Given a string of `text`, return a generator yielding each hypothesized sentence string

Given a `text` string, get candidates and context for feature extraction and classification

Given a `filename` string, slurp the whole file into a string

Given left context `L`, punctuation mark `P`, and right context R`, extract features. Probability distributions for any quant

Given an left context `L`, punctuation mark `P`, and right context `R`, return True iff this observation is hypothesized to be a

Split string `text` into word tokens using the Penn Treebank rules

Return a Detector loaded from the default model. Currently, the default model is trained on WSJ.

Given a string of `text`, compute confusion matrix for the classification task.

Given a string `text`, use it to train the segmentation classifier for `epochs` iterations.

Return a Detector loaded from resource with the specified name. The model name must match a filename existing under /models in this pack

A single sentence is segmented as a single sentence.

Strip non-initial whitespace by default.

Preserve all whitespace when strip=False.

Two sentences joined by newline are segmented as two sentences.

Two sentences joined by space are segmented as two sentences.