Update Readme to reflect recent perf improvements
This commit is contained in:
parent
3a61fd2c55
commit
ae858ead5c
|
@ -14,7 +14,7 @@ which is in turn based on code from Peter Norvig's chapter [Natural Language
|
||||||
Corpus Data][chapter] from the book [Beautiful Data][book] (Segaran and
|
Corpus Data][chapter] from the book [Beautiful Data][book] (Segaran and
|
||||||
Hammerbacher, 2009).
|
Hammerbacher, 2009).
|
||||||
|
|
||||||
For the microbenchmark included in this repository, Instant Segment is ~100x
|
For the microbenchmark included in this repository, Instant Segment is ~500x
|
||||||
faster than the Python implementation. The API was carefully constructed
|
faster than the Python implementation. The API was carefully constructed
|
||||||
so that multiple segmentations can share the underlying state to allow parallel
|
so that multiple segmentations can share the underlying state to allow parallel
|
||||||
usage.
|
usage.
|
||||||
|
|
Loading…
Reference in New Issue