Refactor docs. Migrate form deprecated mkdocs to zensical

.github/dependabot.yml

@@ -6,6 +6,10 @@
 
 version: 2
 updates:
+ - package-ecosystem: "github-actions"
+   directory: "/"
+   schedule:
+     interval: weekly
  - package-ecosystem: "devcontainers"
    directory: "/"
    schedule:

.github/skills/update-docs/SKILL.md

@@ -26,12 +26,12 @@ description: "Update the documentation and changelog, and reconcile GitHub issue
 8. Determine whether changelog entries are managed by CI (for example, `python-semantic-release`).
    - Check release automation config first (for example, `[tool.semantic_release]` in `pyproject.toml` and release workflow files) to identify the authoritative changelog file path and casing.
    - If semantic-release (or equivalent) is configured to write changelog entries, **do not manually append a release section**. Instead, verify that the generated/release-bound changelog content is consistent with steps 2-7 and only make non-conflicting adjustments explicitly requested by the user.
-   - If no automation manages changelog entries, add the changelog to `mkdocs/docs/Changelog.md` (note the path/casing). Follow the existing format. If the file is empty or does not exist, use the Keep a Changelog format (https://keepachangelog.com) with sections: Added, Changed, Deprecated, Removed, Fixed, Security. Include the version number and release date at the top of the new section, and keep the issue/PR links added in step 5.
+   - If no automation manages changelog entries, add the changelog to `docs/Changelog.md` (note the path/casing). Follow the existing format. If the file is empty or does not exist, use the Keep a Changelog format (https://keepachangelog.com) with sections: Added, Changed, Deprecated, Removed, Fixed, Security. Include the version number and release date at the top of the new section, and keep the issue/PR links added in step 5.
 9. Reconcile GitHub issues and milestones per the user's approval in step 7 (skip this step entirely if the GitHub CLI was unavailable in step 4).
    - Assign approved issues and merged PRs that lack a milestone to the release milestone: `gh issue edit <N> --milestone "<title>"` / `gh pr edit <N> --milestone "<title>"`.
    - Close each approved open issue with a comment linking the release: `gh issue close <N> --comment "Resolved in <version> release."`. Note that issues referenced with a closing keyword in a PR merged to the default branch are closed automatically by GitHub — do not double-close; only close issues that are still open after the merge.
    - If the user approved creating a milestone, create it (`gh api repos/:owner/:repo/milestones -f title="<title>" -f state="open"`) and assign the release's issues/PRs to it before closing.
    - Once every issue in the release milestone is closed, close the milestone: find its number via `gh api repos/:owner/:repo/milestones --jq '.[] | select(.title=="<title>") | .number'` then `gh api repos/:owner/:repo/milestones/<number> -X PATCH -f state="closed"`.
    - Report the outcome of each action (issues closed, milestone closed/created, anything skipped) back to the user.
-10. Now, review the documentation throughout the `mkdocs/docs` folder to ensure any version-specific information is updated for the new version. This includes updating version numbers, dates, and any other relevant details. In case new features are implemented, make sure to add documentation for those features as well into the appropriate sections. In case no sections exists, please create a new section for the new feature and add the documentation there. Make sure to format it properly and include any relevant examples or usage instructions.
+10. Now, review the documentation throughout the `docs` folder to ensure any version-specific information is updated for the new version. This includes updating version numbers, dates, and any other relevant details. In case new features are implemented, make sure to add documentation for those features as well into the appropriate sections. In case no sections exists, please create a new section for the new feature and add the documentation there. Make sure to format it properly and include any relevant examples or usage instructions.
 11. After updating the documentation, commit the changes to the repository with a clear commit message indicating that the documentation (and changelog verification/update mode) has been completed for the new release. If changelog sections are CI-generated, avoid committing manual duplicate release-note blocks unless explicitly requested. For example, "Update documentation and changelog workflow alignment for version X.Y.Z release." Reference the reconciled issues in the commit body where appropriate (e.g., `Refs #42, #43`). Commit on the current branch. Do not push or create a pull request unless explicitly instructed. If the repository requires pull requests for the main branch, note this to the user and stop.

