pyqlib 0.9.7

A Quantitative-research Platform

Stars: 37472, Watchers: 37472, Forks: 5822, Open Issues: 350

The microsoft/qlib repo was created 5 years ago and the last code push was 5 days ago.
The project is extremely popular with a mindblowing 37472 github stars!

How to Install pyqlib

You can install pyqlib using pip

pip install pyqlib

or add it to a project with poetry

poetry add pyqlib

Package Details

Author

None

License

None

Homepage

None

PyPi:

https://pypi.org/project/pyqlib/

GitHub Repo:

https://github.com/microsoft/qlib

Classifiers

Related Packages

Errors

A list of common pyqlib errors.

Code Examples

Here are some pyqlib code examples and snippets.

GitHub Issues

The pyqlib package has 350 open issues on GitHub

• enhance observability with structured logging and metrics
• out of memory for instrument "all"
• Save local changes in qlib
• two date ranges for train data
• feat: add comprehensive XPU and MPS support for PyTorch models
• [Bug] SZSE trading calendar fetch failed for years 2000-2004 in data_collector/utils.py
• fix(security): prevent path traversal in user data deletion
• feat: add auto lint fix CI workflow
• ExpressionDFilter result not as expected
• Frequency format error in US data
• Prediction scores change when test segment range changes, causing backtest unreliability
• 兼容mac的mps
• feat(config): add explicit validation for required configuration fields
• refactor: implement deterministic budget allocation in SoftTopkStrategy
• fix(docker): improve Dockerfile build for source installation

See more issues on GitHub