Robert Gruener

icon company @Netflix icon location Pittsburgh icon location Currently working on ML infar @Netflix Formerly did art intelligence @sothebys self driving cars @uber and music recommendations @spotify

Results 6 comments of Robert Gruener

Petastrom fails with big datasets

This is definitely a spark problem. I believe that what is happening is that when spark is serializing the data between the python and java processes it is failing since...

gcloud emulator docs

Nice, I will add a link in the README. A docker version is still useful for portability and integration testing. But it would actually be better to just run the...

Maven import not working

`LATEST_RELEASE` is a placeholder, I probably should have made that more clear in the documentation. The latest actual release would be `0.1.0-beta14` but as I am not longer maintaining this...

fix the element wise aggregation

I can try, i am going away after this week for a little bit. If I have time this week I will prioritize it

Switch average implementation to utilize running average to prevent overflow

Will update docs assuming the change looks ok

Switch average implementation to utilize running average to prevent overflow

> approving to negate my request changes. given the loss of commutativity with the running average computation (due to double multiple / divide errors) - i don't know what is...