.github/workflows/workflow.yml

@@ -85,7 +85,7 @@ jobs:
         run: |
           uv run poe coverage
 
-      - uses: actions/upload-artifact@v4
+      - uses: actions/upload-artifact@v7
         with:
           name: coverage-py${{ matrix.python-version }}
           path: ./coverage.xml
@@ -139,9 +139,113 @@ jobs:
         run: |
           uv run poe test-slow
 
+  release-preflight:
+    name: Release Preflight
+    needs: test-slow
+    runs-on: ubuntu-latest
+    if:
+      ${{ success() && (github.ref == 'refs/heads/main' || github.event_name ==
+      'pull_request') }}
+    # Least privilege: this job checks out and runs untrusted PR code
+    # (scripts/release_preflight.py), so it must never hold a write-scoped token.
+    # Posting the report comment is handled by the separate comment-preflight
+    # job below, which never executes PR code.
+    permissions:
+      contents: read
+    steps:
+      - name: Checkout 🛎
+        uses: actions/checkout@v7
+        with:
+          fetch-depth: 0
+
+      - name: Run release preflight report
+        id: preflight
+        run: |
+          set +e
+          python3 scripts/release_preflight.py --strict --output release-preflight-report.md
+          exit_code=$?
+          echo "exit_code=$exit_code" >> "$GITHUB_OUTPUT"
+          exit 0
+
+      - name: Upload preflight artifact
+        uses: actions/upload-artifact@v7
+        with:
+          name: release-preflight-report
+          path: release-preflight-report.md
+          if-no-files-found: error
+
+      - name: Fail on preflight violations
+        if: steps.preflight.outputs.exit_code != '0'
+        run: |
+          echo "Release preflight failed with exit code ${{ steps.preflight.outputs.exit_code }}."
+          exit 1
+
+  comment-preflight:
+    name: Comment Preflight Report
+    needs: release-preflight
+    runs-on: ubuntu-latest
+    # Isolated commenting job: it does NOT check out or execute any PR code, it
+    # only downloads the report artifact and posts it. This is the only job that
+    # holds a write-scoped token, and it is gated to same-repo PRs (fork PRs
+    # receive a read-only token from GitHub regardless). always() lets the
+    # comment post even when the preflight check fails upstream.
+    if: always() && github.event_name == 'pull_request' && github.event.pull_request.head.repo.full_name == github.repository
+    permissions:
+      pull-requests: write
+    steps:
+      - name: Download preflight artifact
+        uses: actions/download-artifact@v8
+        continue-on-error: true
+        with:
+          name: release-preflight-report
+
+      - name: Comment preflight report on PR
+        uses: actions/github-script@v8
+        with:
+          script: |
+            const fs = require('fs');
+            const path = 'release-preflight-report.md';
+            const marker = '<!-- release-preflight-report -->';
+            const report = fs.existsSync(path)
+              ? fs.readFileSync(path, 'utf8')
+              : 'Preflight report file was not generated.';
+            const body = `${marker}\n## Release Preflight\n\n\
+            _Automated check that complements semantic-release changelog generation._\n\n\
+            \`\`\`markdown\n${report}\n\`\`\``;
+
+            const { owner, repo } = context.repo;
+            const issue_number = context.issue.number;
+
+            const comments = await github.paginate(github.rest.issues.listComments, {
+              owner,
+              repo,
+              issue_number,
+              per_page: 100,
+            });
+
+            const existing = comments.find(
+              (c) => c.user?.type === 'Bot' && c.body?.includes(marker),
+            );
+
+            if (existing) {
+              await github.rest.issues.updateComment({
+                owner,
+                repo,
+                comment_id: existing.id,
+                body,
+              });
+            } else {
+              await github.rest.issues.createComment({
+                owner,
+                repo,
+                issue_number,
+                body,
+              });
+            }
+
   release:
     name: Create Release package
-    needs: test-slow
+    needs: release-preflight
     if: success() && github.ref == 'refs/heads/main' &&
       !contains(github.event.head_commit.message, '[skip release]')
     runs-on: ubuntu-latest
@@ -169,7 +273,7 @@ jobs:
           install-args: "--extra report --extra rna --extra tabpfn --extra tabicl"
 
       - name: Python Semantic Release
-        uses: python-semantic-release/python-semantic-release@v10.2.0
+        uses: python-semantic-release/python-semantic-release@2896129e02bb7809d2cf0c1b8e9e795ee27acbcf # v10.2.0
         id: release
         with:
           github_token: ${{ secrets.PAT_RELEASE_PIPELINE }}
@@ -212,16 +316,16 @@ jobs:
           path: dist
 
