Learn R Programming

tfautograph

This package implements autograph for R.

It lets you use tensors in R control flow expressions like if, while and for and automatically translates those expressions into tensorflow graphs.

It is compatible with and tested against all Tensorflow versions >= 1.15 (including versions 2.x).

Learn More:

Demos

Additional Articles:

Installation

You can install tfautograph from CRAN with:

install.packages("tfautograph")

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("t-kalinowski/tfautograph")

Copy Link

Version

Install

install.packages('tfautograph')

Monthly Downloads

18,878

Version

0.3.2

License

GPL-3

Maintainer

Tomasz Kalinowski

Last Published

September 17th, 2021

Functions in tfautograph (0.3.2)

tf_case

tf.case
ag_loop_vars

Specify loop variables
tf_map

tf.map_fn()
ag_while_opts

specify tf.while_loop options
tf_assert

tf_assert
ag_name

Specify a tensor name
autograph

Autograph R code
tf_cond

tf.cond
ag_if_vars

Specify tf.cond() output structure when autographing if
[[<-.tensorflow.python.ops.tensor_array_ops.TensorArray

TensorArray.write()
tf_switch

tf.switch_case
view_function_graph

Visualizes the generated graph