def cleanUnigramCountFile(inputfile, outputfile, n):
'''Filter the unigram count file, and reduce the number of items in it.'''
ngrok.cleanUnigramCountFile(inputfile, outputfile, n)
[opus.processLanguage(x, directories['expandpath'], directories['combinedpath']) for x in languages]
[opus.combineLanguage(os.path.join(directories['combinedpath'], x), os.path.join(directories['intermediatecountpath'],x+'_combined.txt')) for x in languages]
for language in languages:
print('Counting ngrams...')
countfile = os.path.join(directories['intermediatecountpath'],language+'_counted.txt')
ngrok.countNgrams(os.path.join(directories['intermediatecountpath'],language+'_combined.txt'), countfile, n=1)
print('Rearranging ngrams...')
rearrangedFile = os.path.join(directories['intermediatecountpath'],language+'_2013_rearrange_counts.txt')
ngrok.rearrangeNgramFile(countfile, rearrangedFile , reverse=False)
print('Cleaning ngrams...')
cleanedFile = os.path.join(directories['intermediatecountpath'],language+'_2013_cleaned.txt')
ngrok.cleanUnigramCountFile(rearrangedFile, cleanedFile, numItems, language)
print('Sorting ngrams...')
sortedfile = os.path.join(directories['intermediatecountpath'],language+'_2013_sorted.txt')
ngrok.sortNgramFile(cleanedFile, sortedfile)
print('Collapsing ngrams...')
collapsedfile = os.path.join(directories['intermediatecountpath'],language+'_2013_collapsed.txt')
ngrok.collapseNgrams(sortedfile, collapsedfile)
print('Marginalizing ngrams, first pass...')
ngrok.marginalizeNgramFile(collapsedfile, os.path.join(slowstoragedir,language+'_2013.txt'), n=1, sorttype='numeric')
def cleanUnigramCountFile(inputfile, outputfile, n): '''Filter the unigram count file, and reduce the number of items in it.''' ngrok.cleanUnigramCountFile(inputfile, outputfile, n)
