fucking-awesome-datascience
📝 An awesome Data Science repository to learn and apply for real world problems. With repository stars⭐ and forks🍴
https://github.com/correia-jpv/fucking-awesome-datascience
Last synced: 11 days ago
JSON representation
Agents
Frameworks
- ADK-Rust - Production-ready AI agent development kit for Rust with model-agnostic design (Gemini, OpenAI, Anthropic), multiple agent types (LLM, Graph, Workflow), MCP support, and built-in telemetry.
- Lumen - Agent framework for chatting with data, turning natural language into SQL, transformation pipelines and visualizations. Outputs are declarative specs that can be inspected, edited, reopened in a notebook or composed into a dashboard.
Research & Knowledge Retrieval
- Chunk Tuner - Open-source Python library and MCP server to benchmark document chunking strategies for RAG, score retrieval quality, and recommend configurations for a corpus.
- II-Commons - Daily-updated skill and CLI for deterministic retrieval across arXiv, PubMed/PMC, and supported US policy corpora.
Tools
- Frostbyte MCP - MCP server providing 13 data tools for AI agents: real-time crypto prices, IP geolocation, DNS lookups, web scraping to markdown, code execution, and screenshots. One API key for 40+ services.
- DeepAlpha - AI crypto trading framework using LightGBM + XGBoost ensemble with 72 ML features. 70.9% walk-forward validated accuracy on out-of-sample data. Supports Bybit and Binance. MIT licensed, available on 🌎 PyPI.
- CAJAL - Local AI agent for generating publication-ready scientific papers with real arXiv citations, IMRaD structure, and tribunal scoring. Runs 100% offline via Ollama with 4B-9B models. MIT licensed. 🌎 HuggingFace
- ai-evaluation - Open-source LLM and agent evaluation framework with 50+ metrics, LLM-as-Judge augmentation, and guardrail scanners (jailbreak, PII, prompt-injection). Useful for scoring RAG outputs, agent trajectories, and function-calling behavior in data-science workflows.
- Kitaru - Open-source platform that records real AI agent runs, replays them against changes, and evaluates outcomes before deployment.
Fun
Datasets
- Open Data Sources
- Open data Index
- Public Git Archive
- NAYN.CO Turkish News with categories
- Covid-19
- Covid-19 Google
- 5000 Images of Clothes
- AI Displacement Tracker - Structured dataset tracking 92 AI-attributed workforce reduction events affecting 453,748 workers across 12 countries and 11 sectors. JSON and CSV formats. CC-BY-4.0 licensed.
- FirstData - The world's most comprehensive authoritative data source knowledge base. 210+ curated sources from governments, international organizations, and research institutions. MCP integration for AI agents. MIT licensed.
- latamdata-py - Python package for one-line access to 38 open research datasets from Latin America (health, neuroscience, mental health, economics). pip install latamdata-py.
- ZipCheckup - Free ZIP-level environmental safety data for 42,000+ US ZIP codes: water quality, air quality, PFAS contamination, radon, lead, flood risk, and 11 more verticals. Public REST API, npm/PyPI packages, CC BY 4.0.
- Verified Supplement Evidence - Evidence-graded dietary-supplement dataset covering dosing, bioavailability by form, drug-nutrient interactions, NHANES deficiency prevalence, FDA FAERS adverse-event signals, and cost-per-effective-dose, with every clinical claim citing a PubMed PMID. CC BY 4.0, DOI 10.57967/hf/9356.
- College ROI Dataset - Lifetime return-on-investment estimates for 29,700 US bachelor's programs across 3,392 colleges, built from FREOPP, IPEDS, and BEA regional price data. 5 CSVs with data dictionary, CC BY 4.0, Zenodo DOI.
- BDE Score - AI-powered multi-market stock analysis with transparent BDE scoring across 73 stocks (US/HK/A-share). EU AI Act Art.50 compliant. MIT license.
- Pokemon Card Centering Measurements - 320 measured PSA-style centering annotations (left/right and top/bottom border percentages, tilt) across 302 real eBay-listed Pokemon cards. CSV, CC BY 4.0, Zenodo DOI.
- Pokemon Card Sold-Price Reference by Grade - Median sold price by grade (raw, PSA 9, PSA 10) for 486 Pokemon cards, with sample size and confidence flag per card. CSV, CC BY 4.0, Zenodo DOI.
- HDD Price Index - Daily open dataset of the cheapest new internal 3.5" SATA hard-drive price per terabyte (USD/TB) by capacity tier on Amazon US, with a historical time series. CSV, JSON and JSONL, no login, CC BY 4.0.
- WhatFontIs-Bench - Synthetic benchmark for font family identification with 11,995 images of words set in 600 known fonts, annotated with word and per-letter boxes.
