Custom H2O implementation
August 16, 2018

Custom H2O implementation

Anonymous | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User

Overall Satisfaction with H2O

H2O is used as a core tool across the whole organization. The primary business we are in is measuring the Return on Ad Spend (ROAS) for advertisers, media companies and CPG marketing and product companies.
  • Flexible modeling including Ensemble
  • Open Source - so that we can know what is really happening and can request changes when needed
  • Ability to scale up horizontally by provisioning dynamic clusters
  • Access to core development team and speed of problem resolution and feature additions
  • Better documentation
  • Improve the Visual presentations including charting etc
  • Positive impact: saving in infrastructure expenses - compared to other bulky tools this costs a fraction
  • Positive impact: ability to get quick fixes from H2O when problems arise - compared to waiting for several months/years for new releases from other vendors
  • Positive impact: Access to H2O core team and able to get features that are needed for our business quickly added to the core H2O product
H2O provided all the needed features such as Linear Modeling, Targeted Learning, Predictive Analytics including GLM, Trees, Neural networks and ensemble with ease. We are also able to pick and choose what we want without deploying all the bulky tools unlike others. Able to package H2O jar with our home grown code for remote deployments without worrying able expensive licenses.
It is able to handle large amounts of data. It is best suited when we want to productionalize BI and Analytical applications/features with ease and scale well. Applicable for ensemble learning, data munging, scaled application development.

Not yet ready for fast, quick and dirty prototyping.