trf-feature.h

Go to the documentation of this file.

// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// Copyright 2014-2015 Tsinghua University
// Author: wb.th08@gmail.com (Bin Wang), ozj@tsinghua.edu.cn (Zhijian Ou)
//
// All h, cpp, cc, and script files (e.g. bat, sh, pl, py) should include the above
// license declaration. Different coding language may use different comment styles.


#pragma once
#include "trf-vocab.h"
#include <omp.h>

namespace trf
{
    const int c_nMaxOrder = 100;

    class Seq;
    class FeatStyle;
    class FeatTable;
    class Feat;

#define word_layer 0
#define class_layer 1

    class Seq
    {
    public:
        Mat<VocabID> x;
        int nLen;
        int nMaxLen; 
    public:
        Seq() :nLen(0),nMaxLen(0) {}
        Seq(int len) :nLen(0),nMaxLen(0) { Reset(len); }
        void Reset(int p_len) {
            if (p_len > nMaxLen) {
                Mat<VocabID> newx(2, p_len);
                newx.Fill(0);
                for (int i = 0; i < nMaxLen; i++) {
                    newx[0][i] = x[0][i];
                    newx[1][i] = x[1][i];
                }
                x.Copy(newx);
                nMaxLen = p_len;
            }

            nLen = p_len;
        }
        void Copy(Seq &seq) {
            x.Copy(seq.x);
            nLen = seq.nLen;
            nMaxLen = seq.nMaxLen;
        }
        int GetLen() const { return nLen; }
        void Set(Array<int> &aInt, Vocab *pv);
        void Set(int *pInt, int nLen, Vocab *pv);
        void Random(Vocab *pv);
        void SetClass(Vocab *pv);
        VocabID *GetWordSeq() { return x[word_layer].GetBuf(); }
        VocabID* GetClassSeq() { return x[class_layer].GetBuf(); }
        void Print();
        void Print(File &file);
    };

    class FeatStyle
    {
    public:
        typedef struct {
            char c;  
            short i; 
            short n; 
        } Field;
    public:
        int m_nOrder; 
        Array<Field> m_aFields; 
    public:
        FeatStyle() :m_nOrder(0) {}
        FeatStyle(const char *pstyle) { Set(pstyle); }
        void Set(const char *pstyle);
        bool GetKey(Seq &seq, int nPos, int nOrder, Array<int> &key);
    };
    class FeatTable
    {
    public:
        String m_style; 
        int m_nMinOrder; 
        int m_nMaxOrder; 
        int m_nNum; 
        Array<int> m_aCutoff; 
        Trie<VocabID, int> *m_ptrie; 

        Array<FeatStyle*> m_aStyleInfo;

        Array<Array<int> *> m_aKey; 

    public:
        FeatTable(const char *pstyle = "");
        ~FeatTable();
        void Reset(const char *pstyle);
        int GetNum() const { return m_nNum; }
        int GetMinOrder() const { return m_nMinOrder; }
        int GetMaxOrder() const { return m_nMaxOrder; }
        const char *GetStyle() const { return m_style.GetBuffer(); }
        void LoadFeat(Seq &seq);
        void IndexFeat(int begIdx);
        int CutoffFeat();
        void Find(Array<int> &afeat, Array<int> &key, bool bBeg, bool bEnd);
        void Find(Array<int> &afeat, Seq &seq, int pos, int order);
        void Find(Array<int> &afeat, Seq &seq);
        void FindPosDep(Array<int> &afeat, Seq &seq, int pos);

        void ReadT(File &file, PValue *pValue = NULL);
        void WriteT(File &file, PValue *pValue = NULL);
//      /// Read from binary
//      void ReadB(File &file);
//      /// Write to binary
//      void WriteB(File &file);
    };

    class Feat
    {
    public:
        Array<FeatTable*> m_aTable; 
        int m_nTotalNum; 
    public:
        Feat(int nOrder = 0, bool bClass = true)
        {
            m_nTotalNum = 0;
            if (nOrder > 0) {
                Reset(nOrder, bClass);
            }
        }
        ~Feat() { SAFE_DEL_POINTER_ARRAY(m_aTable); }
        void Reset(int nOrder, bool bClass);
        void Reset(const char *pfeatType);
        int GetMaxOrder();
        int GetNum() const { return m_nTotalNum;  }
        void Find(Array<int> &afeat, Seq &seq, int pos, int order);
        void Find(Array<int> &afeat, Seq &seq);
        void FindClass(Array<int> &afeat, Seq &seq, int pos, int order);
        void FindWord(Array<int> &afeat, Seq &seq, int pos, int order);
        void FindPosDep(Array<int> &afeat, Seq &seq, int pos, int type = 0);
        void LoadFeatFromCorpus(const char *path, Vocab *pv);
        void WriteT(File &file, PValue *pValue = NULL);
        void ReadT(File &file, PValue *pValue = NULL);
    };
}