So, how come we are able to use TensorFlow from R?

Which pc language is most carefully related to TensorFlow? Whereas on the TensorFlow for R weblog, we might in fact like the reply to be R, chances are high it’s Python (although TensorFlow has official bindings for C++, Swift, Javascript, Java, and Go as effectively).

So why is it you’ll be able to outline a Keras mannequin as

library(keras)
mannequin <- keras_model_sequential() %>%
  layer_dense(items = 32, activation = "relu") %>%
  layer_dense(items = 1)

(good with %>%s and all!) – then practice and consider it, get predictions and plot them, all that with out ever leaving R?

The quick reply is, you’ve gotten keras, tensorflow and reticulate put in.
reticulate embeds a Python session inside the R course of. A single course of means a single deal with area: The identical objects exist, and may be operated upon, no matter whether or not they’re seen by R or by Python. On that foundation, tensorflow and keras then wrap the respective Python libraries and allow you to write R code that, actually, appears to be like like R.

This publish first elaborates a bit on the quick reply. We then go deeper into what occurs within the background.

One be aware on terminology earlier than we leap in: On the R aspect, we’re making a transparent distinction between the packages keras and tensorflow. For Python we’re going to use TensorFlow and Keras interchangeably. Traditionally, these have been completely different, and TensorFlow was generally regarded as one doable backend to run Keras on, apart from the pioneering, now discontinued Theano, and CNTK. Standalone Keras does nonetheless exist, however current work has been, and is being, performed in tf.keras. After all, this makes Python Keras a subset of Python TensorFlow, however all examples on this publish will use that subset so we are able to use each to seek advice from the identical factor.

So keras, tensorflow, reticulate, what are they for?

Firstly, nothing of this might be doable with out reticulate. reticulate is an R package deal designed to permit seemless interoperability between R and Python. If we completely needed, we might assemble a Keras mannequin like this:

<class 'tensorflow.python.keras.engine.sequential.Sequential'>

We might go on including layers …

m$add(tf$keras$layers$Dense(32, "relu"))
m$add(tf$keras$layers$Dense(1))
m$layers

[[1]]
<tensorflow.python.keras.layers.core.Dense>

[[2]]
<tensorflow.python.keras.layers.core.Dense>

However who would need to? If this had been the one approach, it’d be much less cumbersome to instantly write Python as a substitute. Plus, as a person you’d must know the entire Python-side module construction (now the place do optimizers stay, at the moment: tf.keras.optimizers, tf.optimizers …?), and sustain with all path and identify modifications within the Python API.

That is the place keras comes into play. keras is the place the TensorFlow-specific usability, re-usability, and comfort options stay.
Performance offered by keras spans the entire vary between boilerplate-avoidance over enabling elegant, R-like idioms to offering technique of superior characteristic utilization. For example for the primary two, contemplate layer_dense which, amongst others, converts its items argument to an integer, and takes arguments in an order that enable it to be “pipe-added” to a mannequin: As a substitute of

mannequin <- keras_model_sequential()
mannequin$add(layer_dense(items = 32L))

we are able to simply say

mannequin <- keras_model_sequential()
mannequin %>% layer_dense(items = 32)

Whereas these are good to have, there may be extra. Superior performance in (Python) Keras largely relies on the flexibility to subclass objects. One instance is customized callbacks. For those who had been utilizing Python, you’d must subclass tf.keras.callbacks.Callback. From R, you’ll be able to create an R6 class inheriting from KerasCallback, like so

CustomCallback <- R6::R6Class("CustomCallback",
    inherit = KerasCallback,
    public = record(
      on_train_begin = operate(logs) {
        # do one thing
      },
      on_train_end = operate(logs) {
        # do one thing
      }
    )
  )

It is because keras defines an precise Python class, RCallback, and maps your R6 class’ strategies to it.
One other instance is customized fashions, launched on this weblog a couple of yr in the past.
These fashions may be educated with customized coaching loops. In R, you utilize keras_model_custom to create one, for instance, like this:

m <- keras_model_custom(identify = "mymodel", operate(self) {
  self$dense1 <- layer_dense(items = 32, activation = "relu")
  self$dense2 <- layer_dense(items = 10, activation = "softmax")
  
  operate(inputs, masks = NULL) {
    self$dense1(inputs) %>%
      self$dense2()
  }
})

Right here, keras will ensure an precise Python object is created which subclasses tf.keras.Mannequin and when known as, runs the above nameless operate().

