omfpandas

A pandas (and parquet) interface for the Open Mining Format package (omf).

When working with OMF files, it is often useful to convert the data to a pandas DataFrame. This package provides a simple interface to do so.

The parquet format is a nice, compact, efficient format to persist pandas DataFrames. This package also provides a simple interface to convert an omf element to a parquet file. When datasets do not fit into memory, parquet files can be read in chunks or by column.

Note: This package only supports omf 2.0, which is currently only a pre-release.

Installation

pip install omfpandas

If you intend to use the parquet functionality, you will need to install the optional dependencies.

pip install omfpandas[io]

Roadmap

• 0.2.0 - Add support for reading a VolumeElement (Block Model) from an OMF file as a pandas DataFrame. Export a VolumeElement as a parquet file.
• 0.3.0 - Add support for writing a DataFrame to an OMF BlockModel. Version 2.0 of the OMF spec is supported.
• 0.4.0 - Convert to omf 2.0 support.
• 0.5.0 - Block model profiling, with reports persisted in the omf file.
• 0.6.0 - Optional block model validation using pandera json schemas.
• 0.7.0 - Add support for low-memory/out-of-core writing an omf element to parquet
• ...