Closed mmcauliffe closed 10 years ago
There is somewhat of a point in that if someone wants to input spelling, have the system automatically find the transcription, and then output the phonological edit distance on that.
You're right in that doing phonological edit distance on actual spelling makes no sense whatsoever. I'll add some error handling to deal with when there are no transcriptions available
Can't we just grey out the option for "spelling" for "phonological" edit distance? They're always entering the spelling, even if they're using a transcription-based algorithm.
We could - but I thinking about it, are you ever going to want to calculate the frequency of alternation on orthography anyways? It might make more sense to just remove the option of using freq_of_alt on spelling whatsoever, and only allow it with transcrpition
but thinking about it\ not "but I thinking about it"
I get the same error as Michael M. whenever I try to calculate F. of A. with phonological edit distance.
Yes, if phonological alignment goes in to the determination of frequency of alternation, then we're never using spelling, so that can be taken out.
Updated the scripts so they no longer call errors. The problem was caused by a mistake in getting the transcription for display purposes. It should be all cleared up now
When running frequency of alternations using Phonological edit distance, selecting "Compare transcription" instead of "Compare spelling" causes the following error:
Exception in Tkinter callback Traceback (most recent call last): File "C:\Python34\lib\tkinterinit.py", line 1487, in call return self.func(_args) File "C:\Users\SpeechInContext\Documents\GitHub\CorpusTools\corpustools\gui\fagui.py", line 148, in calculate_freq_of_alt min_rel=min_rel, max_rel=max_rel, min_pairs_okay=min_pairs_ok) File "C:\Users\SpeechInContext\Documents\GitHub\CorpusTools\corpustools\freqalt\freq_of_alt.py", line 92, in calc_freq_of_alt alignment = al.align(w1, w2) File "C:\Users\SpeechInContext\Documents\GitHub\CorpusTools\corpustools\symbolsim\phono_align_ex.py", line 32, in align similarity_matrix = self.make_similarity_matrix(seq1, seq2) File "C:\Users\SpeechInContext\Documents\GitHub\CorpusTools\corpustools\symbolsim\phono_align_ex.py", line 64, in make_similarity_matrix d[0][y]['f'] = d[0][y-1]['f'] + self.compare_segments('empty', seq2[y-1]) File "C:\Users\SpeechInContext\Documents\GitHub\CorpusTools\corpustools\symbolsim\phono_align_ex.py", line 113, in comparesegments fs2 = self.features[segment2symbol] KeyError: ''
Should there even be the choice of doing phonological edit distance on spelling? I'm assuming that it uses the Transcription anyway.