Matcher is too slow processing text

[sandeep118]

I built a matcher & phrase matcher object which has 10 lakh patterns. For some reason Matcher is taking very long time to process a document ( > 5 mins for each doc) and phrase matcher is doing it in a second on the same data.

I don't know if I am missing something here, but any help is much appreciated.

The reason I need to use Matcher instead of phrase matcher because:

Ex: I have a text file with many phrases such as "United State Of America"
I built phrase matcher object using the above. In the test document if there is phrase
"United State Of \n America" phrase matcher will not return a match because of the extra newline character. Whereas in Matcher I can customize the pattern by adding various things like {'ORTH':'\n','OP':'*'}.

Your Environment

• spaCy version: 2.0.5
• Platform: Darwin-17.2.0-x86_64-i386-64bit
• Python version: 3.6.0
• Models: en_core_web_lg

[ines]

Merging this with the master issue #1971!

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.