In order that’s keras. What concerning the tensorflow package deal? As a person you solely want it when you need to do superior stuff, like configure TensorFlow system utilization or (in TF 1.x) entry parts of the Graph or the Session. Internally, it’s utilized by keras closely. Important inner performance contains, e.g., implementations of S3 strategies, like print, [ or +, on Tensors, so you can operate on them like on R vectors.

Now that we know what each of the packages is “for”, let’s dig deeper into what makes this possible.

Show me the magic: reticulate

Instead of exposing the topic top-down, we follow a by-example approach, building up complexity as we go. We’ll have three scenarios.

First, we assume we already have a Python object (that has been constructed in whatever way) and need to convert that to R. Then, we’ll investigate how we can create a Python object, calling its constructor. Finally, we go the other way round: We ask how we can pass an R function to Python for later usage.

Scenario 1: R-to-Python conversion

Let’s assume we have created a Python object in the global namespace, like this:

So: There is a variable, called x, with value 1, living in Python world. Now how do we bring this thing into R?

We know the main entry point to conversion is py_to_r, defined as a generic in conversion.R:

py_to_r <- function(x) {
  ensure_python_initialized()
  UseMethod("py_to_r")
}

… with the default implementation calling a function named py_ref_to_r:

#' @export
py_to_r.default <- function(x) {
  [...]
  x <- py_ref_to_r(x)
  [...]
}

To seek out out extra about what’s going on, debugging on the R degree gained’t get us far. We begin gdb so we are able to set breakpoints in C++ features:

$ R -d gdb

GNU gdb (GDB) Fedora 8.3-6.fc30
[... some more gdb saying hello ...]
Studying symbols from /usr/lib64/R/bin/exec/R...
Studying symbols from /usr/lib/debug/usr/lib64/R/bin/exec/R-3.6.0-1.fc30.x86_64.debug...

Now begin R, load reticulate, and execute the project we’re going to presuppose:

(gdb) run
Beginning program: /usr/lib64/R/bin/exec/R 
[...]
R model 3.6.0 (2019-04-26) -- "Planting of a Tree"
Copyright (C) 2019 The R Basis for Statistical Computing
[...]
> library(reticulate)
> py_run_string("x = 1")

In order that arrange our state of affairs, the Python object (named x) we need to convert to R. Now, use Ctrl-C to “escape” to gdb, set a breakpoint in py_to_r and kind c to get again to R:

(gdb) b py_to_r
Breakpoint 1 at 0x7fffe48315d0 (2 areas)
(gdb) c

Now what are we going to see once we entry that x?

> py$x

Thread 1 "R" hit Breakpoint 1, 0x00007fffe48315d0 in py_to_r(libpython::_object*, bool)@plt () from /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/reticulate/libs/reticulate.so

Listed here are the related (for our investigation) frames of the backtrace:

Thread 1 "R" hit Breakpoint 3, 0x00007fffe48315d0 in py_to_r(libpython::_object*, bool)@plt () from /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/reticulate/libs/reticulate.so
(gdb) bt
#0  0x00007fffe48315d0 in py_to_r(libpython::_object*, bool)@plt () from /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/reticulate/libs/reticulate.so
#1  0x00007fffe48588a0 in py_ref_to_r_with_convert (x=..., convert=true) at reticulate_types.h:32
#2  0x00007fffe4858963 in py_ref_to_r (x=...) at /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/Rcpp/embrace/RcppCommon.h:120
#3  0x00007fffe483d7a9 in _reticulate_py_ref_to_r (xSEXP=0x55555daa7e50) at /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/Rcpp/embrace/Rcpp/as.h:151
...
...
#14 0x00007ffff7cc5fc7 in Rf_usemethod (generic=0x55555757ce70 "py_to_r", obj=obj@entry=0x55555daa7e50, name=name@entry=0x55555a0fe198, args=args@entry=0x55555557c4e0, 
    rho=rho@entry=0x55555dab2ed0, callrho=0x55555dab48d8, defrho=0x5555575a4068, ans=0x7fffffff69e8) at objects.c:486

We’ve eliminated a number of intermediate frames associated to (R-level) technique dispatch.

As we already noticed within the supply code, py_to_r.default will delegate to a technique known as py_ref_to_r, which we see seems in #2. However what’s _reticulate_py_ref_to_r in #3, the body just under? Right here is the place the magic, unseen by the person, begins.

Let’s take a look at this from a fowl’s eye’s view. To translate an object from one language to a different, we have to discover a widespread floor, that’s, a 3rd language “spoken” by each of them. Within the case of R and Python (in addition to in numerous different instances) this can be C / C++. So assuming we’re going to write a C operate to speak to Python, how can we use this operate in R?

Rcpp : You simply write your C++ operate, and Rcpp takes care of compilation and offers the glue code essential to name this operate from R.

