run entire training set

jackdreilly / SparkADMM

ADMM implementation on Spark Cluster

7 stars 2 forks source link

Open jackdreilly opened 12 years ago

jackdreilly commented 12 years ago

we currently have no way of running

30,000 samples 40,000 features

this needs to be run in a distributed manner because of memory restrictions.

SOMEONE DO THIS