Skip to content

Release 1.2
Compare
Choose a tag to compare
@jonaballe jonaballe released this 15 Aug 21:10
v1.2
a543a0a

Release 1.2 is the first major feature upgrade for tensorflow-compression. Because it contains custom TensorFlow ops, you need to compile this release to use it. If you're on Linux or Darwin (Mac OS), install the pre-compiled binary by running:

pip install tensorflow-compression==1.2

This release requires TensorFlow 1.14.

New features in 1.2:

  • Proper release schedule and binaries. We've tagged existing code versions 1.0 and 1.1, which work without installing binaries with TensorFlow <=1.12 and 1.13, respectively. These versions rely on a binary range coder implementation in tf.contrib, which will not exist any more in TensorFlow 1.14 (the TensorFlow team will remove tf.contrib entirely in release 2.0). For this reason, tensorflow-compression 1.2 ships with its own range coder implementations. Pre-compiled pip packages are provided for Linux and Darwin (Mac).
  • Support for conditional entropy models. This version includes three new classes, GaussianConditional,LaplacianConditional, and LogisticConditional, which implement conditional entropy models (entropy models whose probabilities are computed using another neural network). These are necessary for building the hyperprior models published in our ICLR 2018 paper. This also includes a more flexible range coding implementation, which can encode unbounded integer values. An example training script for the ICLR 2018 model is provided in the examples directory.
  • Support for trained models. The library now includes support for a new file format, TFCI, and packaged metagraphs of models we've published. examples/tfci.py implements an easy-to-use command line interface for converting PNG images to TFCI and back.
  • More flexible SignalConv layers. Among some other minor improvements, the SignalConv* Keras layers now implement rational up-/downsampling factors.
Assets 2
Loading