Pandey, Devansh · Narasimhan, Vagheesh M.

1 files · 112 KB · zipdeclared

Analysis code for a five-layer human-genetic study of the lipid-coronary artery disease axis in UK Biobank (GWAS, LD-score genetic correlation, multivariable Mendelian randomization, colocalization, SuSiE fine-mapping, and exome-wide rare-variant burden), with two-sample MR replication against FinnGen. Code only; no UK Biobank or individual-level data are included.

Fonod, Robert

1 files · 810 MB · zipdeclared

A compact, U-Net-inspired 1D fully convolutional neural network (740,551 trainable parameters) that predicts non-apnea sleep arousals from full-length, multi-channel polysomnographic (PSG) recordings at 5-millisecond resolution. Reference implementation, pretrained models, and evaluation code for the paper published in AI (MDPI), 3(1), 164-179.

Gabiccini, Marco

1 files · 87 KB · zipdeclared

First public release. Python/CasADi companion code for the paper 'Handling, Stability and Limit Conditions of Road Vehicles via MAPs and Singularity Analysis' (M. Gabiccini, submitted to Vehicle System Dynamics). Reproduces all figures and results of the paper (2D MAPs, singular lines / grip-limited boundaries, 3D stability MAP) and includes the rigorous double-track vs single-track comparison study, with regression tests pinning all published numbers.

Zinner, Thomas · De Moor, Katrien · Lange, Stanislav · et al.

1 files · 152 MB · zipdeclared

This repository contains the course materials and operational tooling of TTM4200 Computer Networks, an introductory undergraduate networking course at the Norwegian University of Science and Technology (NTNU). The course combines flipped Team-Based Learning (TBL) with reproducible hands-on labs in a recurring weekly cycle of preparation, readiness assurance tests (iRAT/tRAT), team application activities, and lab work. The archive includes: the weekly readiness assurance tests in Teampy format (rats/), the lecture slide decks including the course AI contract (slides/), the lab specifications and interactive notebooks with automated checkpoints (labs/), and the tooling for building the course VM image and per-team lab VMs on OpenStack (vm/). This is the companion artifact to the experience report "Flipped Team-Based Learning and Labs for a Course on Computer Networking" (SIGCOMM Education Workshop, A4NE '26), which describes the course design, assessment model, and lessons learned. The materials are released so that other educators can adopt the weekly cycle in full or incrementally. Code is MIT-licensed; teaching materials are CC BY 4.0. Material adapted from Kurose & Ross, "Computer Networking: A Top-Down Approach", 8th ed. (Pearson), is excluded from both licenses and remains subject to the original reuse terms - see LICENSE, LICENSE-MATERIALS.md, and NOTICE.md in the archive. Version 1.0.1 replaces v1.0.0, which inadvertently contained unlicensed third-party material.

Emmerson (Yaohushuason), Parker

2 files · 479 KB · pdf, zipdeclared

This treatise unifies three documents into one system, rebuilt so that the order of exposition matches the order of foundation. \textbf{Part~I} states the Descent Calculus. The plenum $\Plenum$ is primitive and uncounted. A number is precipitated by constraining the total. This is the mirror of the inherited foundation, which makes the void primitive and assembles infinity last. The axioms force the Sweeping Law: finitude is precipitated by the interference of infinities, and the count is the number of elementary interferences required to reach positive discreteness. The numeral is genealogical: subscripts are descent record, exponents are descent degrees, and ranges are descent spans. The $n$-wave realization reads integers as resonant overtones of a confined plenum; the integer is legible as a nodal-domain tally. Zero is unreachable: where no positive residue forms, the object-language withholds expression rather than installing a null stage. \textbf{Part~II} gives the calculus in its own glyphs. The counting kernel is a pole-bearing interaction whose contour integral returns the enclosed count. For infinite spectra the admissible kernel is not the raw divergent sum but the descent-regularized pole field: this is not a concession to an external analysis, but the analytic face of counting back from infinity. The exponent $m$ is the degree of a cyclic interaction; its deck orbit is the multiplicity of recoverable branches. The summation $\sum_{[l]\leftarrow\infty}$ is tail-subtraction, leaving the counting comb. The phase-space form is the state-counting integral; its heat-smoothed expansion carries the Euler characteristic of the domain inside the count. Every load-bearing identity is stated with the hypotheses that make it true, because in this system hypotheses are not bureaucratic additions: they are the inscriptions of the procedure by which a count forms. \textbf{Part~III} is the ascent traversal. It crosses the same field by inherited mathematics: apartness rosters, the identity-free tally semigroup, counts as trivializations of \'etale witness fields, monodromy obstruction and classification, and positive coherent logic over inhabited stages. It pairs with Parts~I--II where tallies agree, while remaining silent where genealogies live. \textbf{Part~IV} states the general point: notation is the medium, equations are traversals of a field, the fractal morphism acts on notation itself, and every edge is written as an available continuation. \textbf{Part~V} states the greater argument: strict classical counting requires an origin owing nothing to number, identical units, persistent items, observer-independent totals, global assembly of local counts, and completed formal fixity. Each is blocked. What occurs under the name of counting is therefore one of its weakened kin. The Descent Calculus is the theory of what occurs instead.

