Skip to main content

TFLogger is a standalone library to read and write events for TensorBoard.

Project description


This is a simple library that has been carved out of TensorFlow's codebase for being able to generate the required output format for TensorBoard.


In order to build from source the following packages are required:

  • cmake
  • protobuf

To build the library simply run the following commands:

git clone
cd tflogger
mkdir build
cd build
cmake ..


In order to enable the unit tests add the -DBUILD-TESTS=ON cmake option to the cmake command. This will automatically create a target for building the unit tests and adds it to the ctest targets.

In other words to run the unit tests:

cmake -DBUILD-TESTS=ON [other cmake flags] ..

Python interface

Currently only the python wrapper is available, but further targets could be specified using SWIG.

For building the python wrapper you need to add the -DINTERFACE_PYTHON=ON cmake flag to the cmake command:



Examples of how to use the library are available under example directory.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tflogger-1.0.1.tar.gz (28.1 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page