PyEarthTools: is a Python framework that supports the develoment of machine learning models, big and small, for Earth system science is suitable for students and newcomers, as well as for domain specialists and scientists runs effectively on HPC (supercomputers), cloud, workstations and laptops is suitable for use with megabytes to petabytes of data Source Code: github.com/ACCESS-Community-Hub/PyEarthTools Documentation: pyearthtools.readthedocs.io Tutorial Gallery: available here New Users Guide: available here If you use PyEarthTools for your work or a publication, please cite our work . What's Changed Add SimpleModel and expand tutorials b
Leeuwenburg, Tennessee, Cook, Harrison, Rio, Maxime, Hobeichi, Sanaa, Yang, Edward, Miller, Joel, Mason, Gemma, Ramanathan, Nikeeth, Pill, John, Haddad, Stephen, Stassen, Christian, de Burgh-Day, Catherine, Holmes, Ryan, Potokina, Margarita, Bogacheva, Jenya, James, Matthew, Sullivan, Ben, Hoffmann, Luke, Pegios, Michael
Pipeline, prompts versionados, padrão-ouro, anotação dupla às cegas e classificação por LLM das menções a inteligência artificial nos Formulários de Referência entregues à CVM, 2023–2026. Material suplementar S2 em SUPLEMENTAR.md, incluindo a comparação com um modelo aberto (Gemma 4 via Ollama), reaplicável num computador pessoal. Licenças: código sob MIT; dados em dados_publicos/ (derivados dos dados abertos da CVM) sob Open Database License (ODbL 1.0) — ver dados_publicos/LICENCA.md. English. Pipeline, versioned prompts, gold standard, blind double annotation and LLM classification of mentions of artificial intelligence in the Reference For
Vinhas, Felipe Domingues Vieira
V3-gemma — Multimodal Feature Extraction and ML Analysis Pipeline Summary This release provides a reproducible end-to-end pipeline for extracting multimodal clinical features (I1–I20) with Gemma 3n and analyzing depression-related signals with machine learning. The repository is organized into four packages that follow the analysis workflow in order. Packages included Multimodal_feature_extraction Extracts clinical features I1–I20 from video (frames + audio) and text using Gemma 3n, with threshold post-processing and paper-ready wide CSV outputs. In-house_machine_learning_toolbox Interactive Streamlit-based ML analysis engine (stratified K-fo
전인수
Code and experiment data for the empirical validation of the Behavioral Credibility Trilemma via Best-of-N selection. Accompanies the manuscript (arXiv:2605.25739), prepared for submission to the Journal of Machine Learning Research (JMLR). v1.3 completes the public artifact set: per-model held-out calibration CSVs underlying the binding sets, the pre-specified cross-model replication (Gemma-2-9B, Yi-9B; all five hypotheses replicate 5/5 under per-model binding sets), the comonotone-selector robustness re-analysis, and the uniform-random and no-oracle-proxy control suites (plan amendment E-C).
Lovén, Lauri, Do, Nam, Mehmood, Hassan, Sah, Dinesh Kumar, Tarkoma, Sasu
An experimental five-layer cognitive architecture that wraps a local LLM (via Ollama) in persistent memory, self-monitoring, background reflection, and a continuously updated self-model. Developed and tested on Google's Gemma 4 26B for best results, but model-agnostic — any Ollama-compatible model works. The base model is unmodified — the architecture adds continuity, error correction, and self-regulation through external processing layers. Runs entirely offline on consumer hardware. Includes full source code (Python), web interface, and example configuration.
Schots, Ilja
Highlights Cleaned all sections and included new Apptainer guide for HPC compute containerization and LLM introduction and hyperparameter tuning with Gemma 7b (Google) model. Sections and Subsections Deep Learning and Deep Reinforcement Learning Tutorials (Libraries: Python, PyTorch, Gym, NumPy, Matplotlib and more) Introduction Course Progression Practical Deep Learning with PyTorch Matrices Gradients Linear Regression Logistic Regression Feedforward Neural Network (FNN) Convolutional Neural Network (CNN) Recurrent Neural Network (RNN) Long Short-Term Memory Network (LSTM) Autoencoders (AE) Fully Connected Overcomplete Autoencoders Improving
Ritchie Ng
