{"id":37064269,"url":"https://github.com/adc-trust-ai/trust-free","last_synced_at":"2026-01-14T07:30:38.080Z","repository":{"id":307985842,"uuid":"1030832017","full_name":"adc-trust-ai/trust-free","owner":"adc-trust-ai","description":"An interpretable regression model in Python with Random-Forest-level accuracy","archived":false,"fork":false,"pushed_at":"2025-12-23T17:12:17.000Z","size":69673,"stargazers_count":6,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-12-25T07:21:27.770Z","etag":null,"topics":["ai-safety","compliance","decision-trees","education","explainable-ai","finance","healthcare","interpretable-ml","linear-models","machine-learning","model-trees","python","random-forest","regression-models","research","tree"],"latest_commit_sha":null,"homepage":"https://adc-trust-ai.github.io/trust/","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/adc-trust-ai.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-08-02T12:23:25.000Z","updated_at":"2025-12-23T17:12:21.000Z","dependencies_parsed_at":"2025-08-03T13:27:23.884Z","dependency_job_id":"31d4854a-4af3-4566-a19f-e78175b2888f","html_url":"https://github.com/adc-trust-ai/trust-free","commit_stats":null,"previous_names":["adc-trust-ai/trust-free"],"tags_count":3,"template":false,"template_full_name":null,"purl":"pkg:github/adc-trust-ai/trust-free","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/adc-trust-ai%2Ftrust-free","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/adc-trust-ai%2Ftrust-free/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/adc-trust-ai%2Ftrust-free/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/adc-trust-ai%2Ftrust-free/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/adc-trust-ai","download_url":"https://codeload.github.com/adc-trust-ai/trust-free/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/adc-trust-ai%2Ftrust-free/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28413308,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-14T05:26:33.345Z","status":"ssl_error","status_checked_at":"2026-01-14T05:21:57.251Z","response_time":107,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-safety","compliance","decision-trees","education","explainable-ai","finance","healthcare","interpretable-ml","linear-models","machine-learning","model-trees","python","random-forest","regression-models","research","tree"],"created_at":"2026-01-14T07:30:37.450Z","updated_at":"2026-01-14T07:30:38.067Z","avatar_url":"https://github.com/adc-trust-ai.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# trust-free \u003ca href=\"https://adc-trust-ai.github.io/trust\"\u003e\u003cimg src=\"assets/TRUST_logo_500x500.png\" align=\"right\" height=\"128\" alt=\"TRUST logo\"/\u003e\u003c/a\u003e\n\n[![PyPI version](https://img.shields.io/pypi/v/trust-free.svg)](https://pypi.org/project/trust-free/)\n[![Python](https://img.shields.io/pypi/pyversions/trust-free.svg)](https://pypi.org/project/trust-free/)\n[![Downloads](https://static.pepy.tech/badge/trust-free)](https://pepy.tech/project/trust-free)\n[![License](https://img.shields.io/badge/license-Proprietary-lightgrey.svg)](LICENSE.txt)\n[![User Manual](https://img.shields.io/badge/docs-User_Manual-blue)](https://github.com/adc-trust-ai/trust-free/blob/main/MANUAL.md)\n![OS](https://img.shields.io/badge/OS-Windows%20-blue)\n![OS](https://img.shields.io/badge/OS-macOS%20-blue)\n![OS](https://img.shields.io/badge/OS-Linux%20-blue)\n![Kaggle Compatible](https://img.shields.io/badge/Kaggle-Compatible-blue?logo=kaggle\u0026logoColor=white)\n![Colab Compatible](https://img.shields.io/badge/Google%20Colab-Compatible-blue?logo=googlecolab\u0026logoColor=white)\n\n### Model. Explain. TRUST. All in one package.\n\n**trust-free** is a Python package for fitting interpretable regression models using Transparent, Robust, and Ultra-Sparse Trees (TRUST™) — a new generation of Linear Model Trees (LMTs) with Random-Forest accuracy and intuitive explanations. It is based on my peer-reviewed paper [1], **presented at the 22nd Pacific Rim International Conference on Artificial Intelligence (PRICAI 2025) and to appear in Springer Nature (Lecture Notes in Artificial Intelligence)**.