search

rhiever / sklearn-benchmarks

A centralized repository to report scikit-learn model performance across a variety of parameter settings and data sets.
MIT License
209 stars 53 forks source link

added basic metafeatures/dataset_describe.py #8

Closed harshnisar closed 8 years ago

harshnisar commented 8 years ago

I've added metafeatures/dataset_describe.py. I've implemented a class-method structure. I wasn't too sure but wanted to set the ball rolling.

The public methods of class will only return one value per dataset. These will be the metafeatures. Private methods of class are used for intermediate computations etc.

Dataset (the class) can be initialized with some user-set values (Eg. which of the columns are categorical)

Let me know if you have any suggestions regarding design, module structure etc.

rhiever commented 8 years ago

Great start. Thank you for this!