Hara, Hiromasa · Noguchi, Mitsuhiro · Tanihara, Fuminori · et al.

10 files · 16 GB · zipdeclared

This dataset contains the raw and source data underlying the figures and supplementary figures in the manuscript. GraphPad Prism files (.pzfx) contain the numerical data tables and graph settings used to generate the corresponding figure panels. Raw image data are included in the corresponding compressed files, including files for Fig. 4 double knock-in blastocyst analysis and Fig. 5 FISH analysis. Large image datasets were divided into multiple ZIP files to facilitate upload.

Hanchate, Abhishek · Balhara, Himanshu · Bukkapatnam, Satish

4 files · 637 MB · zipdeclared

This dataset supports the NSF Future Manufacturing Data Challenge and contains multimodal measurements for probabilistic local geometry prediction in single directed energy deposition (DED) laser tracks on SS316L substrates. The dataset includes in-situ thermal image sequences, SEM images of surrounding substrate morphology, and Bruker/Wyko full-field height maps. The common analysis window corresponds to physical part coordinates 20-100 mm. The dataset is intended for developing models that predict local track width, boundary position, contour deviation, edge roughness, or related probabilistic descriptors from process-driven and substrate-driven information.

Buhler, Cassidy Kim · Boettiger, Carl

1 files · 32 KB · zipdeclared

An interactive conservation decision-support tool for California's 30x30 goal — the state's commitment to conserve 30% of its lands and coastal waters by 2030. Built on the Geospatial LLM-Enabled Navigator (GLEN) framework, it pairs a MapLibre interface with a natural-language agent that discovers datasets and runs live SQL queries over cloud-native (Parquet/COG) data through MCP tools. Users can explore the state's conserved-areas inventory alongside biodiversity, habitat, connectivity, freshwater, and climate-hazard data, gaining insight into what is conserved, what lacks protection, and where opportunities remain to fill those gaps. Developed by the Eric and Wendy Schmidt Center for Data Science & Environment (DSE) at UC Berkeley in collaboration with the California Biodiversity Network, aligned to the CA 30x30 Biodiversity Assessment. The tool has been re-platformed onto GLEN — succeeding the earlier Streamlit prototype in this repository — and its data updated to the 2025 CA 30x30 conservation inventory; the repository now holds only dataset configuration and the domain system prompt, with the map, chat, agent loop, and tool orchestration provided by the GLEN framework.

Corilo, Yuri E. · Kew, William R. · McCue, Lee Ann · et al.

1 files · 289 MB · zipdeclared

CoreMS is a comprehensive mass spectrometry framework for software development and data analysis of small molecules.

Soltes, Julian

3 files · 12 MB · csv, pdfdeclared

This paper presents a non-parametric topography of the cosmological landscape, following the principle of maximum entropy to map the relative probability of all isotropic spacetime configurations. The topography is defined as an 11-D probability distribution, derived from a maximally unbiased ensemble of solutions to the Einstein Field Equations (EFE). Uncertainty prevents infinite precision, blurring the ensemble to quantify the relative likelihood of any physical microstate. The result provides a statistical foundation for the manifestation of our universe and its contents, described by probabilities and their respective gradients. This assumes unconstrained numerical coverage of the EFE landscape, mapping mathematically valid solutions that may violate parametric energy conditions. Uncertainty distorts the boundaries between these exotic states and the canonical phase, rationalizing their existence within our universe as quantifiable improbabilities. Resources: The computational implementation and main data ensemble are attached here, as well as maintained at: https://github.com/jgsoltes/Universe-KDE Researchers are encouraged to apply the QUASAR optimizer to their own high-dimensional or non-convex function landscapes. Source code and documentation are maintained at: https://github.com/jgsoltes/hdim-opt

Leray, Sarah · Vicuna, Sebastian · Gómez Sepúlveda, Rodolfo Adrián · et al.

