krl-model-zoo 2.0.2

Community tier: 20 essential open-source models for vision, NLP, time series, audio, and multimodal tasks. FREE forever.

KR-Labs Model Zoo

Open Models. Trusted Intelligence. Shared Progress.

Documentation • Quick Start • Examples • Pricing • Contributing

---

🎉 NEW: KRL Model Zoo v2.0 - Choose Your Tier

We've restructured the Model Zoo into three tiers to better serve researchers, professionals, and enterprises:

Tier	Models	Price	Best For
Community 🌍	20 models	FREE	Learning, prototyping, open research
Professional 💼	65 models	$49/month	Production apps, advanced analytics
Enterprise 🏢	105 models	$299/month	Mission-critical systems, cutting-edge AI

→ You are viewing the Community tier (FREE forever)

Upgrade to Professional → | Explore Enterprise → | Compare Tiers →

---

Highlights

• 20 Production-Ready Models - Essential open-source models for vision, NLP, time series, audio, and multimodal tasks
• Built for Public Data - Native integration with Census, BLS, FRED, CDC, HUD via KRL Data Connectors
• Research-Grade Quality - Peer-reviewed algorithms with comprehensive validation and testing
• Production Ready - Battle-tested in real-world policy analysis and forecasting applications
• Fully Documented - Complete API reference, tutorials, and mathematical formulations
• Open Source - Apache 2.0 License, community-driven development

---

Overview

KRL Model Zoo™ is an open-source library of socioeconomic and econometric models designed for researchers, policymakers, analysts, and community organizations seeking to turn data into actionable intelligence.

It provides modular, production-grade tools for forecasting, regional analysis, anomaly detection, and policy evaluation — all engineered with transparency, reproducibility, and accessibility in mind.

Developed by KR-Labs, the Model Zoo bridges statistical rigor with public-interest purpose, transforming raw data into real-world insight.

Integrated Ecosystem

The Model Zoo is designed to work seamlessly with KRL Data Connectors, our companion library for accessing public datasets. Together, they provide an end-to-end workflow:

Data → Analysis → Insight

• KRL Data Connectors pulls data from Census, BLS, FRED, CDC, HUD, and 20+ other federal sources
• KRL Model Zoo applies statistical models and analytical frameworks
• Results drive evidence-based decisions and policy evaluation

---

Mission

We believe data science should serve humanity.
The KRL Model Zoo exists to make responsible, interpretable, and replicable analytics accessible to everyone shaping economic, social, or community outcomes.

Our models support research, governance, and education by standardizing open methods for understanding change.

---

Core Capabilities

• Unified Framework: Econometric, causal, ML, and Bayesian models under one architecture.
• Transparency by Design: Built-in provenance tracking, version control, and reproducibility.
• Seamless Data Integration: Native compatibility with KRL Data Connectors for federal datasets.
• Community-Driven: Collaborative ecosystem for model sharing, validation, and continuous improvement.
• Policy-Ready Tools: Designed for public datasets like Census, BLS, FRED, CDC, and HUD.
• Educational Utility: Ideal for universities, research institutes, and civic technology programs.

---

Community Tier Models (20 Models - FREE Forever)

Category	Models Included	Use Case
Computer Vision (8)	ResNet-50, MobileNetV2, EfficientNet-B0, YOLO-v5s, Faster R-CNN, U-Net, DeepLabV3, OpenPose	Image classification, object detection, segmentation, pose estimation
NLP (6)	BERT-base, DistilBERT, GPT-2-small, Word2Vec, TextCNN, Sentiment-RoBERTa	Text encoding, generation, classification, sentiment analysis
Time Series (3)	LSTM, GRU, TabNet	Sequential modeling, forecasting, tabular data
Audio (2)	Wav2Vec2-base, SpeechBrain-ASR	Speech recognition, audio processing
Multimodal (1)	CLIP (ViT-B/32)	Vision-language understanding, zero-shot classification

All models are open-source with permissive licenses (MIT/Apache 2.0). Perfect for learning, prototyping, and open research.

🚀 Need More Models?

Professional Tier ($49/month) adds 45 advanced models:

• Larger model variants (ResNet-101/152, BERT-large, EfficientNet B1-B4)
• Domain-specific models (forecasting, recommendation systems, GNNs)
• Advanced architectures (Vision Transformers, T5, BART, XLNet)
• Total: 65 models (20 Community + 45 Professional)

Enterprise Tier ($299/month) includes cutting-edge models:

• 7B LLMs (Llama-2, Mistral, Falcon, Code Llama)
• Latest research (SAM, NeRF, AlphaFold2, Stable Diffusion)
• Specialized models (protein folding, molecular property prediction)
• Total: 105 models (Full catalog with SLA and dedicated support)

