#ifndef SEGMENT_CONV_INC_CPP

#define SEGMENT_CONV_INC_CPP

#include "segment.inc.cpp"

#include "func.inc.cpp"

#include "layer.conv.inc.cpp"

class SegmentConv: public Segment {

public:

  enum {

    KSX = 4,

    KSY = 4,

  };

  const int msx, msy, msz;

  NeuronReal *m_values;

  NeuronReal *b_values;

  SegmentConv(int sx, int sy, int sz, int msz, Weight *weights = nullptr):

    Segment(sx, sy, sz, msz*KSY*KSX*sz, weights), msx((sx - KSX)/2 + 1), msy((sy - KSY)/2 + 1), msz(msz)

  {

    assert(msx > 0);

    assert(msy > 0);

    assert(msz > 0);

    m_values = new NeuronReal[msx*msy*msz + sx*sy*sz];

    b_values = m_values + msx*msy*msz;

    clear();

  }

  ~SegmentConv()

    { delete[] m_values; }  

  void clear() override

    { memset(m_values, 0, sizeof(*m_values)*(msx*msy*msz + sx*sy*sz)); }

  inline void check(int x, int y, int z) {

    Segment::check(x, y, z);

    assert(layout.getD() == sz);

  }

  Quality pass(Barrier &barrier, int x, int y, int z, NeuronReal trainRatio) override {

    check(x, y, z);

    Layout l = layout;

    NeuronReal *f_values = this->f_values + (y*l.sx + x)*l.sz + z;

  }

  __attribute__((always_inline))

  inline Quality pass(Barrier &barrier, NeuronReal *f_values, NeuronReal trainRatio) {

    Layout l = layout;

    int tid = barrier.tid;

    int threads = barrier.threads;

    int sx = this->sx;

    //int sy = this->sy;

    int sz = this->sz;

    int msx = this->msx;

    int msy = this->msy;

    int msz = this->msz;

    int msxz = msx*msz;

    int ksxz = KSX*sz;

    int ksyxz = KSY*ksxz;

    int fv_dkx = l.sz - sz;

    int fv_dky = (l.sx - KSX)*l.sz;

    int fv_dmx = 2*l.sz;

    int fv_dmy = 2*(l.sx - msx)*l.sz;

    int mn_dtz = threads - msx*msy*msz;

    // stage 1: pass from front to mid

    int f_sxz = l.sx*l.sz;

    int f_sz2 = l.sz*2;

    int f_sxz2 = l.sx*f_sz2;

    int m_cnt = msx*msy*msz;

    int mi0 = m_cnt*tid/threads;

    int mi1 = m_cnt*(tid+1)/threads;

    for(int i = mi0; i < mi1; ++i) {

      int my = i/msxz;

      int mx = i%msxz/msz;

      int mz = i%msz;

      AccumReal a = 0;

      int wi = i*ksyxz;

      int fvi = my*f_sxz2 + mx*f_sz2 + mz;

      for(int ky = 0; ky < KSY; ++ky, fvi += f_sxz, wi += ksxz) {

        Weight *iw = &weights[wi];

        NeuronReal *ifv = &f_values[fvi];

        for(int i = 0; i < ksxz; ++i)

          a += ifv[i]*iw[i].w;

      }

      m_values[i] = a > 0 ? a : 0;

    }

    barrier.wait();

    // stage 2: pass from mid to back and verify

    AccumReal qa = 0;

    for(int by = 2 + tid; by < 10; by += threads)

    for(int bx = 2; bx < 10; ++bx)

    for(int bz = 0; bz < sz; ++bz) {

      AccumReal a = 0;

      Neuron &bn = b_neurons[ (by*sx + bx)*sz + bz ];

      for(int ky = by%2; ky < KSY; ky += 2)

      for(int kx = bx%2; kx < KSX; kx += 2) {

        int mx = (bx - kx)/2;

        int my = (by - ky)/2;

        assert(mx >= 0 && mx < msx && (bx - kx)%2 == 0);

        assert(my >= 0 && my < msy && (by - ky)%2 == 0);

        for(int mz = 0; mz < msz; ++mz) {

          Neuron &mn = m_neurons[ (my*msx + mx)*msz + mz ];

          Weight &w = weights[ ((mz*ksy + ky)*ksx + kx)*sz + bz ];

          a += mn.v * w.w;

        }

      }

      if (a > 0) bn.v = a, bn.d = 1; else bn.v = bn.d = 0;

      NeuronReal fn = f_values[ (by*l.sx + bx)*l.sz + bz ];

      NeuronReal d = fn - bn.v;

      bn.d *= d*trainRatio;

      qa += d*d;

    }

    Quality q(qa/(64*sz));

    if (trainRatio <= 0) return q;

    barrier.wait();

    // stage 3: backpass deltas

    for(int mz = tid; mz < msz; mz += threads)

    for(int my = 1; my < 4; ++my)

    for(int mx = 1; mx < 4; ++mx) {

      AccumReal a = 0;

      Neuron &mn = m_neurons[ (my*msx + mx)*msz + mz ];

      for(int ky = 0; ky < ksy; ++ky)

      for(int kx = 0; kx < ksx; ++kx)

      for(int kz = 0; kz < sz;  ++kz) {

        int bx = mx*2 + kx;

        int by = my*2 + ky;