1 files · 1.7 GB · zipdeclared

Curated publication package from the SACBAD master archive (ANID Anillo ATE220055): geospatial products, tabular field-monitoring data, and database documentation for the Petorca, Quilimari, and Ligua basins (central Chile). This deposit is a copy of selected final processed data and documentation from the SACBAD master archive. Raw/source field campaign files are not included; contact project collaborators for access to source data. PACKAGE STRUCTURE • ZENODO_README.txt - package guide • 01_Geospatial_Data/ - cartography, geophysics, hydrography (one ZIP per theme) • 02_Tabular_Data/ - field monitoring by theme (one ZIP per category): geochemistry, stable isotopy, water level/tide (LTC), elevation, tidal data, soil variables, meteorology, geology, eddy covariance (processed only), sapflow, stakeholders • 03_SACBAD_Database_Documentation.zip - user manual, METADATA_SACBAD.xlsx, database report, stakeholder summaries, HTML index COMPANION METADATA (inside 03_SACBAD_Database_Documentation.zip) • METADATA_SACBAD.xlsx • data_files.csv, geospatial_inventory.csv, site_folder_map.csv GEOGRAPHIC SCOPE Petorca, Quilimari, and Ligua river basins, Valparaíso Region, central Chile. RELATED RESOURCES • Reproducible hydroclimate pipeline (software): https://github.com/Centro-de-Cambio-Global-UC/SACBAD • SACBAD project: https://cambioglobal.uc.cl/ • Manuscript (when published): "Assessing the Complex Nature of Climate Change Risks in Semi-Arid Coastal Basins" CITATION Cite this Zenodo deposit and the SACBAD manuscript once published. Third-party hydrometeorological data from DGA and DMC (Chile) remain under their respective terms.

