r"""Upsample a batch of 2D images with the given filter. Accepts a batch of 2D images of the shape `[N, C, H, W]` or `[N, H, W, C]` and upsam

Sum gradients across devices using NCCL ops (fast path).

Slow reference implementation of `fused_bias_act()` using standard TensorFlow ops.

Apply dynamic loss scaling for the given expression.

Compute pairwise distances between two batches of feature vectors.

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

r"""Fused `tf.nn.conv2d()` followed by `downsample_2d()`. Padding is performed only once at the beginning, not between the operations. The fu

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

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

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.

r"""Fused bias and activation function. Adds bias `b` to activation tensor `x`, evaluates activation function `act`, and scales the result by

Finds the python object with the given name.

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

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.

Calculates k-NN precision and recall for two sets of feature vectors.

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

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

Construct a TensorFlow op that updates the variables of this network to be slightly closer to those of the given network.

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

r"""Fused `upsample_2d()` followed by `tf.nn.conv2d()`. Padding is performed only once at the beginning, not between the operations. The fuse

Pickle export.