Constructor. :param dict[str, any] config: configuration following the following format: {

Computes the optimal partitions given the size distributions and computed number of expected false positives for all sub-intervals. Args:

Computes the expected number of false positives caused by using u to approximate set sizes in the interval [l, u], using the real set size dis

Computes the expected number of false positives caused by using u to approximate set sizes in the interval [l, u], assuming uniform distributi

Computes the matrix of expected false positives for all possible sub-intervals of the complete domain of set sizes. Args: counts: the

Download pre-trained common-crawl vectors from fastText's website https://fasttext.cc/docs/en/crawl-vectors.html Parameters

Download pre-trained GloVe vectors from Stanford's website https://fasttext.cc/docs/en/crawl-vectors.html Parameters

Download the pre-trained model file. Parameters ---------- model_name : str The embedding model name to d

Download the pre-trained model file. Parameters ---------- if_exists : str Supported values:

Compute the false positive and false negative probabilities given the LSH parameters. Parameters ---------- b : int

Determine the values for LSH parameters b and r that minimize the probability of false negatives (guarantees high recall). No

Compute the optimal `MinHashLSH` parameter that minimizes the weighted sum of probabilities of false positive and false negative.

Parse a configuration dictionary, optionally fetching data from env variables. :param dict[str, any] config: the configurati