT personnel has not been offered in the market. Market heavyweights like DataRobot, Amazon Web Services, Microsoft, DataBricks and SAS do not offer this particular feature. H2O.ai, on the other hand, set out to overcome this problem.
The company is located in Mountain View, California. H2O.ai announced H2O Hydrogen Torch, an important new addition to its open source platform. This feature is a deep learning training engine that the company says enables every size and business sector to easily generate image, video, and natural language processing (NLP) models of tip without script. These models can be used in the field to uncover new business information about consumers, competitors, the market, and other topics.
H2O Hydrogen Torch was created by Kaggle Grandmasters, the world’s most outstanding data scientist, and the solution automatically handles the problematic elements of developing world-class deep learning models, according to CEO and co-founder Sri Ambati. Until now, coding and tuning accurate deep learning models required considerable knowledge and effort. Since data scientists are some of the highest paid IT professionals, these expenses can be costly.
According to Ambati, Smart LOB employees, data scientists and developers can quickly create models for a variety of image, video and NLP processing use cases, such as identifying or classifying objects. objects in images and videos, sentiment analysis or finding relevant information. in texts, using a simple, no-code user interface.
Monitoring foot movements in public buildings, shopping malls and businesses, for example, and noting the frequency of visitors and their movement from place to place are examples of use cases for video. The platform can be used by retailers to see which sales displays are the most popular. According to the CEO, all data is instantly collected and accessible for queries in the H2O.ai analytics engine.
“There’s a lot of unstructured data out there,” the CEO added, referring to photographs and text in companies. “There are a lot of promises that have yet to be fulfilled.” The goal is to allow users to create state-of-the-art models for many types of use cases. We basically offer them these features in Hydrogen Torch to solve different types of use cases.
According to several analyst estimates, unstructured data accounts for 80-90% of data, but only a tiny fraction of companies can benefit from it, according to the CEO.
Deep learning models have the potential to transform industries such as healthcare, which can use computer-assisted disease detection or diagnosis based on medical images, insurance, which can automate the analysis of claims and damages based on reports and images; and manufacturing, which can use predictive maintenance based on images, videos and other sensor data, according to Ambati.
Image and video processing
According to the CEO, Hydrogen Torch can be taught for classification, regression, object identification, semantic segmentation, and learning metrics on images and videos. Hydrogen Torch, for example, can assess medical X-ray images to detect abnormalities in a medical context, with a “person in the know” making the final judgement. According to the CEO, detecting objects in a manufacturing plant to assess if a part is missing, or learning metrics to alert an online store to duplicate photos on a website, are two other use cases. image based.
natural language processing
Hydrogen Torch can be taught for text classification, regression, token classification, range prediction, sequence-to-sequence analysis, and metric learning for text-based or NLP use cases. Natural language processing has many applications, from estimating consumer happiness from transcribed phone calls to using sequence-to-sequence analysis to summarize massive amounts of text, such as medical transcriptions.
According to the CEO, these models can then be automatically compressed for deployment to external Python environments or directly to H2O MLops in a consumable way for production.
More than 20,000 companies worldwide, including AT&T, Allergan, CapitalOne, Commonwealth Bank of Australia, GlaxoSmithKline, Hitachi, Kaiser Permanente, Procter & Gamble, PayPal, PwC, Unilever and Walgreens, use the H2O.ai platform, which currently offers a free trial, according to the CEO.