py-why / py-why/EconML

CausalForest DML Randomness

Open
#971 1 comment 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

Hi, we recently found that we get different prediction results every time we run the causal forest model with ite_estimates = model.effect(df[features])
ite_ci = model.effect_interval(df[features])

The model was pre-trained, saved and then loaded for predictions.

Is it because the effect() and effect_interval() methods often use bootstrap sampling to generate predictions, which introduces randomness?

How could I reproduce predictions with a CausalForest DML model?

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 with the CausalForest DML effect() and effect_interval() calls described in the issue, then reproduce them after saving and loading the pre-trained model. Compare repeated predictions and determine whether the behavior is expected; done means documenting or correcting how reproducible predictions should be obtained.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.