NSketch

Due to current NDoc limited support for .Net 2.0, the generic types have been omitted from this documentation.

The NSketch library contains several sketch-based algorithms. There are two main motivations in building specific sketch-based algorithm:

The main tradeoff of the sketched algorithms is that the results are only approximate.

This library contains several sketch-based algorithms:

Histograms (see IHistogram) to be used as compact real distribution representations.
Frequent item selection algorithm (see IFrequentItemCollection`1).
Bloom filters (see BloomFilter`1).
General purpose fast hash function (see FastHashFunction).

Class	Description
ExpHistogram	The exponential histogram is high-streaming-performance approximated variant of the streaming histogram.
FastHashFunction	This class provides a fast hashing function.
NaiveHistogram	The class NaiveHistogram provides an implementation of the naive histograms where all the bucket have the same length.
SechapHistogram	Streamed Equi-Cost Approximate Histogram (SECHAP)

Interface	Description
IBucket	The interface IBucket represents an histogram bucket.
IHistogram	The interface IHistogram defines the method required for the streaming histogram algorithm implementation.

NSketch Namespace