\n\nIt includes a **state-of-the-art explainability suite**, providing comprehensive, automatically-generated explanation reports. To see it in action, here's a 30-second demo showcasing the explain() and compare() methods applied to the famous [Medical Insurance Charges](https://www.kaggle.com/datasets/mirichoi0218/insurance) dataset from Kaggle:\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_explain_compare_wm.gif\" alt=\"ExplainCompareGif\" width=\"100%\" style=\"display: block; margin: auto;\" /\u003e\n\n### Proven Performance: Accuracy + Full Interpretability (60 Datasets)\n\n| Model                   | **Test R² ↑** | **Interpretable?** |\n|-------------------------|---------------|--------------------|\n| **TRUST™**              | **0.67**      | ✅ Yes             |\n| Random Forest (RF)      | 0.62          | ❌ No              |\n| Lasso                   | 0.57          | ✅ Yes             |\n| CART                    | 0.49          | ✅ Yes             |\n| Node Harvest (NH)       | 0.47          | ✅ Yes             |\n| M5' (Linear Model Tree) | 0.36          | ⚠️ Partially       |\n\n\u003e In the table above, **TRUST™ is the only fully interpretable model statistically above 0.6 test R²** across varied benchmark datasets — and **6× sparser** than M5' (*17 vs 109 coefficients* on average).  \n\u003e *Source: PRICAI 2025 (Springer LNAI)*\n\nTry it now: `pip install trust-free`. In **Google Colab**: `%pip install trust-free`.\nSee full benchmarks in the [PRICAI 2025 paper](https://arxiv.org/abs/2506.15791)\n\n---\n\nThe package currently supports standard regression and experimental time-series regression tasks. Future releases will also tackle other tasks such as classification.\n\n## Overview\nTRUST™ [1] is a next-generation algorithm based on (sparse) **Linear Model Trees** (LMTs), which I developed as part of my Ph.D. in Statistics at the [University of Wisconsin-Madison](https://www.wisc.edu/). **trust-free** is the official Python implementation of the algorithm.\n\nLMTs combine the strengths of two popular interpretable machine learning models: Decision Trees (non-parametric) and Linear Models (parametric). Like a standard Decision Tree, they partition data based on simple decision rules. However, the key difference lies in how they evaluate these splits and model the data. Instead of using a simple constant (like the average) to evaluate the goodness of a split, LMTs fit a Linear Model to the data within each node.\n\nThis approach means that the final predictions in the leaves are made by a Linear Model rather than a simple constant approximation. This gives Linear Model Trees both the predictive and explicative power of a linear model, while also retaining the ability of a tree-based algorithm to handle complex, non-linear relationships in the data. This way, LMTs can approximate well any Lp function in Lp norm, i.e. can learn almost any function. Importantly, the resulting fitted model is usually compact, making it easier to interpret.\n\nCompared to existing LMT algorithms such as M5 [2], TRUST™ offers unmatched interpretability while approaching the accuracy of black-box models like Random Forests [3] — a combination that is rare in machine learning.\n\n### References\n\n[1] Dorador, A. (2025). *TRUST: Transparent, Robust and Ultra-Sparse Trees*. [arXiv:2506.15791](https://arxiv.org/abs/2506.15791).\n\n[2] Quinlan, J.R. (1992). *Learning with Continuous Classes*. Australian Joint Conference on AI, 343–348.  \n\n[3] Breiman, L. (2001). *Random Forests*. Machine Learning, 45(1), 5–32.\n\n### Recognition\n\n* **Featured:**\n  * [Data Elixir (Issue 546)](https://news.dataelixir.com/t/t-69C03215CCA6CFF02540EF23F30FEDED) (over 60,000 subscribers)\n  * [Data Science Weekly (Issue 616)](https://datascienceweekly.substack.com/p/data-science-weekly-issue-616) (over 68,500 subscribers)\n  * [University of Wisconsin - Madison Department of Statistics website](https://stat.wisc.edu/2025/05/08/department-of-statistics-celebrates-spring-2025-graduates/) (May 2025)\n\n* **Past Talks \u0026 Workshops:**\n  * [BarcelonaTech, Statistics Department](https://eio.upc.edu/en/seminar) (Dec 2025)\n  * [PRICAI 2025](https://www.pricai.org/2025/index.php) (Nov 2025) \n  * [University of Seville, Minerva AI Lab](https://grupo.us.es/minerva/) (Oct 