v0.7.7

Author: paulbkoch

.github/workflows/ci.yml

@@ -685,6 +685,8 @@ jobs:
           token: ${{ secrets.CODECOV_TOKEN }}
 
   testR:
+    permissions:
+      contents: read
     needs: [R]
     strategy:
       matrix:

CHANGELOG.md

@@ -4,6 +4,22 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and the versioning is mostly derived from [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [v0.7.7] - 2026-03-14
+### Added
+- support for python 3.14
+### Changed
+- optimization of EBM predict speed, especially for single samples
+- replaced React with plain JavaScript in the inline visualization bundle
+- reduced the size of the javascript visualization bundle
+- changed the format of floating point and boolean categorical strings
+### Fixed
+- fixed bug in handling the newer pd.StringDtype
+- fixed issue where -0.0 creates a nondeterministic category
+### Removed
+- removed direct python dependencies: ipykernel, dill, pytest-runner, jupyter
+- removed pin on dask<3
+- removed JS dependencies: react, react-cytoscapejs, react-dom, react-hot-loader, react-plotly.js, react-select
+
 ## [v0.7.6] - 2026-02-25
 ### Changed
 - improved speed of EBMs during predict, especially for datasets with small numbers of samples

python/interpret-core/interpret/_version.py

@@ -2,4 +2,4 @@
 # Distributed under the MIT software license
 
 # NOTE: Version is replaced by a regex script.
-__version__ = "0.7.6"
+__version__ = "0.7.7"

python/interpret-core/setup.py

@@ -10,7 +10,7 @@
 from setuptools.command.sdist import sdist
 
 # NOTE: Version is replaced by a regex script.
-version = "0.7.6"
+version = "0.7.7"
 
 
 def _copy_native_code_to_setup():
@@ -224,7 +224,8 @@ def run(self):
         # Explainers
         "lime": ["lime>=0.1.1.33"],
         "sensitivity": ["SALib>=1.3.3"],
-        "shap": ["shap>=0.28.5"],
+        # installing ipywidgets removes crud in SHAP notebooks during fitting
+        "shap": ["shap>=0.28.5", "ipywidgets>=7.4.2"],
         "linear": ["scikit-learn>=1.6.0"],
         "skoperules": ["skope-rules>=1.0.1"],
         "treeinterpreter": ["treeinterpreter>=0.2.2"],
@@ -246,7 +247,6 @@ def run(self):
             "nbformat>=4.4.0",
             "selenium>=3.141.0",
             "pytest-cov>=2.6.1",
-            "ipywidgets>=7.4.2",
         ],
     },
 )

python/interpret/setup.py

@@ -5,7 +5,7 @@
 
 name = "interpret"
 # NOTE: Version is replaced by a regex script.
-version = "0.7.6"
+version = "0.7.7"
 long_description = """
 In the beginning machines learned in darkness, and data scientists struggled in the void to explain them.
 

shared/vis/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@interpretml/interpret-inline",
-  "version": "0.7.6",
+  "version": "0.7.7",
   "description": "Interpret inline library for rendering visualizations across all notebook environments.",
   "main": "index.js",
   "keywords": [],