Top-level synchronous function that creates Examples. Provided for backwards compatibility, i.e. so that gr.Examples(...) can be used to create the Ex

generate samples from z given strategy Args: z: [batch, nsamples, nz] strategy: "beam" or "greedy" or "sample"

Converts a sequence of ids (integer) in a string, using the tokenizer and vocabulary with options to remove special tokens and clean

compute log p(x|z)

this function computes log q(z | x) Args: z: tensor different z points that will be evaluated, with

perform grid search to calculate the true posterior Args: zrange: tensor different z points that will be evaluated

Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).

perform sampling from inference net Returns: Tensor Tensor: samples from infernece nets with shape (batch_size, ns

Split a full s3 path into the bucket name and path.

Make sure we are sharing the input and output embeddings. Export to TorchScript can't handle parameter sharing so we are cloning them ins

Serializes this instance to a JSON string.

Performs invalid character removal and whitespace cleanup on text.

Converts a token (str/unicode) in an id using the vocab.

Checks whether CP is the codepoint of a CJK character.

Checks whether `chars` is a control character.

Checks whether `chars` is a punctuation character.

Checks whether `chars` is a whitespace character.

Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTraine

Splits punctuation on a piece of text.

Strips accents from a piece of text.

Adds whitespace around any CJK character.

Create a 1D, 2D, or 3D average pooling module.

Returns list of utf-8 byte and a mapping to unicode strings. We specifically avoids mapping to whitespace/control characters the bpe code bar

Returns: Tensor1, Tensor2 Tensor1: the tensor latent z with shape [batch, nsamples, nz] Tensor2: the tenor of KL for

Returns: Tensor1, Tensor2 Tensor1: the tensor latent z with shape [batch, nsamples, nz] Tensor2: the tenor of KL for

sampling from the encoder Returns: Tensor1 Tensor1: the tensor latent z with shape [batch, nsamples, nz]

perform grid search to calculate the true posterior this function computes p(z|x) Args: grid_z: tensor di

Constructs a `Config` from a Python dictionary of parameters.

Constructs a `BertConfig` from a json file of parameters.

Given a URL, look for the corresponding dataset in the local cache. If it's not there, download it. Then return the path to the cached file.

This matches the implementation in Denoising Diffusion Probabilistic Models: From Fairseq. Build sinusoidal embeddings. This matches

This matches the implementation in Denoising Diffusion Probabilistic Models: From Fairseq. Build sinusoidal embeddings. This matches