2025) \n    \n\n## Key Advantages: RF Accuracy ⟡ Tree Transparency ⟡ Linear Interpretability\n\n- **Hybrid power**: Trees to capture non-linearity \u0026 interactions + sparse linear (Relaxed Lasso) models in leaves\n- **Superior accuracy**: RF-level accuracy, proven on 60 benchmark datasets\n- **Full transparency**: Every prediction is auditable via tree path + leaf equation\n- **Inclusive**: Explanation reports written in natural language accessible to all audiences\n- **Compliant by design**: 100% Compliant with the EU AI Act and the OECD AI Principles — ideal for high-stakes domains like finance and healthcare\n\n### About this edition\n- ℹ️ Free-tier dataset limits: ≤ 5,000 rows and ≤ 20 columns (intended for proof-of-concept, R\u0026D and teaching)\n- ✅ All core features are fully functional within these bounds\n- ✅ Unlimited scale and [additional features](https://github.com/adc-trust-ai/trust-free/blob/main/trust-pro.md) in the forthcoming **trust-pro** edition\n\n**Want early access to trust-pro?**  \n- Join the [waitlist](https://forms.gle/Gsti4kZ7yG5ZTNqu7) (completely anonymous \u0026 GDPR-compliant)\n- Star ⭐ this repo to stay updated!\n\n### Features in this edition\n\n- Solves regression tasks (including a currently experimental 'time series mode')\n- Interpretable models with accuracy comparable to Random Forests\n- Visual tree structure and comprehensive, automatically-generated explanations on demand\n- Automatically-generated head-to-head comparisons of profiles of interest\n- Multiple variable importance methods (Ghost, Permutation, ALE plots, SHAP values)\n- Automatic missing value handling that learns from missingness itself\n- Automatic detection of potential overfitting.\n- Ability to efficiently use continuous and categorical predictor variables\n- Prediction confidence intervals *[coming in next release]*\n- Novel method to warn about risky predictions on the fly *[coming in next release]*\n- Novel in-leaf regression model delivering even further sparsity *[coming in next release]*\n- Lightning fast training *[coming in next release]*\n\n\n## What's new in version 2.1.4?\n### TL;DR: First version with **expanded platform compatibility**, plus minor improvements in many areas.\n\n## 2.1.4 (2025-11-16)\n- Added:\n  1. **Expanded compatibility (new platforms will be sequentially added)**\n  2. Axis values in radar chart (compare method).\n  3. Greedy feature order optimization (instead of exhaustive) in radar charts with more than 9 features.\n  4. Pie and radar charts and saved to device in explain and compare method retain feature names when run in Jupyter too.\n  5. Visual cues to convey training performance more easily.\n  6. Automatic detection of potential overfitting.\n- Changed:\n  1. Changed prediction logic from recursive to iterative (more efficient).\n  2. Reversed color scheme for bar chart in detailed mode for the compare method.\n  3. Sorted dumbell plot from largest to smallest feature difference in compare method.\n  4. Fixed bug in explain method for rare cases where no feature was statistically relevant.\n  5. More accurate expected time to training completion after cross-validation.\n  6. Swapped cosine similarity for angular similarity in compare() for more intuitive scaling.\n  7. Other minor enhancements in explain() and compare() methods.\n\nCheck CHANGELOG.md to see all past release notes.\n\nComing up in the next release: **TurboSolve**, a smart OLS solver that is always at least as fast as your favorite OLS solver but usually 2x to 10x faster.\nTurboSolve will serve as the high-performance engine for the TRUST algorithm in `trust-free`. Additionally, it will be available as a *standalone, free utility* for OLS problems of any scale, without any constraints on dataset size.\n### 🚀 Performance Benchmarks: TurboSolve vs. Scikit-Learn\n\nThe following benchmarks compare **TurboSolve** against `sklearn.linear_model.LinearRegression`. \nTests were conducted across 100 repetitions with a range of dataset geometries. \n\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/TurboSolve_benchmarks.png\" alt=\"TurboSolve_vs_Sklearn\" width=\"90%\" style=\"display: block; margin: auto;\" /\u003e\n\u003c/div\u003e\n\n**TurboSolve** is designed for efficiency across all data geometries. As shown above, the performance gap widens significantly as dataset size increases, reaching nearly 10x faster execution than standard implementations for large-scale problems.