Difference between revisions of "Wamesa/Transducer"

From LING073
Jump to: navigation, search
(Created page with "==Analyser== ===Evaluation=== My analyser covers <how many> stems. My main yaml file (wad.yaml) has 54/100 tests passing. Instances of infixation ({{tag|p2}}{{tag|sg}} and {{t...")
 
(Analyser)
Line 1: Line 1:
 
==Analyser==
 
==Analyser==
 
===Evaluation===
 
===Evaluation===
My analyser covers <how many> stems.
+
My analyser covers 32 stems.
 
My main yaml file (wad.yaml) has 54/100 tests passing. Instances of infixation ({{tag|p2}}{{tag|sg}} and {{tag|p3}}{{tag|sg}}) make up the vast majority of the tests that are not passing.
 
My main yaml file (wad.yaml) has 54/100 tests passing. Instances of infixation ({{tag|p2}}{{tag|sg}} and {{tag|p3}}{{tag|sg}}) make up the vast majority of the tests that are not passing.
 +
At first, my <code> aq-covtest </code> only covered 2.22% of my lexicon. The most egregious uncovered words were ''yau'' and ''pai'', and when I added these, coverage increased to 10.77%.
  
 
===Notes===
 
===Notes===
I couldn't run aq-covtest on my corpus because it took up all the RAM and CPU, so I don't know what the top unanalysed words are.
+
It is hard to meaningfully say what words my transducer covers, because most of the top uncovered words are actually morphemes, just ones that in some scholars' transcriptions are written as a separate word and in others' transcriptions are glommed onto the stem.
  
  

Revision as of 18:32, 22 February 2017

Analyser

Evaluation

My analyser covers 32 stems. My main yaml file (wad.yaml) has 54/100 tests passing. Instances of infixation (<p2><sg> and <p3><sg>) make up the vast majority of the tests that are not passing. At first, my aq-covtest only covered 2.22% of my lexicon. The most egregious uncovered words were yau and pai, and when I added these, coverage increased to 10.77%.

Notes

It is hard to meaningfully say what words my transducer covers, because most of the top uncovered words are actually morphemes, just ones that in some scholars' transcriptions are written as a separate word and in others' transcriptions are glommed onto the stem.