       - name: Upload package to Test PyPI
-        uses: pypa/gh-action-pypi-publish@release/v1
+        uses: pypa/gh-action-pypi-publish@76f52bc884231f62b9a034ebfe128415bbaabdfc # v1.12.4
         with:
           repository-url: https://test.pypi.org/legacy/
           skip-existing: true
           packages-dir: dist/
 
   publish-pypi:
     name: Publish to PyPI
-    needs: release
-    if: needs.release.outputs.released == 'true'
+    needs: [release, test-pypi-publish]
+    if: needs.release.outputs.released == 'true' && needs.test-pypi-publish.result == 'success'
     environment:
       name: pypi
       url: https://pypi.org/project/mother-ml/
@@ -236,12 +340,12 @@ jobs:
           path: dist/
 
       - name: Publish package distributions to PyPI
-        uses: pypa/gh-action-pypi-publish@release/v1
+        uses: pypa/gh-action-pypi-publish@76f52bc884231f62b9a034ebfe128415bbaabdfc # v1.12.4
         with:
           packages-dir: dist/
 
-  build-mkdocs:
-    name: Build MkDocs
+  build-docs:
+    name: Build Docs
     runs-on: ubuntu-latest
     permissions:
       contents: read
@@ -267,17 +371,16 @@ jobs:
           install-args: "--group docs"
 
       - name: Build docs
-        working-directory: mkdocs
-        run: uv run mkdocs build
+        run: uv run zensical build
 
       - name: Upload GitHub Pages artifact
         uses: actions/upload-pages-artifact@v4
         with:
-          path: mkdocs/site/
+          path: site/
 
-  deploy-mkdocs:
+  deploy-docs:
     name: Publish Docs
-    needs: build-mkdocs
+    needs: build-docs
     if: github.event_name != 'pull_request' && github.ref == 'refs/heads/main'
     runs-on: ubuntu-latest
     permissions:

.gitignore

@@ -44,8 +44,8 @@ ENV/
 env.bak/
 venv.bak/
 
-# mkdocs documentation
-mkdocs/site
+# zensical documentation
+site/
 
 # mypy
 .mypy_cache/

.pre-commit-config.yaml

@@ -1,8 +1,8 @@
 fail_fast: false
-exclude: ^(docs/|examples/|tests/)
+exclude: ^(examples/|test/)
 repos:
   - repo: https://github.com/astral-sh/uv-pre-commit
-    rev: "0.11.2"
+    rev: "0.12.5"
     hooks:
       - id: uv-lock
 
@@ -71,7 +71,7 @@ repos:
         entry: uv run poe docs-python-fences
         pass_filenames: false
         language: system
-        files: ^mkdocs/docs/.*\.md$
+        files: ^docs/.*\.md$
   - repo: https://github.com/nbQA-dev/nbQA
     rev: 1.9.1
     hooks:

CLAUDE.md

@@ -20,7 +20,7 @@ Uses `uv` + `poe` (poethepoet). Run tasks with `uv run poe <task>`.
 - **Style (check)**: `uv run poe check-style` = `check-sort-imports` (isort, black profile) + `check-format` (ruff format). Apply with `uv run poe style`.
 - **Static analysis**: `uv run poe check-static-analysis` = `check-lint` (ruff, E+F) + `check-types` (mypy on `src`, strict: `disallow_untyped_defs`).
 - **Lint autofix**: `uv run poe lint`.
-- **Docs**: `uv run poe docs` (build) / `uv run poe serve-docs` (mkdocs serve from `mkdocs/`).
+- **Docs**: `uv run poe docs` (build) / `uv run poe serve-docs` (zensical serve; config `zensical.toml` at repo root, content in `docs/`).
 - **Pre-commit**: `uv run poe install-hook` then `uv run poe check-hook`.
 
 Line length: ruff 120, pylint 100. Note: `CONTRIBUTING.md` references some task names (`poe test`, `test-acceptance`, `check-docs`) that do not exist in `pyproject.toml` — use the verified names above.

README.md

@@ -1,13 +1,11 @@
-# Mother-ML
+# Mother-ML - A ML framework that takes care.
 
-A ML framework that takes care.
-
-Mother is a machine-learning framework for predicting properties from chemical molecules. The major features are:
+Mother is a machine-learning framework for predicting properties from mainly chemical molecules. The major features are:
 
 - 🔬 **SMILES** preprocessing
 - 💾 Generating of **feature vectors** from molecules
 - 📈 Grouping and cross-validation, based on chemical similarity
-- 💻 Model Training: Standard catboost models, and feature selection methods
+- 💻 Model Training: Standard models like catboost, lasso, or random forest, and custom models (e.g.: TabPFN), and feature selection methods
 - 🚴 Training, cross-validation, and hyperparameter optimization of machine-learning models
 - 🌀 Handling Gene expression data from transcriptomics experiments including different normalisation techniques
 - ✨ ~~Explainability analysis with *SHAP*~~ (Currently not supported, will be added in a later release)
@@ -16,7 +14,18 @@ Mother is a machine-learning framework for predicting properties from chemical m
 
 Mother provides methods for each of these steps in the form of sklearn transformer objects. By that, all methods are designed to be easily accessible and usable in a modular way. The methods can be combined to ML workflows with [sklearn pipelines, column transformers, and feature unions](https://scikit-learn.org/dev/modules/compose.html).
 