\n\n| Scenario ($n \\times p$) | TurboSolve (ms) | Sklearn (ms) | Speedup | Mean Rel. Error (%) | Global Max Error (%) |\n| :--- | :--- | :--- | :--- | :--- | :--- |\n| **Tall \u0026 Lean** ($5k \\times 20$) | 0.54 ± 0.20 | 1.54 ± 0.08 | **2.85x** | $2.41 \\times 10^{-12}$ | $4.45 \\times 10^{-11}$ |\n| **Underdetermined** ($50 \\times 200$)* | 0.20 ± 0.02 | 0.76 ± 0.06 | **3.80x** | $3.43 \\times 10^{-4}$ | $0.0129$ |\n| **Big Data** ($100k \\times 100$) | 27.86 ± 1.24 | 273.96 ± 2.68 | **9.83x** | $9.54 \\times 10^{-13}$ | $2.04 \\times 10^{-11}$ |\n\n*\\*Note: In the n \u003c\u003c p case, **TurboSolve** utilizes a data-driven micro-ridge penalty to maintain stability and speed, accounting for the slight increase in relative error.*\n\n## Installation\n\nYou can install this package using pip:\n\n```bash\npip install trust-free\n```\n\u003e 📦 **Note:** The package name on PyPI is `trust-free`, but the module you import in Python is `trust`: `from trust import TRUST`.\n\n### Platform Compatibility\n\n| Platform / Environment   | OS \u0026 Arch         | Python    | Status      |\n|--------------------------|-------------------|-----------|-------------|\n| **macOS ARM64** (M1–M4)  | macOS 11+ ARM64   | 3.11–3.12 | ✅ Working  |\n| **macOS Intel** (x86_64) | macOS 11+ Intel   | 3.11–3.12 | ✅ Working  |\n| **Linux Intel/AMD**      | manylinux x86_64  | 3.11–3.12 | ✅ Working  |\n| **Google Colab**         | Linux x86_64      | 3.12      | ✅ Working  |\n| **Kaggle Notebooks**     | Linux x86_64      | 3.11      | ✅ Working* |\n| **Linux ARM64**          | manylinux ARM64   | 3.11–3.12 | ✅ Working  |\n| **Windows Intel/AMD**    | Windows 11 x86_64 | 3.11–3.12 | ✅ Working  |\n\n*If Kaggle shows a dependency-compatibility issue message upon installation via %pip install trust-free you may safely ignore it and hit \"Restart and run up to selected cell\" (assuming your selected cell is the one installing trust-free).\n\nFor a fully reproducible development environment with all dependencies, see SETUP.md.\n\n## Usage\n\nHere are two basic examples of how to use the TRUST™ algorithm:\n\n```python\nfrom trust import TRUST # note the import name is trust, not trust-free\nfrom sklearn.datasets import make_regression\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import r2_score, mean_squared_error\n```\n\n### 🧪 Example 1: Sparse Synthetic Regression (n=5000, p=20)\n```python\nX, y, coefs = make_regression(n_samples=5000, n_features=20, n_informative=10, coef=True, noise=0.1, random_state=123)\nprint(coefs)\n# x2 = 80.9\n# x3 = 91.4\n# x7 = 64.1\n# x8 = 44.6\n# x10 = 96.2\n# x12 = 90.5\n# x14 = 45.3\n# x17 = 39.8\n# x18 = 90.6\n# x19 = 33.2\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123)\n# Instantiate and fit your model\nmodel = TRUST()\nmodel.fit(X_train, y_train)\n# Predict and print results\ny_pred = model.predict(X_test)\nprint(\"Predictions:\", y_pred[:5])\nprint(\"True y values:\", y_test[:5])\nprint(\"test R\\u00B2:\", r2_score(y_test, y_pred))\n```\n\n```python\n# Obtain (conditional) variable importance by Ghost method (based on Delicado and Pena, 2023)\nmodel.varImp(X_test, y_test, corAnalysis=True, filename=\"Synthetic\")\n# Unconditional variable importance by permutation (with added debiasing and uncertainty quantification steps)\nmodel.varImpPerm(X_test, y_test, R=20, B=20, U=10, filename=\"Synthetic\")\n```\n\u003cdiv align=\"center\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/varImpScores_plot_Synthetic.png\" alt=\"varImp\" width=\"49.88%\" /\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/varImpPermScores_plot_Synthetic.png\" alt=\"varImpPerm\" width=\"47%\" /\u003e\n\u003c/div\u003e\n\n```python\n# Obtain prediction explanation for first observation\nmodel.explain(X_test[0,:], mode=\"detailed\", actual=y_test[0], filename=\"Synthetic\") \n```\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_explain1.png\" alt=\"Explain1\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/Pie_chart_Synthetic.png\" alt=\"PieChart\" width=\"50%\" style=\"display: