Key concepts

Using a Module

Instantiating a Module

A TensorFlow Hub module is imported into a TensorFlow program by creating a Module object from a string with its URL or filesystem path, such as:

m <- hub_load("path/to/a/module_dir")

This adds the module’s variables to the current TensorFlow graph.

Caching Modules

When creating a module from a URL, the module content is downloaded and cached in the local system temporary directory. The location where modules are cached can be overridden using TFHUB_CACHE_DIR environment variable.

For example, setting TFHUB_CACHE_DIR to /my_module_cache:

Sys.setenv(TFHUB_CACHE_DIR = "/my_module_cache")

and then creating a module from a URL:

m <- hub_load("")

results in downloading and unpacking the module into /my_module_cache.

Applying a Module

Once instantiated, a module m can be called zero or more times like a Python function from tensor inputs to tensor outputs:

y <- m(x)

Each such call adds operations to the current TensorFlow graph to compute y from x. If this involves variables with trained weights, these are shared between all applications.

Modules can define multiple named signatures in order to allow being applied in more than one way (similar to how Python objects have methods). A module’s documentation should describe the available signatures. The call above applies the signature named “default”. Any signature can be selected by passing its name to the optional signature= argument.

If a signature has multiple inputs, they must be passed as a dict, with the keys defined by the signature. Likewise, if a signature has multiple outputs, these can be retrieved as a dict by passing as_dict=True, under the keys defined by the signature. (The key "default" is for the single output returned if as_dict=FALSE) So the most general form of applying a Module looks like:

outputs <- m(list(apples=x1, oranges=x2), signature="fruit_to_pet", as_dict=TRUE)
y1 = outputs$cats
y2 = outputs$dogs

A caller must supply all inputs defined by a signature, but there is no requirement to use all of a module’s outputs. Module consumers should handle additional outputs gracefully.

Creating a new Module

General approach

A Hub Module is simply a TensorFlow graph in the SavedModel format. In order to create a Module you can run the export_savedmodel function with any TensorFlow object.

For example:


mnist <- dataset_mnist()

input <- layer_input(shape(28,28), dtype = "int32")

output <- input %>% 
  layer_flatten() %>% 
  layer_lambda(tensorflow::tf_function(function(x) tf$cast(x, tf$float32)/255)) %>% 
  layer_dense(units = 10, activation = "softmax")

model <- keras_model(input, output)

model %>% 
    loss = "sparse_categorical_crossentropy",
    optimizer = "adam",
    metrics = "acc"

model %>% 
  fit(x = mnist$train$x, y = mnist$train$y, validation_split = 0.2, epochs =1 )

save_model_tf(model, "my_module/", include_optimizer = FALSE)

After exporting the model to the SavedModel format you can load it using hub_load, and use it for predictions for example:

module <- hub_load("my_module/")

predictions <- module(mnist$test$x) %>% 
  tf$argmax(axis = 1L) 

mean(as.integer(predictions) == mnist$test$y)

Exporting a module serializes its definition together with the current state of its variables in session into the passed path. This can be used when exporting a module for the first time, as well as when exporting a fine tuned module.

Module publishers should implement a common signature when possible, so that consumers can easily exchange modules and find the best one for their problem.


Training the variables of an imported module together with those of the model around it is called fine-tuning. Fine-tuning can result in better quality, but adds new complications. We advise consumers to look into fine-tuning only after exploring simpler quality tweaks.

For consumers

To enable fine-tuning, instantiate the module with hub_module(..., trainable = TRUE) to make its variables trainable and import TensorFlow’s REGULARIZATION_LOSSES. If the module has multiple graph variants, make sure to pick the one appropriate for training. Usually, that’s the one with tags {"train"}.

Choose a training regime that does not ruin the pre-trained weights, for example, a lower learning rate than for training from scratch.

For publishers

To make fine-tuning easier for consumers, please be mindful of the following:

Hosting a Module

TensorFlow Hub supports HTTP based distribution of modules. In particular the protocol allows to use the URL identifying the module both as the documentation of the module and the endpoint to fetch the module.


When a URL such as is used to identify a module to load or instantiate, the module resolver will attempt to download a compressed tar ball from the URL after appending a query parameter ?tf-hub-format=compressed.

The query param is to be interpreted as a comma separated list of the module formats that the client is interested in. For now only the "compressed" format is defined.

The compressed format indicates that the client expects a tar.gz archive with the module contents. The root of the archive is the root of the module directory and should contain a module e.g.:

# Create a compressed module from an exported module directory.
$ tar -cz -f module.tar.gz --owner=0 --group=0 -C /tmp/export-module/ .

# Inspect files inside a compressed module
$ tar -tf module.tar.gz