Cortenn

Synopsis

Cortenn implements data formatting and manipulation portions for the Tenncor libraries help developers write math equations for machine learning.

This project is a multiplex layer for Tenncor and some data-manipulation library. This library exist to avoid changing Tenncor too much.

Components

• LLO (Low Level Operators)

This module is implements basic operations for Tenncor's ADE Tensor objects generated through pybinder. Additionally, llo also defines data format and (de)serialization methods required by PBM.

• OPT (Optimizer)

This module specifies graph optimization through ADE's visitor pattern.

• PBM (Protobuf Marshaller)

This module marshals any ADE graph, but requires data serialization functors when saving and loading.

• Pybinder

This generator extends Tenncor's AGE generator. In this instance, on top of generating the ADE operators specified in LLO, pybinder generates pybind11 binding code.

Building

Cortenn uses bazel 0.15+.

Download bazel: https://docs.bazel.build/versions/master/install.html