causalml
causalml copied to clipboard
uber
Uplift modeling and causal inference with machine learning algorithms
**Describe the bug** When i run below code with my data : ```python # LR Filter with order 2` method = 'LR' f_imp = filter_method.get_importance(data_train, data_train.columns.tolist(), y_name='label', method=method, experiment_group_column='ff_rate', control_group=0,...
**Describe the bug** Less of a bug than a question: The T-learner takes the mean of the treatment effect `te` which is calculated over all subjects (so the mean over...
## Proposed changes This PR resolves #780 by removing the environment files in `envs/` as well as associated documentation and github build workflows. ## Types of changes What types of...
**Describe the bug** Maintaining environment files in `envs/` and their builds via `.github/workflows/test-conda-venv-install.py` https://github.com/uber/causalml/blob/a0315660d9b14f5d943aa688d8242eb621d2ba76/.github/workflows/test-conda-venv-install.yml#L1-L68 is cumbersome and not easily addressed across platforms due to cython and other platform-specific build dependencies....
## Proposed changes This PR resolves the libxcrypt dependency issue described in #767. It also: - modifies docs/ to update the install process so that it includes libxcrypt (#767) -...
## Proposed changes This PR adds an explicit `treatment` input argument to `CausalTreeRegressor.fit()` and `CausalRandomForestRegressor.fit()`. Originally, it used `sample_weight` for treatment assignments to use the `sklearn.tree` APIs without modification, but,...
**Is your feature request related to a problem? Please describe.** As raised in #798, Uplift Trees/Uplift RandomForest don't support inputs with NaNs. **Describe the solution you'd like** Adding support for...
## Proposed changes Adding basic pre-commit hooks. This will make it so that black formatting is run before every commit. ## Types of changes What types of changes does your...
**Is your feature request related to a problem? Please describe.** [`pygam`](https://github.com/dswah/pyGAM) pins `scipy < 1.12`, which blocks the use of `numpy 2.x` and `Cython 3.x`. **Describe the solution you'd like**...