py-why / py-why/EconML

Results differ between versions '0.8.1' and '0.9.0b1'

Open
#387 26 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
4.8k
Forks
827
PR merge metrics
No merged PRs in 30d

Description

I have upgraded econml from '0.8.1' to '0.9.0b1' (pip install -e git+https://github.com/microsoft/EconML.git@master#egg=econml) and I changed nothing to my code.
The results I obtain are however very different when fitting the following estimator:

est_forest_gb = ForestDML(
                            model_y = GradientBoostingRegressor(n_estimators = 200, random_state = 42), 
                            model_t = GradientBoostingRegressor(n_estimators = 200, random_state = 42), 
                            n_estimators = 200,
                            n_crossfit_splits = [(fold0, fold1), (fold1, fold0)],
                            random_state = 123)

Do you have an idea of what could explain the difference?
Thanks.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at the ForestDML estimator with the supplied GradientBoostingRegressor settings and compare behavior between econml 0.8.1 and 0.9.0b1. Read the 26-comment thread and review changes between those versions; done means the cause of the differing results is reproduced and clearly identified.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, scikit-learn
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.