Tomer Kaftan

The Least Squares Estimator should have the option to use a local least squares solver. A cost model needs to be built for it, and it needs to be added...

Integrate Block Operators more neatly with the DAG

2

Currently so as to be easily chainable with the rest of the code, block operators (such as block solves and block transformers) take a single complete RDD and manually split...

Figure out future DAG operators

15

There are some concepts we've talked about (and have in some cases touched using FunctionNodes #121) that we need to figure out how to cleanly integrate into our pipelines interfaces....

In Timit features -> solve, we need to add a seed to the Cauchy distribution after @Vaishaal's now-merged pull request into Breeze to add an implicit RandBasis parameter to the...

Create a KeystoneML style guide

1

Should contain information such as: - What should be a Transformer vs. Estimator vs. LabelEstimator vs. FunctionNode vs. Evaluator vs. Loader vs. util method - What packages to put things...

We're specifying types in places the scala compiler can infer it (e.g. NewsgroupsPipeline `NGramsFeaturizer[String]` and anywhere we're using the cacher node) We should clean these up.

Make all nodes case classes or have companion object constructor

1

Discussion with @etrain suggests that we should be very consistent about how nodes are constructed. Because: - @etrain doesn't like `new` - including empty parens in some cases (case classes...

Updates Shark support to 0.9.1 Adds Spark SQL 1.2.0 support