Here the scenario: A new kaggle competition, a new dataset. Gigabites? ouch! Cold shivers as you anticipate hours waiting to extract features, train models and middle of the night cold feet as you’re just checking that your script is still running.

A data set is said to be large when it exceeds 20% of the available RAM for a single machine. Which for your standard MacBook Pro with 8Gb of RAM, corresponds to a meager 2Gb dataset — size that is becoming more and more frequent these days.

Of course before you actually run out of memory, your machine will slow down to a crawl and your frustration will increase in inverse proportions.

To deal with large data, first aid kit strategies consists in sampling the data to only consider a subset of the whole data or reaching out for more RAM by going to the cloud. Amazon offers boxes with plenty of RAM for pennies/hour. Other options are to use libraries such as Apache Spark’s MLlib, or platforms such as H2O or Dato’s GraphLab Create. R also has a streaming package.

However, if scikit-learn is your weapon of choice for machine learning, you should stick with it and make the best of its out-of-core processing capabilities.

The rest of the story is on the Open Data Science Conference Blog: Riding on Large Data with Scikit-learn by yours truly.

If you liked this post, please share it on twitter And leave me your feedback, questions, comments, suggestions below. Much appreciated :)