Home | History | Annotate | Download | only in src 
      1 /*
      2  * Copyright (C) 2013 The Android Open Source Project
      3  *
      4  * Licensed under the Apache License, Version 2.0 (the "License");
      5  * you may not use this file except in compliance with the License.
      6  * You may obtain a copy of the License at
      7  *
      8  *      http://www.apache.org/licenses/LICENSE-2.0
      9  *
     10  * Unless required by applicable law or agreed to in writing, software
     11  * distributed under the License is distributed on an "AS IS" BASIS,
     12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     13  * See the License for the specific language governing permissions and
     14  * limitations under the License.
     15  */
     16 
     17 #ifndef LATINIME_MULTI_BIGRAM_MAP_H
     18 #define LATINIME_MULTI_BIGRAM_MAP_H
     19 
     20 #include <cstring>
     21 #include <stdint.h>
     22 
     23 #include "defines.h"
     24 #include "binary_format.h"
     25 #include "hash_map_compat.h"
     26 
     27 namespace latinime {
     28 
     29 // Class for caching bigram maps for multiple previous word contexts. This is useful since the
     30 // algorithm needs to look up the set of bigrams for every word pair that occurs in every
     31 // multi-word suggestion.
     32 class MultiBigramMap {
     33  public:
     34     MultiBigramMap() : mBigramMaps() {}
     35     ~MultiBigramMap() {}
     36 
     37     // Look up the bigram probability for the given word pair from the cached bigram maps.
     38     // Also caches the bigrams if there is space remaining and they have not been cached already.
     39     int getBigramProbability(const uint8_t *const dicRoot, const int wordPosition,
     40             const int nextWordPosition, const int unigramProbability) {
     41         hash_map_compat<int, BigramMap>::const_iterator mapPosition =
     42                 mBigramMaps.find(wordPosition);
     43         if (mapPosition != mBigramMaps.end()) {
     44             return mapPosition->second.getBigramProbability(nextWordPosition, unigramProbability);
     45         }
     46         if (mBigramMaps.size() < MAX_CACHED_PREV_WORDS_IN_BIGRAM_MAP) {
     47             addBigramsForWordPosition(dicRoot, wordPosition);
     48             return mBigramMaps[wordPosition].getBigramProbability(
     49                     nextWordPosition, unigramProbability);
     50         }
     51         return BinaryFormat::getBigramProbability(
     52                 dicRoot, wordPosition, nextWordPosition, unigramProbability);
     53     }
     54 
     55     void clear() {
     56         mBigramMaps.clear();
     57     }
     58 
     59  private:
     60     DISALLOW_COPY_AND_ASSIGN(MultiBigramMap);
     61 
     62     class BigramMap {
     63      public:
     64         BigramMap() : mBigramMap(DEFAULT_HASH_MAP_SIZE_FOR_EACH_BIGRAM_MAP) {}
     65         ~BigramMap() {}
     66 
     67         void init(const uint8_t *const dicRoot, int position) {
     68             BinaryFormat::fillBigramProbabilityToHashMap(dicRoot, position, &mBigramMap);
     69         }
     70 
     71         inline int getBigramProbability(const int nextWordPosition, const int unigramProbability)
     72                 const {
     73            return BinaryFormat::getBigramProbabilityFromHashMap(
     74                    nextWordPosition, &mBigramMap, unigramProbability);
     75         }
     76 
     77      private:
     78         // Note: Default copy constructor needed for use in hash_map.
     79         hash_map_compat<int, int> mBigramMap;
     80     };
     81 
     82     void addBigramsForWordPosition(const uint8_t *const dicRoot, const int position) {
     83         mBigramMaps[position].init(dicRoot, position);
     84     }
     85 
     86     hash_map_compat<int, BigramMap> mBigramMaps;
     87 };
     88 } // namespace latinime
     89 #endif // LATINIME_MULTI_BIGRAM_MAP_H
     90