diff options
Diffstat (limited to 'native/jni/src')
| -rw-r--r-- | native/jni/src/correction.cpp | 3 | ||||
| -rw-r--r-- | native/jni/src/defines.h | 2 | ||||
| -rw-r--r-- | native/jni/src/unigram_dictionary.cpp | 9 |
3 files changed, 11 insertions, 3 deletions
diff --git a/native/jni/src/correction.cpp b/native/jni/src/correction.cpp index 5ae34cd02..f7ef7efc0 100644 --- a/native/jni/src/correction.cpp +++ b/native/jni/src/correction.cpp @@ -55,6 +55,7 @@ inline static void dumpEditDistance10ForDebug(int *editDistanceTable, } AKLOGI("[ %d, %d, %d, %d, %d, %d, %d, %d, %d, %d, %d ]", c[0], c[1], c[2], c[3], c[4], c[5], c[6], c[7], c[8], c[9], c[10]); + (void)c; } } } @@ -977,7 +978,7 @@ int Correction::RankingAlgorithm::calcFreqForSplitMultipleWords( } const int freq = freqArray[i]; // Demote too short weak words - if (wordLength <= 4 && freq <= MAX_FREQ * 2 / 3 /* heuristic... */) { + if (wordLength <= 4 && freq <= SUPPRESS_SHORT_MULTIPLE_WORDS_THRESHOLD_FREQ) { multiplyRate(100 * freq / MAX_FREQ, &totalFreq); } if (wordLength == 1) { diff --git a/native/jni/src/defines.h b/native/jni/src/defines.h index dfc5238a0..19f843446 100644 --- a/native/jni/src/defines.h +++ b/native/jni/src/defines.h @@ -228,6 +228,8 @@ static inline void prof_out(void) { #define TWO_WORDS_CORRECTION_WITH_OTHER_ERROR_THRESHOLD 0.35 #define START_TWO_WORDS_CORRECTION_THRESHOLD 0.185 +/* heuristic... This should be changed if we change the unit of the frequency. */ +#define SUPPRESS_SHORT_MULTIPLE_WORDS_THRESHOLD_FREQ (MAX_FREQ * 58 / 100) #define MAX_DEPTH_MULTIPLIER 3 diff --git a/native/jni/src/unigram_dictionary.cpp b/native/jni/src/unigram_dictionary.cpp index 3c826e918..8d2f4cd73 100644 --- a/native/jni/src/unigram_dictionary.cpp +++ b/native/jni/src/unigram_dictionary.cpp @@ -222,6 +222,7 @@ int UnigramDictionary::getSuggestions(ProximityInfo *proximityInfo, short unsigned int* w = outWords + j * MAX_WORD_LENGTH; char s[MAX_WORD_LENGTH]; for (int i = 0; i <= MAX_WORD_LENGTH; i++) s[i] = w[i]; + (void)s; AKLOGI("%s %i", s, frequencies[j]); } } @@ -503,8 +504,12 @@ bool UnigramDictionary::getSubStringSuggestion( freqArray, wordLengthArray, currentWordIndex + 1, isSpaceProximity, outputWord); if (DEBUG_DICT) { DUMP_WORD(outputWord, tempOutputWordLength); - AKLOGI("Split two words: %d, %d, %d, %d, (%d) %d", freqArray[0], freqArray[1], pairFreq, - inputLength, wordLengthArray[0], tempOutputWordLength); + for (int i = 0; i < currentWordIndex + 1; ++i) { + AKLOGI("Split %d,%d words: freq = %d, length = %d", i, currentWordIndex + 1, + freqArray[i], wordLengthArray[i]); + } + AKLOGI("Split two words: freq = %d, length = %d, %d, isSpace ? %d", pairFreq, + inputLength, tempOutputWordLength, isSpaceProximity); } addWord(outputWord, tempOutputWordLength, pairFreq, queuePool->getMasterQueue()); } |