block; margin: auto;\" /\u003e\n\u003c/div\u003e\n\n### 🩺 Example 2: Diabetes Dataset (n=442, p=10)\n```python\nimport pandas as pd\nfrom sklearn import datasets\nfrom sklearn.preprocessing import LabelEncoder\n\nDiabetes = pd.DataFrame(datasets.load_diabetes().data)\nDiabetes.columns = datasets.load_diabetes().feature_names\ndiab_target = datasets.load_diabetes().target\nDiabetes.insert(len(Diabetes.columns), \"Disease_marker\", diab_target)\nDiabetes_X = Diabetes.iloc[:,:-1]\n# Binary encoding (0/1) for 'sex'\nle = LabelEncoder()\nDiabetes_X.loc[:, 'sex'] = le.fit_transform(Diabetes_X['sex']).astype(str)\nDiabetes_y = Diabetes.iloc[:,-1]\nRLT_Diabetes = TRUST(max_depth=1)\nRLT_Diabetes.fit(Diabetes_X,Diabetes_y)\ny_pred_TRUST = RLT_Diabetes.predict(Diabetes_X)\n```\n```python\n# Tree plotting requires Graphviz to be installed in your system path\n# You can use e.g. Homebrew: brew install graphviz or Conda: conda install -c conda-forge graphviz\nRLT_Diabetes.plot_tree(\"Diabetes\") #will save \"tree_plot_Diabetes.png\" in your working directory\n```\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/tree_plot_Diabetes.png\" alt=\"tree\" width=\"50%\" style=\"display: block; margin: auto;\" /\u003e\n\u003c/div\u003e\n\n```python\n# Obtain variable importance with 2 different methods: Ghost and permutation\nRLT_Diabetes.varImp(Diabetes_X, Diabetes_y, corAnalysis=True, filename=\"Diabetes\") #Ghost method\nRLT_Diabetes.varImpPerm(Diabetes_X, Diabetes_y, filename=\"Diabetes\") #Permutation method\n```\n\u003cdiv align=\"center\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/varImpScores_plot_Diabetes.png\" alt=\"varImp2\" width=\"49%\" /\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/varImpPermScores_plot_Diabetes.png\" alt=\"varImp3\" width=\"48.25%\" /\u003e\n\u003c/div\u003e\n\n\n```python\n# Obtain prediction explanation for second observation\nRLT_Diabetes.explain(Diabetes_X.iloc[1,:], aim=\"decrease\", actual=Diabetes_y[1], filename=\"Diabetes\")\n```\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_explain2.png\" alt=\"Explain2\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_explain3.png\" alt=\"Explain3\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_explain4.png\" alt=\"Explain4\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\u003c/div\u003e\n\n```python\n# Compare the second and fourth observations head-to-head\nRLT_Diabetes.compare(Diabetes_X.iloc[1,:], Diabetes_X.iloc[3,:], filename=\"Diabetes\")\n```\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_compare1.png\" alt=\"Compare1\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/Radar_chart_Diabetes.png\" alt=\"Radar\" width=\"50%\" style=\"display: block; margin: auto;\" /\u003e\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/trust-free_compare2.png\" alt=\"Compare2\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\n\u003cimg src=\"https://raw.githubusercontent.com/adc-trust-ai/trust-free/main/assets/Pie_charts_Diabetes.png\" alt=\"Pies\" width=\"97%\" style=\"display: block; margin: auto;\" /\u003e\n\u003c/div\u003e\n\n### More Examples on Kaggle Datasets\n- [Medical Insurance Charges (1.82M views, 360K downloads)](https://www.kaggle.com/datasets/mirichoi0218/insurance)\n- [Life Satisfaction in the EU (own contribution)](https://www.kaggle.com/datasets/albertdorador/eu-life-satisfaction-eurostat-un-oecd)\n\n\n## License\n\nThis software is provided under a Proprietary - Permissive Binary Only license. See LICENSE.txt for details.\n\n## More Information\n\nFor more details, documentation, and information about the full upcoming 'pro' version of the TRUST™ algorithm, please visit our official website:\n\nhttps://adc-trust-ai.github.io/trust/\n\nFurther details about the TRUST™ algorithm can be found in our preprint on arXiv:\n\nhttps://www.arxiv.org/abs/2506.15791\n\nCopyright © 2025 Albert Dorador Chalar. All rights reserved. TRUST™ is a trademark of Albert Dorador Chalar.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fadc-trust-ai%2Ftrust-free","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fadc-trust-ai%2Ftrust-free","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fadc-trust-ai%2Ftrust-free/lists"}