        Neuron &bn = b_neurons[ (by*sx + bx)*sz + kz ];

        Weight &w = weights[ ((mz*ksy + ky)*ksx + kx)*sz + kz ];

        a += bn.d * w.w;

      }

      mn.d *= a;

    }

    barrier.wait();

    // stage 4: update weights

    for(int mz = tid; mz < msz; mz += threads)

    for(int by = 4; by <  8; ++by)

    for(int bx = 4; bx <  8; ++bx)

    for(int bz = 0; bz < sz; ++bz) {

      Neuron &bn = b_neurons[ (by*sx + bx)*sz + bz ];

      NeuronReal fv = f_values[ (by*l.sx + bx)*l.sz + bz ];

      for(int ky = by%2; ky < ksy; ky += 2)

      for(int kx = bx%2; kx < ksx; kx += 2) {

        int mx = (bx - kx)/2;

        int my = (by - ky)/2;

        assert(mx >= 1 && mx < 4 && (bx - kx)%2 == 0);

        assert(my >= 1 && my < 4 && (by - ky)%2 == 0);

        Neuron &mn = m_neurons[ (my*msx + mx)*msz + mz ];

        Weight &w = weights[ ((mz*ksy + ky)*ksx + kx)*sz + bz ];

        w.w += bn.d*mn.v + mn.d*fv;

      }

    }

    return q;

  }

  Quality testPass(int x, int y, int z, NeuronReal trainRatio) override {

    check(x, y, z);

    Layout l = layout;

    // stage 1: pass

    clear();

    for(int my = 0; my < msy; ++my)

    for(int mx = 0; mx < msx; ++mx)

    for(int mz = 0; mz < msz; ++mz) {

      AccumReal a = 0;

      for(int ky = 0; ky < KSY; ++ky)

      for(int kx = 0; kx < KSX; ++kx)

      for(int kz = 0; kz < sz;  ++kz) {

        int fx = x + mx*2 + kx;

        int fy = y + my*2 + ky;

        int fz = z + kz;

        NeuronReal fv = f_values[ (fy*l.sx + fx)*l.sz + fz ];

        Weight &w = weights[ ((mz*KSY + ky)*KSX + kx)*sz + kz ];

        a += fv * w.w;

      }

      NeuronReal &mv = m_values[ (my*msx + mx)*msz + mz ];

      if (a < 0) { mv = 0; continue; }

      mv = a;

      for(int ky = 0; ky < KSY; ++ky)

      for(int kx = 0; kx < KSX; ++kx)

      for(int kz = 0; kz < sz;  ++kz) {

        int bx = mx*2 + kx;

        int by = my*2 + ky;

        int bz = kz;

        NeuronReal &bv = b_values[ (by*sx + bx)*sz + bz ];

        Weight &w = weights[ ((mz*KSY + ky)*KSX + kx)*sz + kz ];

        bv += a * w.w;

      }

    }

    // stage 2: finalize values and verify

    AccumReal qa = 0;

    for(int by = 0; by < sy; ++by)

    for(int bx = 0; bx < sx; ++bx)

    for(int bz = 0; bz < sz; ++bz) {

        NeuronReal fn = f_values[ ((y + by)*l.sx + x + bx)*l.sz + z + bz ];

        NeuronReal &bv = b_values[ (by*sx + bx)*sz + bz ];

        if (bv > 0) {

          NeuronReal d = fn - bv;

          bv = d*trainRatio;

          qa += d*d;

        } else {

          bv = 0;

          qa += fn*fn;

        }

    }

    Quality q(qa/(KSX*KSY*sz));

    if (trainRatio <= 0) return q;

    // stage 3: backpass deltas and update weights

    for(int my = 0; my < msy; ++my)

    for(int mx = 0; mx < msx; ++mx)

    for(int mz = 0; mz < msz; ++mz) {

      NeuronReal mv = m_values[ (my*msx + mx)*msz + mz ];

      if (!mv) continue;

      AccumReal a = 0;

      for(int ky = 0; ky < KSY; ++ky)

      for(int kx = 0; kx < KSX; ++kx)

      for(int kz = 0; kz < sz;  ++kz) {

        int bx = mx*2 + kx;

        int by = my*2 + ky;

        int bz = kz;

        NeuronReal bv = b_values[ (by*sx + bx)*sz + bz ];

        Weight &w = weights[ ((mz*KSY + ky)*KSX + kx)*sz + kz ];

        a += bv * w.w;

      }

      for(int ky = 0; ky < KSY; ++ky)

      for(int kx = 0; kx < KSX; ++kx)

      for(int kz = 0; kz < sz;  ++kz) {

        int bx = mx*2 + kx;

        int by = my*2 + ky;

        int bz = kz;

        NeuronReal fv = f_values[ ((y + by)*l.sx + x + bx)*l.sz + z + bz ];

        NeuronReal bv = b_values[ (by*sx + bx)*sz + bz ];

        Weight &w = weights[ ((mz*KSY + ky)*KSX + kx)*sz + kz ];

        w.w += bv*mv + fv*a;

      }

    }

    return q;

  }

  bool saveDemo() override

    { return !filename || saveConvDemoImage(filename, msz, KSX, KSY, sz, weights); }

};

#endif