TensorFlow v1.3 Released

[This article was first published on TensorFlow for R, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

The final release of TensorFlow v1.3 is now available. This release of TensorFlow marks the initial availability of several canned estimators, including:

  • DNNClassifier
  • DNNRegressor
  • LinearClassifier
  • LinearRegressor
  • DNNLinearCombinedClassifier
  • DNNLinearCombinedRegressor.

The tfestimators package provides a high level R interface for these estimators.

Full details on the release of TensorFlow v1.3 are available here: https://github.com/tensorflow/tensorflow/releases/tag/v1.3.0

You can update your R installation of TensorFlow using the install_tensorflow function:

library(tensorflow)
install_tensorflow()

Note that you should also provide any options used in your original installation (e.g. method = "conda", version = "gpu", etc. )

cuDNN 6.0

TensorFlow v1.3 is built against version 6.0 of the cuDNN library from NVIDIA. Previous versions were built against cuDNN v5.1, so for installations running the GPU version of TensorFlow this means that you will need to install an updated version of cuDNN along with TensorFlow v1.3.

Updated installation instructions are available here: https://tensorflow.rstudio.com/tensorflow/installation_gpu.html.

Version 1.4 of TensorFlow is expected to migrate again to version 7.0 of cuDNN.

To leave a comment for the author, please follow the link and comment on their blog: TensorFlow for R.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.