Our Work

Collaborated as a team of four to develop and manage a project designed to enhance our collaboration and project management skills. Together, we conceptualized and built a solution that effectively ad…

Collaborated as a team of four to develop and manage a project designed to enhance our collaboration and project management skills. Together, we conceptualized and built a solution that effectively addressed stakeholder objectives while strategically balancing resource constraints. This project encompassed the full project lifecycle, including planning, execution, and delivery, showcasing our ability to manage competing priorities and achieve results.

v2 focuses on a simpler fork setup, clearer customization guidance, and a reliable automated release path. 🚀 Features / 新功能 feat(ui): refine API docs utilities (#58) by @ZL-Asica feat(config): improv…

v2 focuses on a simpler fork setup, clearer customization guidance, and a reliable automated release path. 🚀 Features / 新功能 feat(ui): refine API docs utilities (#58) by @ZL-Asica feat(config): improve fork customization defaults (#59) by @ZL-Asica feat(i18n): add four locale support (#72) by @ZL-Asica feat(ui): clarify avatar generation and improve mobile navigation (#73) by @ZL-Asica 🛠️ Improvements / 改进 refactor(ui): refine avatar generator design (#60) by @ZL-Asica perf(avatar): cache transformed responses (#71) by @ZL-Asica 📖 Documentation / 文档更新 docs: reorganize contributor guides (#61) by @ZL-Asica docs: simplify customization and harden v2 release flow (#74) by @ZL-Asica 📦 Dependencies / 依赖更新 chore(deps): update dependencies (#70) by @ZL-Asica

This dataset compiles hourly atmospheric, geospatial, and model-derived information for the integrated characterization of solar and wind energy resources in Mexico. The meteorological time series wer…

This dataset compiles hourly atmospheric, geospatial, and model-derived information for the integrated characterization of solar and wind energy resources in Mexico. The meteorological time series were obtained from the MERRA-2 atmospheric reanalysis via the NASA POWER API for ten selected study zones. The temporal coverage spans from January 1, 2021, to December 31, 2025, with 43,824 hourly timestamps for each zone. Variables include Global Horizontal Irradiance (GHI), Direct Normal Irradiance (DNI), Diffuse Horizontal Irradiance (DHI), ambient temperature, surface albedo, wind speed, wind direction, and atmospheric pressure. The dataset is structured to facilitate the reuse of both original time series and derived products. The repository includes raw and processed CSV files, geospatial layers (GeoTIFF, GeoJSON), Python scripts used in Google Colab, graphical outputs, and site-specific datasets utilized for machine learning (XGBoost) modeling. This data can support comparative resource assessments, temporal complementarity studies, preliminary site selection, and hybrid renewable energy systems modeling.

An Agent Skill that makes Word (.docx) documents with Thai or mixed Thai–English text render correctly: no spelling squiggles under every Thai word, Thai lines that wrap inside words, working bullets …

An Agent Skill that makes Word (.docx) documents with Thai or mixed Thai–English text render correctly: no spelling squiggles under every Thai word, Thai lines that wrap inside words, working bullets and bold. The agent writes Markdown; one bundled command builds the document and checks it without changing a character of the content. Two zero-dependency implementations, Python and JavaScript, produce byte-identical files; a checker reports the five causes of broken Thai in any .docx, and a repair command clears them in a new file without changing its text.

Python code that reproduces the figures of the article "Why the aperiodic slope is not a simple proxy for excitation–inhibition balance" (Sheheitli, Johnson & Vitek) 2026. The script is a di…

Python code that reproduces the figures of the article "Why the aperiodic slope is not a simple proxy for excitation–inhibition balance" (Sheheitli, Johnson & Vitek) 2026. The script is a direct numerical evaluation of the equations in the main text. It does not load any input files. Requirements: Python 3 with NumPy and Matplotlib. Run python generate_figures.py to write the figures to an output directory.

No description provided.

Version 1.1.0. Adds result files and scripts for the exact randomisation, the tariff-exposure and announcement-date specifications, the exact decomposition of the sustainability profile shift, and the…

Version 1.1.0. Adds result files and scripts for the exact randomisation, the tariff-exposure and announcement-date specifications, the exact decomposition of the sustainability profile shift, and the allocation-rule dispersion and information checks. The claim registry now holds 324 claims.Data, derived series and analysis code for a study of United States imports of 32 materials underpinning quantum technologies, reconstructed by partner and month at ten-digit tariff resolution from January 2019 to June 2026, across China's export controls of 2023 to 2025. Contents. data/ the reconstructed panels and the tables any result depends on. results/ the derived series, one file per analysis step. code/ the scripts that build the panels and produce every result, with code/paper/ the checks that rebuild the supplementary tables.  Verification. code/claim_registry.py holds every numeric statement in the article with the sample it is computed on, the rule that selected that sample, its metric and its inference, and recomputes each from the files here:  cd code && python claim_registry.py It reports 324 of 324 claims verified and 41 of 41 declared p-values recomputed. MANIFEST.csv lists every staged file with its size and SHA-256. Sources. United States Census Bureau international trade API and United Nations Comtrade for trade; World Bank World Development Indicators and Worldwide Governance Indicators, Ember and Our World in Data for country indicators. All publicly accessible. Credentials and raw API response caches are excluded; the fetch scripts and their request logs are included.

This archive accompanies the manuscript "Interaction-Aware QUBO Post-Pruning Converts Gradient-Boosted Trees into Low-Redundancy Predictive Signatures." It contains the original Python analysis script…

This archive accompanies the manuscript "Interaction-Aware QUBO Post-Pruning Converts Gradient-Boosted Trees into Low-Redundancy Predictive Signatures." It contains the original Python analysis scripts, processed result tables, small public benchmark data caches, manuscript source, and figure files for the benchmark and biological case studies (PBMC, DepMap, and Perturb-seq), including representative D-Wave quantum-annealing spot checks. The README, RUN_REPRODUCTION.md, and data_sources.md describe inspection of archived results and the steps and public sources needed to rerun analyses. Large third-party raw datasets and private API credentials are intentionally excluded. The original code is licensed under MIT; original processed result tables under CC0; original manuscript text and figures under CC BY 4.0. Third-party source datasets retain their own terms. The corresponding license and data notices are included in the archive.