Fonod, Robert

1 files · 59 MB · zipdeclared

Geo-trax (GEO-referenced TRAjectory eXtraction) is a comprehensive pipeline that extracts high-accuracy, georeferenced vehicle trajectories from high-altitude drone imagery. Built for quasi-stationary aerial monitoring of urban traffic, it turns raw bird's-eye view (BEV) drone footage into precise, real-world vehicle trajectories. The framework combines YOLO detection, multi-object tracking, and video stabilization with a robust orthophoto-based georeferencing stage, producing geo-coordinated, lane-resolved trajectories that are spatially and temporally consistent and ready for large-scale traffic analysis and simulation. It is optimized for urban intersections and arterial corridors, where high-fidelity, vehicle-level insights drive intelligent transportation systems and digital twin applications. 📌 Important: If you use this code in your work, kindly acknowledge it by citing the following article: Robert Fonod, Haechan Cho, Hwasoo Yeo, Nikolas Geroliminis (2025). Advanced computer vision for extracting georeferenced vehicle trajectories from drone imagery , Transportation Research Part C: Emerging Technologies, vol. 178, 105205. DOI: 10.1016/j.trc.2025.105205 Why Geo-trax 🛰️ Real-world output : georeferenced, lane-resolved trajectories (WGS84 + local CRS) with per-vehicle speed, acceleration, and estimated dimensions, straight from raw BEV drone video. 🎯 Accurate detection : YOLOv8s vehicle detector reaching 0.951 mAP@50 , trained on more than 19,000 annotated aerial images. 🚗 Flexible tracking : four vehicle classes and six selectable multi-object trackers (BoT-SORT, ByteTrack, OC-SORT, and more). 🌀 Drone-motion robust : homography-based stabilization ( Stabilo ) plus orthophoto image registration for consistent, cross-flight coordinates; both optionally CUDA-accelerated. 📊 Proven at scale : powered the Songdo Traffic dataset (roughly 700,000 trajectories across 20 intersections , fleet of 10 drones ; see Real-World Deployment below). ⚙️ One command, one config : geotrax batch runs the whole pipeline; a single YAML drives every stage, with four tuned presets included. Features Detection : YOLOv8s on aerial BEV imagery; detects car (incl. vans), bus, truck, and motorcycle; optional SAHI sliced inference for improved small-object recall (--sahi). Tracking : six multi-object trackers (BoT-SORT default); see Tracking for a comparison; optional per-track frame-gap interpolation. Stabilization : homography-based trajectory correction via Stabilo 🌀, tuned with Stabilo-Optimize 🎯; optional CUDA acceleration (--stab-gpu). Georeferencing : frame-to-orthophoto registration; outputs lat/lon, local CRS, speed, acceleration, and lane assignment per vehicle; optional CUDA acceleration (--geo-gpu). Visualization : track overlays on original, stabilized, or static-reference video, in five rendering modes (incl. oriented bounding boxes). Analysis : trajectory maps, kinematic distributions, and class/dimension charts, per-video or aggregated across drones and sessions. Scaling & tooling : batch-processes directory trees and aggregates multi-drone data; includes standalone utilities for end-to-end data preparation, training, evaluation, and validation. 🚀 Planned Enhancements Comprehensive documentation in a dedicated docs/ folder. A tools/README.md index already covers the auxiliary scripts. Modularized, OOP-based pipeline with custom reference frame support and georeferencing leveraging Stabilo's image-matching backend. Per-class confidence thresholds. Batch inference and multi-thread processing. Real-world map visualization (e.g., MovingPandas, contextily) and interactive web app. 🔗 Related Projects Geo-trax integrates with and complements several specialized tools: Stabilo 🌀 : Python library for video and trajectory stabilization using robust homography transformations. Supports various feature detectors, RANSAC algorithms, and user-defined masks. Used as Geo-trax's core stabilization engine. Stabilo-Optimize 🎯 : benchmarking and hyperparameter optimization framework for Stabilo. Evaluates stabilization performance through ground truth-free assessment using random perturbations. Used to fine-tune Geo-trax stabilization parameters. HBB2OBB 📦 : converts horizontal bounding boxes to oriented bounding boxes using SAM segmentation models. Can enhance Geo-trax outputs when object orientation is needed for downstream analysis. Real-World Deployment: The Songdo Experiment Geo-trax was validated in a large-scale urban traffic monitoring campaign in Songdo, South Korea, where it processed footage from a fleet of 10 drones to produce the Songdo Traffic dataset. The detection model was trained on the companion Songdo Vision dataset. Both are described in the associated publication. Songdo campaign 📍 Location Songdo International Business District, South Korea 📅 Duration 4 days (October 4 to 7, 2022) 🚁 Fleet 10 drones (DJI Mavic 3), 140 to 150 m altitude, 4K at 29.97 fps 🔭 Coverage 20 busy intersections 🚗 Result ~700,000 georeferenced vehicle trajectories 🎥 Demo of Geo-trax applied to the Songdo experiment: https://youtu.be/gOGivL9FFLk Citation If you use Geo-trax in your research or software, please cite: Journal article (preferred for any use of the framework): @article{fonod2025advanced, title = {Advanced computer vision for extracting georeferenced vehicle trajectories from drone imagery}, author = {Fonod, Robert and Cho, Haechan and Yeo, Hwasoo and Geroliminis, Nikolas}, journal = {Transportation Research Part C: Emerging Technologies}, volume = {178}, pages = {105205}, year = {2025}, publisher = {Elsevier}, doi = {10.1016/j.trc.2025.105205}, url = {https://doi.org/10.1016/j.trc.2025.105205} } Software archive (when referencing or building on the code itself): @software{fonod2026geo-trax, author = {Fonod, Robert}, title = {Geo-trax: A Comprehensive Framework for Georeferenced Vehicle Trajectory Extraction from Drone Imagery}, year = {2026}, month = jul, version = {1.3.0}, doi = {10.5281/zenodo.12119542}, url = {https://github.com/rfonod/geo-trax}, license = {MIT} } Contributions Early code received key contributions from Haechan Cho (georeferencing) and Sohyeong Kim (video/flight-log merging). Community contributions are welcome: open a GitHub Issue or submit a pull request. License This project is distributed under the MIT License. See the LICENSE file for more details. Full Changelog https://github.com/rfonod/geo-trax/compare/v1.2.0...v1.3.0

Reeve, Richard

1 files · 37 KB · zipdeclared

Research Software metadata crosswalk between Project.toml, codemeta.json, .zenodo.json and LICENSE to provide consistent metadata