Compare All Tiers →

---

Tier Comparison

Feature	Community 🌍	Professional 💼	Enterprise 🏢
Price	FREE	$49/month	$299/month
Model Count	20 models	65 models	105 models
Computer Vision	8 base models	+18 advanced models	+15 cutting-edge models
NLP	6 base models	+12 advanced models	+10 large LLMs (7B+)
Time Series	3 models	+6 forecasting models	+4 foundation models
Audio/Speech	2 models	+4 models (Whisper, TTS)	—
Multimodal	CLIP-base	—	+4 models (BLIP, Flamingo, LLaVA)
Specialized	—	Recommendation, GNNs	AlphaFold2, ESM-2, Stable Diffusion
License	Open source (MIT/Apache)	Commercial license	Enterprise license
Support	Community (GitHub Issues)	Email support	Dedicated support + SLA
Updates	Major releases	Priority updates	Early access to new models
Usage Tracking	No tracking	Basic analytics	Advanced metrics + monitoring
Code Obfuscation	No	No	Yes (PyArmor protection)
SLA	No	No	99.5% uptime guarantee
Best For	Students, researchers, hobbyists	Startups, production apps, consultancies	Enterprises, mission-critical systems

Why Upgrade?

Community → Professional (+$49/month):

• 3x more models (20 → 65)
• Advanced variants (BERT-large, ResNet-152, T5, BART)
• Domain-specific tools (forecasting, recommendations, GNNs)
• Email support for production issues
• ROI: Better accuracy = better decisions = business value

Professional → Enterprise (+$250/month):

• Full model catalog (65 → 105)
• Cutting-edge AI (7B LLMs, SAM, AlphaFold2, Stable Diffusion)
• 99.5% uptime SLA
• Dedicated support engineer
• Code protection (PyArmor obfuscation)
• ROI: Enterprise-grade reliability + exclusive models = competitive advantage

---

Practical Impact

The Model Zoo powers work that matters:

• Labor & Employment: Forecasting job trends, analyzing workforce shifts, and tracking equity gaps.
• Housing & Urban Development: Modeling affordability, detecting displacement, and identifying price volatility.
• Income & Inequality: Measuring economic disparity, mobility, and opportunity over time.
• Public Health: Linking health indicators with economic and environmental conditions.
• Regional Development: Assessing industrial strengths, resilience, and competitiveness.

Each model is field-tested, policy-relevant, and community-accessible.

---

Quick Start

Installation

Install from PyPI (recommended):

pip install krl-model-zoo

📝 Note about PATH Warnings:
During installation, you may see warnings like:
WARNING: The scripts tqdm, transformers-cli are installed in '/path/to/bin' which is not on PATH.

This is normal and can be safely ignored. These warnings come from dependencies (like transformers, tqdm) installing command-line tools. The KRL Model Zoo library itself works perfectly via Python imports - you don't need these CLI tools.

Only add the directory to PATH if you specifically need to use dependency CLI commands. See INSTALLATION.md for detailed solutions.

Check your environment: Run python -m krl_model_zoo.env_check for diagnostics and recommendations.

Install with optional dependencies:

# For development
pip install krl-model-zoo[dev]

# For testing
pip install krl-model-zoo[test]

# For documentation
pip install krl-model-zoo[docs]

# All optional dependencies
pip install krl-model-zoo[all]

Install from source:

git clone https://github.com/KR-Labs/krl-model-zoo.git
cd krl-model-zoo
pip install -e .

Complete ecosystem setup:

# Install Model Zoo + Data Connectors for seamless data access
pip install krl-model-zoo krl-data-connectors

Complete Workflow Example

# Step 1: Fetch data using KRL Data Connectors
from krl_data_connectors import BLSConnector
from krl_models.econometric import SARIMAModel
from krl_models.core import ModelInputSchema, ModelMetadata
import numpy as np

# Get unemployment data from BLS
bls = BLSConnector()
unemployment_data = bls.get_series('LNS14000000', start_year=2015, end_year=2024)

# Step 2: Prepare data for modeling
# Create input schema with actual field names
schema = ModelInputSchema(
    y=unemployment_data['value'].values,  # Required: target as numpy array
    X=None,  # No exogenous variables for univariate series
    params={},
    feature_names=None
)

# Create model metadata
meta = ModelMetadata(
    name='unemployment_forecast',
    version='1.0.0',
    author='YourName',
    description='SARIMA forecast for US unemployment',
    tags=['unemployment', 'BLS', 'forecast']
)

# Step 3: Create and fit model
model = SARIMAModel(
    input_schema=schema,
    params={'order': (1, 1, 1), 'seasonal_order': (0, 0, 0, 0), 'trend': 'c'},
    meta=meta
)

# Fit and forecast
result = model.fit()
forecast = model.forecast(steps=12, confidence_level=0.95)

# Access forecast results
print(f"Forecast values: {forecast.forecast_values}")
print(f"Lower CI: {forecast.ci_lower}")
print(f"Upper CI: {forecast.ci_upper}")

