Library API¶

For quick reference these are a list of all provided functions, grouped by which reference page they are on.

`signatory.signature`	Applies the signature transform to a stream of data.
`signatory.Signature`	`torch.nn.Module` wrapper around the `signatory.signature()` function.
`signatory.signature_channels`	Computes the number of output channels from a signature call.
`signatory.extract_signature_term`	Extracts a particular term from a signature.
`signatory.signature_combine`	Combines two signatures into a single signature.
`signatory.multi_signature_combine`	Combines multiple signatures into a single signature.

`signatory.logsignature`	Applies the logsignature transform to a stream of data.
`signatory.LogSignature`	`torch.nn.Module` wrapper around the `signatory.logsignature()` function.
`signatory.logsignature_channels`	Computes the number of output channels from a logsignature call with `mode in ("words", "brackets")`.
`signatory.signature_to_logsignature`	Calculates the logsignature corresponding to a signature.
`signatory.SignatureToLogSignature`	`torch.nn.Module` wrapper around the `signatory.signature_to_logsignature()` function.

Calculates signatures and logsignatures on intervals of an input path.

Invert the signature with the insertion algorithm: reconstruct a stream of data given its signature.

`signatory.Augment`	Augmenting a stream of data before feeding it into a signature is often useful; the hope is to obtain higher-order information in the signature.
`signatory.all_words`	Computes the collection of all words up to length `depth` in an alphabet of size `channels`.
`signatory.lyndon_words`	Computes the collection of all Lyndon words up to length `depth` in an alphabet of size `channels`.
`signatory.lyndon_brackets`	Computes the collection of all Lyndon words, in their standard bracketing, up to length `depth` in an alphabet of size `channels`.

Reference pages