Zapater, Marina

1 files · 1.9 MB · zipdeclared

This archive contains the source code and supporting implementation material associated with the developments of the REDS Team at HEIG-VD/HES-SO within the framework of the IGNITE project (funded by HES-SO). The work presents an FPGA-based processing chain for real-time Online Recursive Independent Component Analysis, ORICA, targeting high-density electroencephalography, hdEEG, with 130 channels sampled at 1 kHz and scalability considerations toward 256 channels. The implementation targets a heterogeneous Xilinx Zynq UltraScale+ System-on-Chip platform, where the Processing System performs data acquisition, buffering, and whitening, while the Programmable Logic accelerates the recursive ICA adaptation stages. The design follows a multi-rate execution model that separates sample-level streaming operations from window-level matrix adaptation, enabling real-time processing within a 1-second observation window. The released code includes the material needed to reproduce and extend the main implementation and validation flow, including fixed-point ORICA simulation, hardware-oriented validation scripts, FPGA accelerator components, and system-integration support for PS/PL communication. The artifact is intended to support reproducibility, further research on embedded blind source separation, and future development of low-latency edge-processing pipelines for high-density biomedical signals. The corresponding paper demonstrates that the proposed implementation achieves real-time execution for 130-channel hdEEG, with an end-of-window ORICA latency of 1.85 ms, a total processing pipeline of approximately 711 ms including whitening, and an estimated system power consumption of 3.86 W on the target Zynq UltraScale+ platform. The fixed-point implementation is validated against MATLAB EEGLAB and a float64 ORICA reference, showing statistically comparable source separation quality. All results have been submitted for publication in a journal, which is currently under review.

Grisa, César Augusto

1 files · 9.8 MB · zipdeclared

Candidate proof synthesis of an explicit-threshold strengthening of the generic branch of Lengyel's Conjecture 4 (INTEGERS 21 (2021), #A86) on 2-adic valuations of central Delannoy number differences: v2(D(u*2^(n+1)+b) - D(u*2^n+b)) = n + 2*v2(b(b+1)) for every b>=1, odd positive u, and every n >= max{2, v2(b(b+1))} (Lengyel's conjecture asserts the generic law for n sufficiently large). The archive assembles four separately reproducible modules (ramified Mahler-Taylor bound M3; high-scale signed increment M4; four-chart derivative-residue transport M5a; finite transition certificate M5b with 1,047,440 exhaustively checked residue classes) together with exact reproducers, outputs, SHA-256 manifests, and a complete audit-resolution ledger. v5.2 answered an external GAP (not FAIL) audit; v5.3 and v5.4 add archive-integrity, attribution, and citation repairs from two further independent audits, with no mathematical change. This is a candidate proof, not a referee-validated theorem; see PROOF_STATUS.md.

Hoffmann, Daniel

2 files · 6.3 MB · pdf, zipdeclared

An introductory course on quantitative concepts in biology for undergraduates in biology and natural sciences.

Abrams, Cameron F.

1 files · 112 MB · zipdeclared

Pestifer is a Python package for automating the preparation of biomolecular simulation systems for NAMD using the CHARMM force field. It drives VMD/psfgen to build PSF/PDB topology and coordinate files from PDB or mmCIF structures, handling missing loops, glycosylation, disulfide bonds, mutations, membrane embedding, and solvation.

Fritsch, Miriam · Pflüger, Stefan · de Boer, Remco E. · et al.

1 files · 343 KB · zipdeclared

QRules is a Python package for **validating and generating particle reactions** using quantum number conservation rules. The user only has to provide a certain set of boundary conditions (initial and final state, allowed interaction types, expected decay topologies, etc.). QRules will then span the space of allowed quantum numbers over all allowed decay topologies and particle instances that correspond with the sets of allowed quantum numbers it has found. The resulting state transition objects are particularly useful for **amplitude analysis / Partial Wave Analysis** as they contain all information (such as expected masses, widths, and spin projections) that is needed to formulate an amplitude model.

Samy, M. · Lazem, Shaimaa · Gabr, M. M. · et al.

1 files · 3.2 MB · zipdeclared

Initial public release of the reproducibility package for the manuscript: "Fairness Auditing in Public Tabular Benchmarks: A Multi-Dataset Study of Bias and Mitigation". This release includes processed result tables, post-processing code, generated figures, robustness analyses, dataset provenance, fingerprints, and documentation. Raw public benchmark datasets are not redistributed.