Experiment-Predictium v2.3.0-VM-RR: A No-Code Scientific Machine Learning Platform with MOF-Predictium, High-Throughput Screening, and Inter-Laboratory Reproducibility Tools
Experiment-Predictium v2.3.0-VM-RR Release date: 19 October 2026 Overview This release consolidates the MOF-specific development carried out during the 2026 Virtual Mobility activity "MOF-Predictium: A No-Code Machine Learning Platform for MOF Property Prediction, Benchmarking, and High-Throughput Screening Using CoRE MOF DB" within COST Action CA22147 - EU4MOFs. Main additions MOF-Predictium as an eighth model option, combining 1D convolution, multi-head self-attention, and gradient-boosted residual correction. Benchmarking with 80/20 holdout evaluation, 5-fold cross-validation, RMSE, MAE, MAPE, and R2. Advanced MOF VM Tools for multi-target
Resource information
- Format
- zip
- Original repository
- 23548460
- DOI
- 10.5281/zenodo.23548460
- Credit basis
- Creators in Zenodo record metadata
- Access
- Open files available from the original Zenodo deposit
- Last checked
- Nov 5, 2026
- License
- mit-license
- Last record update
- Nov 5, 2026
About this resource
Experiment-Predictium v2.3.0-VM-RR Release date: 19 October 2026 Overview This release consolidates the MOF-specific development carried out during the 2026 Virtual Mobility activity "MOF-Predictium: A No-Code Machine Learning Platform for MOF Property Prediction, Benchmarking, and High-Throughput Screening Using CoRE MOF DB" within COST Action CA22147 - EU4MOFs. Main additions MOF-Predictium as an eighth model option, combining 1D convolution, multi-head self-attention, and gradient-boosted residual correction. Benchmarking with 80/20 holdout evaluation, 5-fold cross-validation, RMSE, MAE, MAPE, and R2. Advanced MOF VM Tools for multi-target analysis, Isolation Forest anomaly screening, data augmentation, ensemble uncertainty, learning curves, OOD assessment, and chemistry-aware stratified evaluation. High-throughput candidate screening, PCA/UMAP property-space visualization, Pareto-front analysis, and inverse-design support. MOF-NP Round-Robin and Inter-Laboratory Reproducibility module for common templates, automated multi-laboratory data merging, provenance tracking, matched-condition analysis, within-lab and between-lab variability, ICC(2,1), protocol metadata review, and discordance flags. Updated documentation, tutorial data, reproducibility guidance, screenshots, and COST/EU4MOFs acknowledgement. Scientific scope The round-robin functionality is a research-software extension inspired by scientific discussions within EU4MOFs. It is not an officially adopted EU4MOFs toxicology protocol, and this release does not claim that a future experimental round-robin study has already been performed. Scientific studies using CoRE MOF or other external datasets should document the exact dataset release, accessible structures, property-label availability, curation, provenance, filtering rules, and split design used for the reported analysis. Funding acknowledgement The MOF-specific development of Experiment-Predictium was supported through a Virtual Mobility Grant within COST Action CA22147 - EU4MOFs (grant reference E-COST-GRANT-CA22147-6e4f1131). COST is supported by the European Union. Installation See README.md for source installation. A Windows release asset contains ExperimentPredictorApp.exe and the required application files. The current Windows build still requires a compatible .NET 9 Desktop Runtime and a Python installation with packages from python_scripts/requirements.txt . Archiving and citation Release DOI: https://doi.org/10.5281/zenodo.23462328 Concept DOI (all versions): https://doi.org/10.5281/zenodo.18215803 Repository: https://github.com/abscienceai/Experiment-Predictium