-All methods can be used as sklearn `transformer` or `estimator`. Combination with other methods, or own methods and models (e.g. using mother preprocessing with other model) is therefore straightforward. To be as compatible as possible, every transformer can be constructed using a dictionary containing the required parameters. However, to provide some convenience to the users, a settings class [MotherSettings](https://github.com/Bayer-Group/MotherML/blob/main/src/mother/settings.py). This class can be used to store all relevant settings for your ML project.
+All methods can be used as sklearn `transformer` or `estimator`. Combination with other methods, or own methods and models (e.g. using mother preprocessing with other model) is therefore straightforward. To be as compatible as possible, every transformer can be constructed using a dictionary containing the required parameters. However, to provide some convenience to the users, a settings class [MotherSettings](https://github.com/Bayer-Group/MotherML/blob/main/src/mother/settings.py) is provided. This class can be used to store all relevant settings for your ML project.
+
+## Why Mother?
+
+Mother keeps the parts of a molecular ML workflow connected without requiring you to leave the scikit-learn ecosystem.
+
+- **Tune an entire workflow from one place.** Mother model wrappers define their own Optuna search spaces, and `PipelineWithHyperparameterRooting`, `ColumnTransformerWithHyperparameterRooting`, and `FeatureUnionWithHyperparameterRooting` collect them across nested steps using familiar `step__parameter` names. You can therefore tune preprocessing, feature selection, and the final estimator together while retaining scikit-learn composition.
+- **Exchange compatible transformers and estimators.** Preprocessing, molecular feature generation, feature selection, cross-validation grouping, and models follow scikit-learn's estimator and transformer interfaces. Mother components work in `Pipeline`, `ColumnTransformer`, and `FeatureUnion`, and can be combined with your own scikit-learn components.
+- **Use consistent, analysis-ready outputs.** Components preserve pandas-friendly tabular data where appropriate. `predict_uncertainty(...)` exposes a common prediction and uncertainty schema across model backends, while `mother_cv(...)` returns fold-level predictions together with evaluation metadata for straightforward comparison and downstream analysis.
+- **Evaluate chemical models more realistically.** Chemistry-aware grouping, including Tanimoto-similarity groups, makes it practical to use group-aware splits that better test generalisation to dissimilar compounds.
+- **Choose models without rewriting the workflow.** The model registry discovers integrated model wrappers and presents a common interface for CatBoost, random forest, lasso, TabPFN when installed, and compatible custom models.
+- **Make workflows reproducible and configurable.** `MotherSettings` keeps input, preprocessing, feature generation, cross-validation, model, and tuning configuration in one validated object that can be loaded from or written to YAML.
 
 ## Usage
 

mkdocs/docs/Changelog.md → docs/Changelog.md

@@ -6,6 +6,34 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 
+## v1.1.3 (2026-08-31)
+
+### Bug Fixes
+
+- Fix use count based fingerprints as default, including TabPFN precision handling and test updates
+  ([#75](https://github.com/Bayer-Group/MotherML/pull/75),
+  [`74cac2ec`](https://github.com/Bayer-Group/MotherML/commit/74cac2ec0f93fa7f436c5fc6f4f5c8f711ae1a8a))
+
+- (tabpfn) Restore upstream TabPFN precision handling with float32 for all operations
+
+
+## v1.1.2 (2026-08-25)
+
+### Bug Fixes
+
+- Solving the `max_features` parameter not being correctly passed to the superclass
+  ([#40](https://github.com/Bayer-Group/MotherML/pull/40),
+  [`4fd6869c`](https://github.com/Bayer-Group/MotherML/commit/4fd6869c3b3218cedf96060bcf4afe84ba70ece5))
+
+
+## v1.1.1 (2026-08-21)
+
+### Chores
+
+- Update changelog for version 1.1.0 with bug fixes and new features
+  ([`8a69fa24`](https://github.com/Bayer-Group/MotherML/commit/8a69fa246b449f257dacd93d0cbbd28b28090532))
+
+
 ## v1.1.0 (2026-08-20)
 
 ### Bug Fixes