Skip to content

Runtime linkin - eager support #430

@coldhearti

Description

@coldhearti

Using the "eager" flag simultaneously with the "tensorflow_runtime_linking" flag results in the compilation of the entire tensorflow C api. This is not ideal.

I have compiled an optimized tensorflow for an aarch64 production environment, so runtime linking is what I'm looking for. My project uses the eager API (has worked fine so far). I'm trying to compile the project using target=aarch64_unknown_linux_gnu and features= "eager", "tensorflow_runtime_linking", "ndarray".

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions