\ft\tiResult of Applying Noisy Harmonic Grammar to JesneySyllables.txt

\cn2-19-2017, 8:00 p.m.

\cnOTSoft 2.5, release date 1/17/2017

\hrNHG Results for JesneySyllables.txt	2-19-2017	\pn
\ks

\h1Weights Found
\ts2
Constraints	Weights Found
*Complex	1.000
*NoCoda	1.000
Max	1.000
\te\ke

\ks

\h1Matchup to Input Frequencies
\ts5
/CCVC/	Input Frequencies	Generated Frequencies	Input Number (exact)	Generated Number
CV	0.000	0.333	0	33263
CVC	0.000	0.166	0	16626
CCV	0.000	0.166	0	16572
CCVC	0.000	0.335	0	33539
\te\ke
\ts4\ks
/CCVC/	*Complex (1)	*NoCoda (1)	Max (1)
\wsF\we CV (0.333)			**
CVC (0.166)		* 	*
CCV (0.166)	* 		*
CCVC (0.335)	* 	*	
\te\ke
\ks

\h1Hasse Diagram
The following Hasse diagram (in output folder at JesneySyllablesHasse.gif) summarizes the rankings obtained.

\hf
\ke


\h1Testing the Grammar:  Details
The grammar was tested for 100000 cycles.
Average error per candidate:  6.956 percent
Learning time:  0.000 minutes


\h1Parameter Values Used

   Noise was set at 1.

   Negative weights were permitted.
   The number of times that a negative weight was recorded during learning and testing was 47157 out of 300000 chances.

   The number of ties between winner and rival was  3 out of 400000 chances.  Tie is within .00001.

   Noise was applied to constraints, hence the same for all candidates.

   Noise was applied before weights multiplied by violations.
\h2Initial weights
\ts3
Constraint	Initial Ranking	Final Ranking
*Complex	1.000	1.000
*NoCoda	1.000	1.000
Max	1.000	1.000
\te
\ke
\ks
\h2Schedule for GLA Parameters
\ts6
Stage	Trials	PlastMark	PlastFaith
1	0	1.000	1.000
2	0	0.100	0.100
3	0	0.010	0.010
4	0	0.001	0.001
\te
There were a total of 0 learning trials.
\ke
