EasyGLM
Fit insurance pricing GLMs, check model performance and export rating tables from a local browser workbench.
Install and open
Requires Python 3.10–3.14. Run in a terminal:
python -m pip install --upgrade easy-glm
easy-glm-workbench
Your browser opens automatically. Keep the terminal open while you work. Restart the workbench after upgrading.
Install as easy-glm; import in Python as easy_glm. The standard installation
includes the workbench, GLM fitting, CatBoost interactions and Optuna tuning.
Build a model
- Load data. Open your file or try the French motor claim-frequency or Swedish motorcycle claim-cost example. Supports CSV, Parquet, Excel, Arrow/Feather and SAS files.
- Choose variables. Set the target, exposure/weight and predictors. Check for missing data, possible leakage and redundant predictors before fitting.
- Explore. See observed rates and exposure by variable.
- Fit. Choose a model family, training/holdout split, factor shapes and interactions, then click Fit model.
- Review. Check actual versus expected, lift, Gini and variable importance. Compare alternative models.
- Adjust. Smooth, cap/floor or edit relativities, then apply your changes. The original fit stays available for comparison.
Supports Poisson, Gamma, Tweedie, Gaussian, binomial and inverse Gaussian GLMs, with regularisation and two-stage interactions.
Export and keep your work
| Export | What you get |
|---|---|
| Excel | Applied rating tables |
Scorer (.easyglm) |
Applied rates for scoring new data |
| Project JSON | Model setup, applied adjustments and named snapshots |
| HTML report | Data summaries, importance, coefficient paths, rating factors and diagnostics |
| Python script | Reproduce the model from its source data |
Before closing, save the project JSON and scorer. Reopening a project requires its source data and a refit; fitted runs and session Undo are not saved in project JSON.
Already working in Python?
Open a pandas or Polars dataframe in the workbench:
import easy_glm
easy_glm.launch_workbench(data=df)
Workbench walkthrough · Python examples · Changelog
Why I built this
This started with my wish to port R's aglm to Python and give it a GUI. It's built primarily for my own pricing work and will probably have plenty of bugs, because… vibecode yo. Check the results before using them.
1 maintainer
Metadata
1 maintainer
Release files for easy-glm 0.472
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| easy_glm-0.472.tar.gz | 812.2 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| easy_glm-0.472-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 1.4 MB
Release files / easy_glm-0.472.tar.gz
| Download URL | easy_glm-0.472.tar.gz |
|---|---|
| Size | 812.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0a00d3975bad8cf468fbf21959bce2c507a33f969ba7557aa0dce736e53f4c32
|
|
BLAKE2b-256 checksum How to use checksums |
623a6685bec5b8a1b01edf7a7bc78bc2e2400d45216ba7e04194fd4707520034
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency logRelease files / easy_glm-0.472-py3-none-any.whl
| Download URL | easy_glm-0.472-py3-none-any.whl |
|---|---|
| Size | 590.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
f0b01d76b58a2694333d96e3b06c806c3ea4e45f2c6844d272f9b4e2c82a2390
|
|
BLAKE2b-256 checksum How to use checksums |
0b7cee0d336d4f7f32cacdc9eaec86cb1b89137ff82ac3a0776dcc2541193776
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency log
