Write data & AI pipelines in (SQL, Spark, Pandas) and deploy to the cloud, simplified
-
Updated
Mar 31, 2026 - Python
Write data & AI pipelines in (SQL, Spark, Pandas) and deploy to the cloud, simplified
Autolume is a no-coding generative AI system allowing artists to train, craft, and explore their own models.
SmallGBM is a gradient boosting library designed for small datasets (n < 1000). It combines robust leaf weight estimation with stochastic split selection to outperform XGBoost and LightGBM — with lower variance, no hyperparameter tuning, and a native C core.
Official PyTorch implementation for ChimeraMix: Image Classification on Small Datasets via Masked Feature Mixing (IJCAI 2022)
HELO is a lightweight and hybridized cryptographic system. This stands for "Hybrid Encryption Lightweight Optimization". It is made for robust security in IoT devices without reducing its performance. Also, it ensures integrity, authenticity, and confidentiality during the P2P data transmission.
Train a FastGAN on your own photographs, on one GPU, up to 3072×2048.
[OPEN Alloy] The official implement of ATON for high entropy alloy discovery
A reproducible, leakage-proof feature selector for omics classification. Find the right features in n<<p data, not just fewer of them.
Physics-informed knot decoder for short-crack growth curves (synthetic demo, M.Sc. thesis method)
Udacity Small Data course (https://www.udacity.com/course/small-data--cd12528) transfer learning project.
Small-data redoxmer stability audit comparing molecule-level validation with chemistry-platform transfer
Out-of-distribution molecular property prediction from a few analog series: data sets, ensemble models, and MMP-based pseudo-labeling
XGBoost end-to-end on tiny real-world data: forecasts the #5 NASCAR Cup car at Talladega (top-10, DNF, finish) with leak-free features, walk-forward backtesting, built-in SHAP explanations and a Streamlit dashboard.
Solution for Udacity Small Data course project.
Four tabular foundation models vs random forest, HistGradientBoosting and tuned XGBoost on six small UCI datasets, CPU only
pip package for data analysis stability evaluation against small data change.
To associate your repository with the small-data topic, visit your repo's landing page and select "manage topics."