ludwig

v0.17.1 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 16d LLM Frameworks

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

computer-vision data-centric data-science deep machine-learning deeplearning

+11 more

fine-tuning learning llama llama2 llm llm-training machinelearning mistral natural-language natural-language-processing pytorch

Affected surfaces

breaking_upgrade

Summary

AI summary

Bug fixes, new features including StudioCallback and per-call callbacks, plus upgrade notes replacing the failed 0.17.0 release.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Low	Introduce `StudioCallback` to write `metrics.jsonl` and `trials.jsonl` for Ludwig Studio integration. Introduce `StudioCallback` to write `metrics.jsonl` and `trials.jsonl` for Ludwig Studio integration. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high	—
Feature	Low	Add per-call callbacks on `train()` and `predict()` methods. Add per-call callbacks on `train()` and `predict()` methods. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high	—
Feature	Low	Implement SIGUSR1/SIGUSR2 pause/resume support in training process. Implement SIGUSR1/SIGUSR2 pause/resume support in training process. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high	—
Dependency
Dependency	Low	Replace `pip install ludwig==0.17.0` with `pip install ludwig==0.17.1` due to 0.17.0 never being published to PyPI. Replace `pip install ludwig==0.17.0` with `pip install ludwig==0.17.1` due to 0.17.0 never being published to PyPI. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: low	—
Dependency	Low	Replace `pip install ludwig==0.17.0` with `pip install ludwig==0.17.1` due to 0.17.0 never being published. Replace `pip install ludwig==0.17.0` with `pip install ludwig==0.17.1` due to 0.17.0 never being published. Source: granite4.1:30b@2026-05-19-audit Confidence: low	—
Dependency	Low	Update installation instruction to use `pip install ludwig==0.17.1` (0.17.0 never published). Update installation instruction to use `pip install ludwig==0.17.1` (0.17.0 never published). Source: granite4.1:30b@2026-05-19-audit Confidence: low	—
Bugfix
Bugfix	Medium	Fix preprocessing pipeline crash for output features without `preprocessing` config (e.g., `anomaly` output type). Fix preprocessing pipeline crash for output features without `preprocessing` config (e.g., `anomaly` output type). Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high	—
Bugfix	Medium	Fix preprocessing pipeline crash for output features without `preprocessing` config (e.g., `anomaly`). Fix preprocessing pipeline crash for output features without `preprocessing` config (e.g., `anomaly`). Source: granite4.1:30b@2026-05-19-audit Confidence: high	—
Bugfix	Low	Fix v0.17.0 PyPI deployment issue causing duplicate-upload error. Fix v0.17.0 PyPI deployment issue causing duplicate-upload error. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high	—
Refactor	Low	Hardened Hyperopt executor for ray-free environments; OptunaExecutor no longer requires Ray. Hardened Hyperopt executor for ray-free environments; OptunaExecutor no longer requires Ray. Source: granite4.1:8b-q6_K@2026-05-19 Confidence: high	—

Full changelog

What's changed

Bug fixes

Fix preprocessing pipeline crash for output features without preprocessing config (e.g. anomaly output type) — KeyError in build_preprocessing_parameters, build_dataset, build_metadata, and build_data
Fix v0.17.0 PyPI deployment: the v0.17.0 tag was created before the version bump commit landed, causing the PyPI build to produce ludwig-0.16.2 and fail with a duplicate-upload error

New features

StudioCallback: writes metrics.jsonl and trials.jsonl for Ludwig Studio integration
Per-call callbacks on train() and predict()
Hyperopt executor hardened for ray-free environments (OptunaExecutor no longer requires Ray)
SIGUSR1/SIGUSR2 pause/resume support in training

Upgrade notes

Replace pip install ludwig==0.17.0 with pip install ludwig==0.17.1 (0.17.0 was never successfully published to PyPI)

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track ludwig

Get notified when new releases ship.

About ludwig

Low-code framework for building custom LLMs, neural networks, and other AI models

All releases →