Run this network for the given NumPy array(s), and return the output(s) as NumPy array(s). Args: input_transform: A dict speci

Do general housekeeping and keep the state of the context up-to-date. Should be called often enough but not in a tight loop.

Check that TensorFlow session has been initialized.

Flush, close possible files, and remove stdout/stderr mirroring.

Find variable by local or global name.

Get or create variable representing log2 of the current dynamic loss scaling factor.

Traverses the object name and returns the last (rightmost) python object.

Replace tags in the given path template and return either Windows or Linux formatted path.

Run the specified ops in the default session.

Internal helper for creating autosummary accumulators.

Forcefully enter the specified name scope, ignoring any surrounding scopes.

Close the context and clean up. Should only be called once.

Copy the values of all variables from the given network, including sub-networks.

Flush written text to both stdout and a file, if open.

Searches for the underlying module behind the name to some python object. Returns the module and the object name (original name with module part r

Get the current user name.

Check whether the input is a valid Tensorflow expression, i.e., Tensorflow Tensor, Variable, or Operation.

Re-initialize all variables of this network, excluding sub-networks.

Write text to stdout (and a file) and optionally flush.

Create a new run dir with increasing ID number at the start.

Reads all directory names in a given directory (non-recursive) and returns the next (increasing) run id. Assumes IDs are numbers at the start of the d

Copy all necessary files into the run dir. Assumes that the dir exists, is local, and is writable.

Apply dynamic loss scaling for the given expression.

Create a clone of this network with its own copy of the variables.

Create tf.Session based on config dict.

Create the necessary ops to include autosummaries in TensorBoard report. Note: This should be done only once per graph.

Finds the python object with the given name.

Construct TensorFlow expression(s) for the output(s) of this network, given the input expression(s).

Convert a normal path back to its template representation.

Determine whether the given object is a top-level function, i.e., defined at module scope using 'def'.

Determine whether the given object is a valid URL string.

Returns a list of (layer_name, output_expr, trainable_vars) tuples corresponding to individual layers of the network. Mainly intended to be us

Print a summary table of the network structure.

Reset internal state of the underlying optimizer.

Wrap the actual run function call for handling logging, exceptions, typing, etc.

Set the values of given tf.Variables. Equivalent to the following, but more efficient and does not bloat the tf graph: tflib.run([tf.assign(v

Undo the effect of dynamic loss scaling for the given expression.

Pickle export.

Pickle import.

Forcefully enter the specified variable scope, ignoring any surrounding scopes.

Construct training op to update the registered variables based on their gradients.

Ask the user the question until the user inputs a valid answer.

Create a new autosummary. Args: name: Name to use in TensorBoard value: TensorFlow expression or python value to track

Finds the python object with the given name and calls it as a function.

Create new network with the given parameters, and copy all variables from this network.

Convert a minibatch of images from uint8 to float32 with configurable dynamic range. Can be used as an input transformation for Network.run().

Convert a minibatch of images from float32 to uint8 with configurable dynamic range. Can be used as an output transformation for Network.run().

Convert a normal path to template and the convert it back to a normal path with given path type.

Takes in a list of tuples of (src, dst) paths and copies files. Will create all necessary directories.

Copy the values of all variables from the given network, excluding sub-networks.

Copy the values of all trainable variables from the given network, including sub-networks.

Create tf.Variable with large initial value without bloating the tf graph.

Exponent in base 2.

Shortcut function for flattening a tensor.

Convert the seconds to human readable string with days, hours, minutes and seconds.

Given a type name string (or an object having a __name__ attribute), return matching Numpy and ctypes types that have the same size in bytes.

How much time passed between the previous two calls to update.

Get the directory path of the module containing the given object name.

How much time has passed since the last call to update.

How much time has passed since the creation of the context.

Return the fully-qualified name of a top-level function.

Get the value of a given variable as NumPy array. Note: This method is very inefficient -- prefer to use tflib.run(list_of_vars) whenever poss

Get the local name of a given variable, without any surrounding name scopes.

Function decorator for declaring custom import handlers.

Initialize TensorFlow session using good default settings.

Initialize all tf.Variables that have not already been initialized. Equivalent to the following, but more efficient and does not bloat the tf gra

Linear interpolation.

Linear interpolation with clip.

List all files recursively in a given directory while ignoring given file and directory names. Returns list of tuples containing both absolute and

Logarithm in base 2.

Download the given URL and return a binary-mode file object to access the data.

Register the gradients of the given loss function with respect to the given variables. Intended to be called once per GPU.

Re-initialize all trainable variables of this network, including sub-networks.

Re-initialize all variables of this network, including sub-networks.

Call FileWriter.add_summary() with all summaries in the default graph, automatically finalizing and merging them on the first call.

Set the global username override value.