So py_ref_to_r actually is written in C++:

// [[Rcpp::export]]
SEXP py_ref_to_r(PyObjectRef x) {
  return py_ref_to_r_with_convert(x, x.convert());
}

py_ref_to_r <- operate(x) {
  .Name(`_reticulate_py_ref_to_r`, x)
}

which lastly wraps the “actual” factor, the C++ operate py_ref_to_R we noticed above.

By way of py_ref_to_r_with_convert in #1, a one-liner that extracts an object’s “convert” characteristic (see under)

// [[Rcpp::export]]
SEXP py_ref_to_r_with_convert(PyObjectRef x, bool convert) {
  return py_to_r(x, convert);
}

we lastly arrive at py_to_r in #0.

Earlier than we take a look at that, let’s ponder that C/C++ “bridge” from the opposite aspect – Python.
Whereas strictly, Python is a language specification, its reference implementation is CPython, with a core written in C and rather more performance constructed on prime in Python. In CPython, each Python object (together with integers or different numeric sorts) is a PyObject. PyObjects are allotted by means of and operated on utilizing pointers; most C API features return a pointer to at least one, PyObject *.

So that is what we count on to work with, from R. What then is PyObjectRef doing in py_ref_to_r?
PyObjectRef isn’t a part of the C API, it’s a part of the performance launched by reticulate to handle Python objects. Its essential objective is to verify the Python object is mechanically cleaned up when the R object (an Rcpp::Atmosphere) goes out of scope.
Why use an R surroundings to wrap the Python-level pointer? It is because R environments can have finalizers: features which might be known as earlier than objects are rubbish collected.
We use this R-level finalizer to make sure the Python-side object will get finalized as effectively:

Rcpp::RObject xptr = R_MakeExternalPtr((void*) object, R_NilValue, R_NilValue);
R_RegisterCFinalizer(xptr, python_object_finalize);

inline void python_object_finalize(SEXP object) {
  PyObject* pyObject = (PyObject*)R_ExternalPtrAddr(object);
  if (pyObject != NULL)
    Py_DecRef(pyObject);
}

SEXP py_to_r(PyObject* x, bool convert) {
  //...
}

and – however wait. Didn’t py_ref_to_r_with_convert move it a PyObjectRef? So how come it receives a PyObject as a substitute? It is because PyObjectRef inherits from Rcpp::Atmosphere, and its implicit conversion operator is used to extract the Python object from the Atmosphere. Concretely, that operator tells the compiler {that a} PyObjectRef can be utilized as if it had been a PyObject* in some ideas, and the related code specifies how one can convert from PyObjectRef to PyObject*:

operator PyObject*() const {
  return get();
}

PyObject* get() const {
  SEXP pyObject = getFromEnvironment("pyobj");
  if (pyObject != R_NilValue) {
    PyObject* obj = (PyObject*)R_ExternalPtrAddr(pyObject);
    if (obj != NULL)
      return obj;
  }
  Rcpp::cease("Unable to entry object (object is from earlier session and is now invalid)");
}


else if (scalarType == INTSXP)
  return IntegerVector::create(PyInt_AsLong(x));

For different objects, sometimes there’s extra motion required; however basically, the operate is “simply” a giant if–else tree.

So this was state of affairs 1: changing a Python object to R. Now in state of affairs 2, we assume we nonetheless have to create that Python object.

Situation 2:

As this state of affairs is significantly extra advanced than the earlier one, we’ll explicitly focus on some features and miss others. Importantly, we’ll not go into module loading, which might deserve separate remedy of its personal. As a substitute, we attempt to shed a light-weight on what’s concerned utilizing a concrete instance: the ever-present, in keras code, keras_model_sequential(). All this R operate does is

operate(layers = NULL, identify = NULL) {
  keras$fashions$Sequential(layers = layers, identify = identify)
}

How can keras$fashions$Sequential() give us an object? When in Python, you run the equal

tf.keras.fashions.Sequential()

class Sequential(coaching.Mannequin):
  def __init__(self, layers=None, identify=None):
    # ...
  # ...

So this time, earlier than – as all the time, ultimately – getting an R object again from Python, we have to name that constructor, that’s, a Python callable. (Python callables subsume features, constructors, and objects created from a category that has a name technique.)

So when py_to_r, inspecting its argument’s kind, sees it’s a Python callable (wrapped in a PyObjectRef, the reticulate-specific subclass of Rcpp::Atmosphere we talked about above), it wraps it (the PyObjectRef) in an R operate, utilizing Rcpp:

Rcpp::Perform f = py_callable_as_function(pyFunc, convert);

The cpython-side motion begins when py_callable_as_function then calls py_call_impl. py_call_impl executes the precise name and returns an R object, a SEXP. Now you could be asking, how does the Python runtime understand it shouldn’t deallocate that object, now that its work is completed? That is taken of by the identical PyObjectRef class used to wrap cases of PyObject *: It will probably wrap SEXPs as effectively.

