H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Means, PCA, Generalized Additive Models (GAM), RuleFit, Support Vector Machine (SVM), Stacked Ensembles, Automatic Machine Learning (AutoML), etc.
The above papers are looking at images as input data, which benefit greatly from convolutional layers, which we currently don’t support. Hence, arbitrary images are not currently handled well in H2O’s Deep Learning, but any structured data is (where each feature/predictor/column has a consistent meaning). For a general dataset with say 17 column of (age, income, zip code, number of pets, etc.), it’s not as straight-forward to look at these kinds of visualizations, but I’m sure they can still be very interpretable if you know your features well.
http://arxiv.org/pdf/1412.0035v1.pdf http://arxiv.org/pdf/1506.02753.pdf
The above papers are looking at images as input data, which benefit greatly from convolutional layers, which we currently don’t support. Hence, arbitrary images are not currently handled well in H2O’s Deep Learning, but any structured data is (where each feature/predictor/column has a consistent meaning). For a general dataset with say 17 column of (age, income, zip code, number of pets, etc.), it’s not as straight-forward to look at these kinds of visualizations, but I’m sure they can still be very interpretable if you know your features well.