Skip to main content

​
Changelog

All notable changes to Synthefy will be documented in this file. We follow Semantic Versioning for our releases.
Subscribe to updates: Get notified about new releases by watching our GitHub repository or joining our Discord community.

​
[Unreleased]

​
Added

  • Enhanced API documentation with comprehensive examples
  • New webhook configuration options
  • Improved error handling and retry logic

​
Changed

  • Updated Python SDK with better performance
  • Improved Docker image size and startup time

​
Fixed

  • Fixed authentication issues with long-lived tokens
  • Resolved memory leaks in batch processing

​
[1.2.0] - 2025-10-01

​
Major Features

Added support for advanced forecasting models:
  • Prophet Integration: Better handling of seasonality and holidays
  • LSTM Networks: Deep learning models for complex patterns
  • Auto-ARIMA: Automatic model selection and parameter tuning
Enhanced webhook system with:
  • Event Filtering: Subscribe to specific job events
  • Retry Logic: Improved reliability with exponential backoff
  • Signature Verification: Enhanced security for webhook endpoints
New batch processing capabilities:
  • Multiple Time Series: Process hundreds of series simultaneously
  • Parallel Execution: Faster processing with concurrent jobs
  • Progress Tracking: Real-time updates on batch job progress

​
Enhancements

  • Python SDK: Added support for custom model parameters
  • Docker Agent: Improved health checks and monitoring
  • API: New endpoints for job management and monitoring
  • Documentation: Comprehensive guides and examples

​
Bug Fixes

  • Fixed memory issues with large datasets
  • Resolved authentication token refresh problems
  • Improved error messages for better debugging

​
Documentation

​
[1.1.0] - 2025-09-15

​
Major Features

Docker Support

Full Docker containerization with health checks and monitoring

Kubernetes Deployment

Production-ready Kubernetes manifests and Helm charts

Webhook Integration

Real-time notifications for job completion

Batch Forecasting

Process multiple time series in parallel

​
Enhancements

  • Performance: 3x faster model training and inference
  • Reliability: Improved error handling and recovery
  • Monitoring: Enhanced logging and metrics collection
  • Security: Better authentication and authorization

​
Bug Fixes

  • Fixed data validation issues with missing values
  • Resolved timezone handling problems
  • Improved memory usage for large datasets

​
[1.0.0] - 2025-09-01

​
Initial Release

The first stable release of Synthefy with core forecasting capabilities:
  • Time Series Forecasting: Accurate predictions with confidence intervals
  • Python SDK: Easy-to-use Python library for data scientists
  • HTTP API: RESTful API for any programming language
  • Multiple Models: Support for various forecasting algorithms
  • ARIMA: Auto-regressive integrated moving average
  • Exponential Smoothing: Holt-Winters and variants
  • Linear Regression: Trend-based forecasting
  • Seasonal Decomposition: STL and X-13ARIMA-SEATS
  • Python Package: pip install synthefy-forecasting
  • Docker Container: docker pull synthefy/agent
  • HTTP API: RESTful endpoints with JSON responses
  • Cloud Deployment: AWS, GCP, and Azure support

​
Performance

  • Speed: Sub-second inference for most models
  • Accuracy: State-of-the-art forecasting performance
  • Scalability: Handle thousands of concurrent requests
  • Reliability: 99.9% uptime SLA

​
[0.9.0] - 2025-08-15

​
Beta Release

Public beta with core functionality:
  • Basic forecasting models
  • Python SDK (beta)
  • HTTP API (beta)
  • Documentation site

​
Known Issues

  • Limited model selection
  • Basic error handling
  • No webhook support
  • Limited documentation

​
Migration Guides

​
Upgrading from 1.1.x to 1.2.0

1

Update Python SDK

pip install --upgrade synthefy-forecasting
2

Update Docker Image

docker pull synthefy/agent:1.2.0
3

Review API Changes

Check the API Reference for any breaking changes

​
Upgrading from 1.0.x to 1.1.0

1

Backup Configuration

Save your current configuration files
2

Update Dependencies

pip install --upgrade synthefy-forecasting
3

Test New Features

Verify webhook configuration and batch processing

​
Deprecation Notices

​
Deprecated in 1.2.0

  • Legacy API endpoints: Some v0 endpoints will be removed in v2.0
  • Old authentication method: API key in query parameters (use headers instead)

​
Removed in 1.1.0

  • Python 3.7 support: Minimum version is now Python 3.8
  • Legacy Docker tags: Only latest and versioned tags supported

​
Security Updates

​
1.2.0 Security Improvements

  • Enhanced webhook signature verification
  • Improved API key rotation
  • Better input validation and sanitization
  • Updated dependencies with security patches

​
1.1.0 Security Updates

  • Fixed authentication bypass vulnerability
  • Improved rate limiting
  • Enhanced logging for security events

​
Performance Improvements

1.2.0

  • 50% faster model training
  • 30% reduction in memory usage
  • Improved batch processing

1.1.0

  • 3x faster inference
  • Better caching mechanisms
  • Optimized Docker images

​
Community Contributions

We’d like to thank our community contributors:

​
1.2.0 Contributors

  • @alex-forecaster: Improved Prophet integration
  • @data-scientist: Enhanced documentation examples
  • @dev-ops: Kubernetes deployment improvements

​
1.1.0 Contributors

  • @ml-engineer: Performance optimizations
  • @docker-expert: Container improvements
  • @api-developer: Webhook implementation
Want to contribute? Check out our Development Guide to get started with contributing to Synthefy.

​
Support and Feedback

Report Issues

Found a bug? Let us know!

Feature Requests

Have an idea? We’d love to hear it!

Community

Join our Discord for discussions

Support

Get help from our support team