Whereas much more might be stated about what occurs earlier than we lastly get to work with that Sequential mannequin from R, let’s cease right here and take a look at our third state of affairs.

Situation 3: Calling R from Python

Not surprisingly, typically we have to move R callbacks to Python. An instance are R knowledge mills that can be utilized with keras fashions .

On the whole, for R objects to be handed to Python, the method is considerably reverse to what we described in instance 1. Say we kind:

This assigns 1 to a variable a within the python essential module.
To allow project, reticulate offers an implementation of the S3 generic $<-, $<-.python.builtin.object, which delegates to py_set_attr, which then calls py_set_attr_impl – yet one more C++ operate exported by way of Rcpp.

Let’s concentrate on a unique facet right here, although. A prerequisite for the project to occur is getting that 1 transformed to Python. (We’re utilizing the only doable instance, clearly; however you’ll be able to think about this getting much more advanced if the item isn’t a easy quantity).

For our “minimal instance”, we see a stacktrace like the next

#0 0x00007fffe4832010 in r_to_py_cpp(Rcpp::RObject_Impl<Rcpp::PreserveStorage>, bool)@plt () from /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/reticulate/libs/reticulate.so
#1  0x00007fffe4854f38 in r_to_py_impl (object=..., convert=convert@entry=true) at /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/Rcpp/embrace/RcppCommon.h:120
#2  0x00007fffe48418f3 in _reticulate_r_to_py_impl (objectSEXP=0x55555ec88fa8, convertSEXP=<optimized out>) at /residence/key/R/x86_64-redhat-linux-gnu-library/3.6/Rcpp/embrace/Rcpp/as.h:151
...
#12 0x00007ffff7cc5c03 in dispatchMethod (sxp=0x55555d0cf1a0, dotClass=<optimized out>, cptr=cptr@entry=0x7ffffffeaae0, technique=technique@entry=0x55555bfe06c0, 
    generic=0x555557634458 "r_to_py", rho=0x55555d1d98a8, callrho=0x5555555af2d0, defrho=0x555557947430, op=<optimized out>, op=<optimized out>) at objects.c:436
#13 0x00007ffff7cc5fc7 in Rf_usemethod (generic=0x555557634458 "r_to_py", obj=obj@entry=0x55555ec88fa8, name=name@entry=0x55555c0317b8, args=args@entry=0x55555557cc60, 
    rho=rho@entry=0x55555d1d98a8, callrho=0x5555555af2d0, defrho=0x555557947430, ans=0x7ffffffe9928) at objects.c:486

Whereas r_to_py is a generic (like py_to_r above), r_to_py_impl is wrapped by Rcpp and r_to_py_cpp is a C++ operate that branches on the kind of the item – principally the counterpart of the C++ r_to_py.

Extending Python Information.

In official phrases, what reticulate does it embed and lengthen Python.
Embed, as a result of it permits you to use Python from inside R. Lengthen, as a result of to allow Python to name again into R it must wrap R features in C, so Python can perceive them.

As a part of the previous, the specified Python is loaded (Py_Initialize()); as a part of the latter, two features are outlined in a brand new module named rpycall, that can be loaded when Python itself is loaded.

PyImport_AppendInittab("rpycall", &initializeRPYCall);

PyMethodDef RPYCallMethods[] = {
   METH_KEYWORDS, "Name an R operate" ,
   METH_KEYWORDS, "Name a Python operate on the principle thread" ,
  { NULL, NULL, 0, NULL }
};

World Interpreter Lock, this isn’t mechanically the case when different implementations are used, or C is used instantly. So call_python_function_on_main_thread makes positive that except we are able to execute on the principle thread, we wait.

That’s it for our three “spotlights on reticulate”.

Wrapup

It goes with out saying that there’s rather a lot about reticulate we didn’t cowl on this article, akin to reminiscence administration, initialization, or specifics of information conversion. Nonetheless, we hope we had been in a position to shed a bit of sunshine on the magic concerned in calling TensorFlow from R.

R is a concise and stylish language, however to a excessive diploma its energy comes from its packages, together with people who let you name into, and work together with, the skin world, akin to deep studying frameworks or distributed processing engines. On this publish, it was a particular pleasure to concentrate on a central constructing block that makes a lot of this doable: reticulate.

Thanks for studying!