Source code for the multimodal LLM evaluation pipeline described in "Multimodal Large Language Models for Clinical Scar Assessment: Agreement, Bias, and Reproducibility." Implements a five-stage pipeline (image ingestion, preprocessing, multi-provider API dispatch, JSON schema validation, and Excel/CSV export) used to evaluate five MLLMs (GPT-5.4, Claude Sonnet 5, Gemini 3.1 Pro, Qwen3.7-Plus, GLM-5V-Turbo) against blinded human observer scores on the POSAS Observer Scale. Ships without patient images, API keys, or run outputs; researchers supply their own image dataset and API credentials.
Abdullah, Ahmed
This repository contains the MATLAB/SPM12 analysis scripts that implement the first-level fMRI model and region-of-interest representational similarity analyses used to quantify six-fold grid-cell-like representations and their temporal stability across runs. The repository includes: A first-level GLM in which translation events are grouped into six directional phase regressors. Each regressor combines movement directions separated by 60°, producing phase offsets of 0°, 10°, 20°, 30°, 40°, and 50°. An ROI-based representational similarity analysis that extracts voxelwise beta patterns for the six phase conditions from each run. Calculation of
Segen, Vladislava, Wolbers, Thomas
\This archive contains the analysis code and public input files used for transcriptomic and survival analyses reported in a study of a three-gene prognostic signature in adrenocortical carcinoma. It includes a 77-patient TCGA-ACC discovery workflow with fully nested leave-one-out cross-validation and external evaluation in GEO cohorts GSE10927 and GSE19750. The package contains project-relative Python and R scripts, public input datasets, selected legacy comparison outputs, and run instructions. The authors generated and independently cross-checked the reported results in R (survival/glmnet); MIT applies to code only, while bundled public dat
Marabathula, Ruben Babu, Mulugu, Apurva Nandini, Nakka, Sravani, Shaik, Babulla, Manjari, Kiran, Hamed, Alsufyani, Ammar, AL-Farga, Pinjari, Osman Basha
A minimalist, browser-friendly data science library for modern JavaScript (ESM). Provides statistical analysis, machine learning, multivariate analysis, and visualization tools inspired by scikit-learn and R.
Parent, Essi
RikaiCode is a sophisticated, browser-based tool designed to turn complex codebases into structured, actionable data. It flattens repositories for LLM context, assigns intelligent quality grades (A++ to C), scans for security vulnerabilities, and provides AI-powered architectural insights using GLM-4.7-Flash.
Sahu, Vihaan
A Python-based Gradient Boosted Classifier model for adaptive parallax correction of Geostationary Lightning Mapper (GLM) group-level data is provided as a serialized binary file created using Python's pickle module (Python, 2026), named unified_lightning_gbc_model.p. The model is demonstrated in an included Jupyter notebook file (Beg et al., 2021), named demo_adaptive_parallax_correction.ipynb, written to download and use sample Level-2 GLM data from GOES-East on 17 July 2024 between 2346 and 2347 UTC (GOES-R Algorithm Working Group and GOES-R Series Program Office, 2018a). Supporting GOES-East Advanced Baseline Imager (ABI) Cloud Top Height
Stough, Sarah
Description This repository contains the data and code necessary to reproduce the main results of the study. The paper proposes a unified framework integrating survey-weighted penalized prediction (LASSO), Double/Debiased Machine Learning (DML), Two-Stage Residual Inclusion (2SRI), and design-consistent bootstrap inference. Repository Structure /data ckd_survey_sim.csv simulation_metrics.csv metrics_all_sim.csv stability_sel_prob.csv /code SimulacionGeneral.R Empirical Study.R /docs README.docx Requirements R version 4.3 or higher with the following packages: survey, glmnet, dplyr, tidyr, ggplot2, haven, purrr, gridExtra Reproducibility Guide
Fernando, Rojas
This is the first public release of GOES GLM bolide detection pipeline. This release corresponds to the paper: Jeffrey C. Smith, et al., Finding Fireballs in Lightning: A Daily Pipeline to Find Meteors in Weather Satellite Data, In prep. for JGR: Machine Learning and Computation
Jeff Smith
This release contains the final version of my Insurance Claims Frequency–Severity Modelling project, built using the freMTPL2 French Motor Third Party Liability Insurance dataset (678,013 policies). The pipeline integrates statistical and machine learning methods to model claim frequency and severity. Frequency is estimated using Poisson and Negative Binomial regression, while severity is modelled with Gamma and Lognormal regression. To complement these approaches, Random Forest and XGBoost models are trained and benchmarked against traditional GLMs using metrics such as RMSE, MAE, and R‑squared. The combined predictions yield pure premium es
whitneysdata
This repository contains the complete, annotated R script (Supplementary Code S1) accompanying the manuscript "A Reproducible Simulation Framework for Benchmarking Artificial Intelligence Models in Automated Biodiversity Monitoring Under Imperfect Detection" submitted to Scientific Reports (Special Issue: Automated Biodiversity Monitoring). The script implements a fully reproducible virtual ecologist simulation framework that benchmarks three modelling approaches - Generalized Linear Models (GLM), Random Forest, and Extreme Gradient Boosting (XGBoost) - for species richness estimation under imperfect and heterogeneous detection probability. S
Kakkara, Athira
Initial public release of the code for base modeling, pretraining, ensemble learning, and process-guided loss design for lake water temperature prediction in Lake Mendota. This repository includes: Base model training and evaluation Pretraining on GLM simulation data and finetuning on observations Depth-wise ensemble modeling Process-guided loss with energy conservation constraint GLM simulation scripts Monthly energy balance validation scripts Notes: The repository was developed primarily for Google Colab + Google Drive. The current implementation is intended mainly for comparative evaluation across model settings. Some paths may need to be
Lan, Xin, Luo, Lifeng, Deng, Yue, Tan, Pang-Ning, Zheng, Zihan
This repository contains the analysis code for the manuscript: "Integrated Proteomic and Lipidomic Signatures Link Eating Timing to Cardiovascular Risk in Chinese Children" Published in Nature Communications (2025) Contents: -Cardiovascular data processing scripts -Mixed-effects modeling -Differential expression analysis (DESeq2, limma) -Machine learning feature selection (Random Forest, SVM, XGBoost, Elastic Net) -Mediation analysis -Data visualization Software Requirements: -R 4.3.2 -SAS 9.4 -Required R packages: DESeq2 1.40.0, limma 3.56.0, WGCNA 1.72, randomForest 4.7-1.1, glmnet 4.1-8, e1071 1.7-14, xgboost 1.7.5, mediation 4.5.0 Contact
LQ-doctor
