automl/auto-sklearn

Avoid costly re-building of pipelines

Open

#443 opened on Mar 21, 2018

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (7,270 stars) (1,265 forks)batch import
Good first issueenhancement

Description

Currently, SMAC suggests hyperparameter configurations which are independent of the dataset size. For example, the hyperparameter classifier:max_features which is specified between zero and one is transformed according to max_features = int(n_features ** classifier:max_features). Assuming the dataset in question has only 10 features, SMAC does not know that most values of the tuned hyperparameter map to the same hyperparameter applied to the actual model. Therefore, one needs to track the 'actual' hyperparameters after transformation and check whether they are re-used, and return a cached function value to SMAC if done so.

Initial experiments suggest that 1-2% of the overall runs are actually re-optimizations.

Contributor guide

Tech stack
python
Domain
machine learningperformance
Issue type
performance
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
3-5 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
Pythonauto sklearnhyperparameter tuning
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
First, understand how SMAC maps hyperparameters to model values by examining the transformation code in auto sklearn, likely in the smac module. Then, implement a caching mechanism that stores the actual hyperparameter values after transformation and returns cached results when re optimizations are detected. Look at the existing SMAC interface to integrate the cache without breaking functionality.
Avoid costly re-building of pipelines · automl/auto-sklearn#443 | Good First Issue