Standalone Usage (Without Data Connectors)

from krl_models.econometric import SARIMAModel
from krl_models.core import ModelInputSchema, ModelMetadata
import pandas as pd
import numpy as np

# Use your own data
data = pd.read_csv('your_data.csv')

# Create input schema
schema = ModelInputSchema(
    y=data['value'].values,  # Your target variable
    X=None,  # Optional: exogenous features
    params={},
    feature_names=None
)

# Create metadata
meta = ModelMetadata(
    name='my_forecast',
    version='1.0.0',
    author='YourName',
    description='Custom SARIMA forecast',
    tags=['custom', 'forecast']
)

# Forecast with SARIMA
model = SARIMAModel(
    input_schema=schema,
    params={'order': (1, 1, 1), 'seasonal_order': (1, 1, 1, 12), 'trend': 'c'},
    meta=meta
)
result = model.fit()
forecast = model.forecast(steps=12)

print(f"12-month forecast: {forecast.forecast_values}")

Explore Example Notebooks

Visit /examples/notebooks/ for guided walkthroughs:

• End-to-End Workflows: Fetching data with Connectors + analyzing with Model Zoo
• Forecasting labor market trends with BLS data
• Regional specialization analysis using Census CBP data
• Housing market volatility with HUD Fair Market Rent data
• Anomaly detection in CDC health indicators
• Multi-source integration: combining FRED, BLS, and Census data

---

Community & Collaboration

KR-Labs thrives on open collaboration and shared intelligence.

You can:

• Contribute new models, tutorials, or datasets
• Report issues or propose features
• Share research Mapplications and case studies
• Join our growing community of open-data practitioners

See our Contributing Guide for details.
Join the discussion at GitHub Discussions.

---

Governance & Roadmap

The Model Zoo evolves through the KR-Labs Gate Framework, ensuring transparent, high-quality development:

Gate	Description	Status
Gate 1 – Foundation	Time series & econometric core	Open-Source
Gate 2 – Domain Models	Regional & anomaly detection	Open-Source
Gate 3 – Ensembles	Meta-models and hybrid systems	Enterprise App
Gate 4 – Advanced Extensions	Causal inference, Bayesian, network models	Enterprise App

Each Gate reflects a maturity milestone balancing innovation, stability, and community feedback.

---

KR-Labs Ecosystem

The Model Zoo is part of a broader open-source intelligence platform:

Repository	Purpose	Status
krl-model-zoo	Statistical models & analytical frameworks	Production
krl-data-connectors	Unified API for 20+ federal data sources	Production

Why This Matters

Traditional Approach:

1. Manually download CSVs from multiple government websites
2. Write custom parsers for each data format
3. Clean and normalize data structures
4. Finally begin analysis

KR-Labs Approach:

1. pip install krl-data-connectors krl-model-zoo
2. Write analysis code immediately
3. Results in minutes, not days

Key Benefits:

• Single API for Census, BLS, FRED, CDC, HUD, and more
• Pre-validated data compatible with Model Zoo methods
• Reproducible workflows from data fetch to final analysis
• Version control for both data retrieval and model parameters
• Community support for common use cases and patterns

---

Documentation

📚 Full Documentation on ReadTheDocs (Coming Soon)

Quick Links

• Quickstart Guide – Get started in 5 minutes
• User Guide – Comprehensive usage documentation
• API Reference – Complete API documentation
• Contributing Guide – Join our open-source community
• Development Guide – Architecture and development practices
• Testing Guide – Quality assurance and test framework

Example Code

• Python Examples – Standalone Python scripts demonstrating model usage
• Jupyter Notebooks – Interactive tutorials with real datasets

Related Documentation

• KRL Data Connectors – Integrate with 20+ federal data sources

---

Licensing

• Software: Apache 2.0 License – Free for academic and commercial use
• Documentation: CC-BY-SA-4.0

KR-Labs™ and KRL Model Zoo™ are trademarks of Deloatch, Williams, Faison, & Parker, LLLP.

---

Citation

If you use the KRL Model Zoo in research or analysis, please cite:

@software{krl-model-zoo,
  author = {Deloatch, Brandon C.},
  title = {KRL Model Zoo: Open-Source Socioeconomic Modeling Framework},
  Year = {2025},
  publisher = {KR-Labs},
  version = {1.0.0},
  url = {https://github.com/KR-Labs/krl-model-zoo}
}

---

Contact & Support

• Website: krlabs.dev
• Email: [e-mail]
• GitHub Discussions: Join the conversation

---

KR-Labs™ | Building Open Intelligence for the Public Good

Version 1.0.0 • Apache 2.0 License

---

Legal & Trademark Notice

© 2025 KR-Labs. All rights reserved.

KR-Labs™ and KRL Model Zoo™ are trademarks of Deloatch, Williams, Faison, & Parker, LLLP

Software License: Apache 2.0 – Free for commercial and academic use
Documentation License: CC-BY-SA-4.0

---