Deep Learning Gradiet Check test detecs too large error

h2oai / h2o-3

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.

Apache License 2.0

6.9k stars 2k forks source link

On master branch, in DeepLearningGradientCheck class, specifically in gradientCheck() method, approx. on line 389, an assert failed and the error propagated was way above expected threshold.

{code:java} java.lang.AssertionError: Error too large: 0.17404528880393297 >= 0.02 at hex.deeplearning.DeepLearningGradientCheck.gradientCheck() {code}

http://mr-0xc1:8080/view/H2O-3/job/h2o-3-nightly-pipeline/job/master/2240/testReport/junit/hex.deeplearning/DeepLearningGradientCheck/Java_8_JUnit___gradientCheck/

h2oai / h2o-3

Deep Learning Gradiet Check test detecs too large error #9294