Literature and Media
Bloggers
- Greg Reda - Greg Reda Personal Blog
- Drew Conway - Personal Web Page
- Noah Iliinsky - Personal Blog
- Clare Corthell - The Open Source Data Science Masters
- Emilio Ferrara's web page
Books
- Mining Massive Datasets - free e-book comprehended by an online course
- Advances in Genetic Programming, Vol. 3 - Free Download
Podcasts
Presentations
Other Awesome Lists
Comics
- awesome-awesomeness
- Awesome Machine Learning
- lists
- awesome-dataviz
- awesome-python
- Data Science IPython Notebooks.
- awesome-r
- awesome-datasets
- awesome-Machine Learning & Deep Learning Tutorials
- Awesome Data Science Ideas
- Machine Learning for Software Engineers
- Awesome Machine Learning On Source Code
- Awesome Community Detection
- Awesome Graph Classification
- Awesome Decision Tree Papers
- Awesome Fraud Detection Papers
- Awesome Gradient Boosting Papers
- Awesome Computer Vision Models
- Awesome Monte Carlo Tree Search
- 100 NLP Papers
- Awesome Game Datasets
- Data Science Interviews Questions
- Awesome Explainable Graph Reasoning
- Awesome Drug Synergy, Interaction and Polypharmacy Prediction
- Data Science Projects
- Awesome Data Analysis - A curated list of data analysis tools, libraries and resources.
- Awesome Evidence Synthesis - A curated list of open-source tools for systematic reviews, meta-analysis, and evidence synthesis.
- ML/AI Interview Prep - 500+ ML/AI interview Q&A with runnable code — covers ML fundamentals, deep learning, NLP, PyTorch, scikit-learn pipelines, and system design
Hobby
Socialize
GitHub Groups
Source
Hobby
Sponsors
Table of Contents
- TabGAN - Synthetic tabular data generation using GANs, Diffusion Models, and LLMs with adversarial filtering and privacy metrics.
The Data Science Toolbox
Comparison
- datacompy - DataComPy is a package to compare two Pandas DataFrames.
- Boosting
- Temporal difference learning
- Regression
- Linear Regression
- Ordinary Least Squares
- Logistic Regression
- Stepwise Regression
- Multivariate Adaptive Regression Splines
- Softmax Regression
- Locally Estimated Scatterplot Smoothing
- KNN (K-Nearest Neighbors)
- SVM (Support Vector Machine)
- Decision Trees
- ID3 algorithm
- C4.5
- Ensemble Learning
- Bagging
- Random Forest
- AdaBoost
- Density-based clustering
- Fuzzy clustering
- Mixture models
- Dimension Reduction
- Latent Dirichlet Allocation (LDA)
- Neural Networks
- Self-Organized Maps
- Adaptive resonance theory
- Hidden Markov Models (HMM)
- Heuristic approaches
- Q Learning
- SARSA (State-Action-Reward-State-Action) algorithm
- k-Means
- Apriori
- EM (Expectation-Maximization)
- PageRank
- Naive Bayes
- CART (Classification and Regression Trees)
- Multilayer Perceptron
- Convolutional Neural Network (CNN)
- Recurrent Neural Network (RNN)
- Boltzmann Machines
- Autoencoder
- Generative Adversarial Network (GAN)
- Transformer
- ML System Designs)
Deep Learning Packages
- PyTorch
- torchvision
- torchtext
- torchaudio
- ignite
- PyToune
- skorch
- PyVarInf
- pytorch_geometric
- GPyTorch
- pyro
- Catalyst
- pytorch_tabular
- Yolov3
- Yolov5
- Yolov8
- TensorFlow
- TensorLayer
- TFLearn
- Sonnet
- tensorpack
- TRFL
- Polyaxon
- NeuPy
- tfdeploy
- TensorFlow Fold
- tensorlm
- Mesh TensorFlow
- Ludwig
- TF-Agents
- TensorForce
- keras-contrib
- Hyperas
- Elephas
- Hera
- Spektral
- qkeras
- keras-rl
- Talos
- altair
- amcharts
- anychart
- bokeh
- Comet
- slemma
- cartodb
- Cube
- d3plus
- Data-Driven Documents(D3js)
- dygraphs
- exhibit
- gephi
- ggplot2
- tensorflow-upstream
- Glue
- Netron
- Resseract Lite
- vizzu
- Wrangler
- r2d3
- TensorWatch
- TensorLight
- PyTorchNet
- pytorch_tabular
- torchvista - Interactive notebook-based tool to visualize the forward pass of any PyTorch model.
- FlexViz - Python library for interactive, cross-filtered dashboards that stay responsive on 100M+ rows by aggregating with Polars on the server.
- TorchDR - GPU and multi-GPU dimensionality reduction with a scikit-learn-compatible API.
- OpenLanguageModel - PyTorch-native library for building, training and teaching transformer language models, with architectures written as ordinary nn.Modules.