Compare MLlib, XGBoost, and scikit-learn

Pricing

 
Free Trial Unavailable
Free Trial Unavailable
Free Trial Unavailable
MLlib
Free Trial Unavailable
XGBoost
Free Trial Unavailable
scikit-learn
Free Trial Unavailable

Ratings

Meets Requirements
Meets Requirements
7.8
9.2
9.6
Ease of Use
Ease of Use
8.9
8.9
9.6
Ease of Setup
Ease of Setup
8.7
8.5
9.6
Ease of Admin
Ease of AdminNot enough data available
8.3
9.4
Quality of Support
Quality of SupportNot enough data available
7.6
9.4
Ease of Doing Business With
Ease of Doing Business WithNot enough data available
8.3
9.2
Product Direction (% positive)
Product Direction (% positive)
4.2
6.5
9.7
Meets Requirements
MLlib
7.8
XGBoost
9.2
scikit-learn
9.6
Ease of Use
MLlib
8.9
XGBoost
8.9
scikit-learn
9.6
Ease of Setup
MLlib
8.7
XGBoost
8.5
scikit-learn
9.6
Ease of Admin
MLlib
Not enough data available
XGBoost
8.3
scikit-learn
9.4
Quality of Support
MLlib
Not enough data available
XGBoost
7.6
scikit-learn
9.4
Ease of Doing Business With
MLlib
Not enough data available
XGBoost
8.3
scikit-learn
9.2
Product Direction (% positive)
MLlib
4.2
XGBoost
6.5
scikit-learn
9.7

Reviewers' Company Size

Small-Business (50 or fewer emp.)
Small-Business
(50 or fewer emp.)
0%
50.0%
34.1%
Mid-Market (51-1000 emp.)
Mid-Market
(51-1000 emp.)
50.0%
16.7%
22.0%
Enterprise (> 1000 emp.)
Enterprise
(> 1000 emp.)
50.0%
33.3%
43.9%
MLlib
Small-Business
0%
Mid-Market
50.0%
Enterprise
50.0%
XGBoost
Small-Business
50.0%
Mid-Market
16.7%
Enterprise
33.3%
scikit-learn
Small-Business
34.1%
Mid-Market
22.0%
Enterprise
43.9%

Reviewers' Industry

 
Financial Services
33.3%
Computer Software
25.0%
Computer Software
41.5%
 
Computer Software
33.3%
Financial Services
16.7%
Information Technology and Services
17.1%
 
Wireless
16.7%
Research
8.3%
Higher Education
9.8%
 
Telecommunications
16.7%
Marketing and Advertising
8.3%
Hospital & Health Care
7.3%
 
Information Technology and Services
8.3%
Education Management
4.9%
 
Other
0.0%
Other
33.3%
Other
19.5%
MLlib
Financial Services
33.3%
Computer Software
33.3%
Wireless
16.7%
Telecommunications
16.7%
Other
0.0%
XGBoost
Computer Software
25.0%
Financial Services
16.7%
Research
8.3%
Marketing and Advertising
8.3%
Information Technology and Services
8.3%
Other
33.3%
scikit-learn
Computer Software
41.5%
Information Technology and Services
17.1%
Higher Education
9.8%
Hospital & Health Care
7.3%
Education Management
4.9%
Other
19.5%

Reviews

Most Helpful Favorable Review
Most Helpful Favorable Review
GF
G2 User in Financial Services

MLlib now works on the new DataFrame API and thus is very easy to use.

G
G2 User

Runs well in basically every situation, handles missing data well, relatively lightweight.

Rishab G.
G2 User in Computer Software

Documentation has great explanation and is very easy to implement.

Most Helpful Critical Review
Most Helpful Critical Review
Saeid A.
G2 User in Outsourcing/Offshoring

It is rigid with some of the algorithms, specially with advanced one like neural network. For instance, you are unable to change activation functions of a neural network. You can either use Sigmoid for all the layers, or tanh which is not really making...

GF
G2 User in Financial Services

There's not much to dislike. It's been pretty popular as a decision tree algorithm and rightly remains a reliable choice for data science applications. Only wished it was developed sooner!

 
MLlib
Most Helpful Favorable Review
GF
G2 User in Financial Services

MLlib now works on the new DataFrame API and thus is very easy to use.

Most Helpful Critical Review
Saeid A.
G2 User in Outsourcing/Offshoring

It is rigid with some of the algorithms, specially with advanced one like neural network. For instance, you are unable to change activation functions of a neural network. You can either use Sigmoid for all the layers, or tanh which is not really making...

XGBoost
Most Helpful Favorable Review
G
G2 User

Runs well in basically every situation, handles missing data well, relatively lightweight.

Most Helpful Critical Review
GF
G2 User in Financial Services

There's not much to dislike. It's been pretty popular as a decision tree algorithm and rightly remains a reliable choice for data science applications. Only wished it was developed sooner!

scikit-learn
Most Helpful Favorable Review
Rishab G.
G2 User in Computer Software

Documentation has great explanation and is very easy to implement.

Screenshots

 No screenshots providedNo screenshots providedNo screenshots provided
MLlib
No screenshots provided
XGBoost
No screenshots provided
scikit-learn
No screenshots provided
Request More InformationRequest More InformationRequest More InformationRequest More InformationRequest More InformationRequest More Information