GF

Verified User in Financial Services

Enterprise (> 1000 emp.)

6/26/2018

"Trial usage but positive experience"

3.5/5

What do you like best about MLlib?

The best features of the program include increasing speeds of computations and has high quality algorithms. Review collected by and hosted on G2.com.

What do you dislike about MLlib?

I dislike that we have not fully implemented the product so I am not fully informm Review collected by and hosted on G2.com.

GW

Verified User in Wireless

Mid-Market (51-1000 emp.)

6/12/2018

"Full fledged ML library support for Spark"

5/5

What do you like best about MLlib?

Easy to follow if you have the basic understanding of the ML algorithms.

Clear interface to tune in parameters for the algorithm Review collected by and hosted on G2.com.

What do you dislike about MLlib?

I have not faced any issue as such with this library. Review collected by and hosted on G2.com.

SE

Scott E.

Principal Consultant, Data Science

Computer Software

Mid-Market (51-1000 emp.)

7/31/2017

"MLlib is a great API in Spark, which is a great framework that is constantly evolving"

4.5/5

What do you like best about MLlib?

It's FAST, and you can access the API from Python, R, SQL, Scala, or Java -- so works great for data science teams that use multiple languages. Review collected by and hosted on G2.com.

What do you dislike about MLlib?

Sometimes, a feature that is new is only available from one of the languages, but that is OK, they try and expand it to the other languages ASAP! Review collected by and hosted on G2.com.

GF

Verified User in Financial Services

Enterprise (> 1000 emp.)

4/13/2017

"analytics in finance"

4.5/5

What do you like best about MLlib?

MLlib now works on the new DataFrame API and thus is very easy to use. Review collected by and hosted on G2.com.

What do you dislike about MLlib?

Some of the functionality is still not reachable from PySpark Review collected by and hosted on G2.com.