public static void encode_rows_cpu(IntPtr input, IntPtr output)//input float, output int
{
var m = (int)THWrapper.THFloatTensor_size(input, 0);
var n = (int)THWrapper.THFloatTensor_size(input, 1);
int l = 1 + (n - 1) / matmul.ENCODE_BIT;
THWrapper.THIntTensor_resize2d(output, m, l);
var test1 = THWrapper.THIntTensor_nDimension(output);
var dim0 = THWrapper.THIntTensor_size(output, 0);
var dim1 = THWrapper.THIntTensor_size(output, 1);
//THIntTensor_resize2d(output, m, l);
/*int m = input->size[0];
* int n = input->size[1];
* int l = 1 + (n - 1) / ENCODE_BIT;
*
* THIntTensor_resize2d(output, m, l);
* float* a = THFloatTensor_data(input);
* uint32_t* b = (uint32_t*)THIntTensor_data(output);
*/
var a = THWrapper.THFloatTensor_data(input);
var b = THWrapper.THIntTensor_data(output);
encode_rows_cpu_kernel(a, b, m, n);
//var temp = InternalArray.FromTHIntTensor(output);
}
public static void encode_cols_cpu(IntPtr input, IntPtr output)//input float, output int
{
/* int n = input->size[0];
* int k = input->size[1];
* int l = 1 + (n - 1) / ENCODE_BIT;
*
* THIntTensor_resize2d(output, l, k);
* float* a = THFloatTensor_data(input);
* uint32_t* b = (uint32_t*)THIntTensor_data(output);
*
* encode_cols_cpu_kernel(a, b, n, k);*/
var n = (int)THWrapper.THFloatTensor_size(input, 0);
var k = (int)THWrapper.THFloatTensor_size(input, 1);
int l = 1 + (n - 1) / matmul.ENCODE_BIT;
THWrapper.THIntTensor_resize2d(output, l, k);
var a = THWrapper.THFloatTensor_data(input);
var b = THWrapper.THIntTensor_data(output);
encode_cols_cpu_kernel(a, b, n, k);
//var res1 = BitConverter.ToUInt32(BitConverter.GetBytes(THWrapper.THIntTensor_get2d(output, 0, 7)), 0);
//var res2 = THWrapper.THIntTensor_get2d(output, 7, 0);
}
public static void fpbinary_gemm_cpu(IntPtr a, IntPtr b, IntPtr c, int m, int nn, int k, int transb, int beta, int alpha, IntPtr alphas)
{
if (THWrapper.THFloatTensor_nDimension(c) != 2 || THWrapper.THFloatTensor_size(c, 0) * THWrapper.THFloatTensor_size(c, 1) < m * k)
{
THWrapper.THFloatTensor_resize2d(c, m, k);
//THFloatTensor_resize2d(c, m, k);
}
/*
* uint32_t* A = (uint32_t*)THIntTensor_data(a);
* uint32_t* B = (uint32_t*)THIntTensor_data(b);
* float* C = THFloatTensor_data(c);
* float* D = THFloatTensor_data(alphas);
*/
var A = THWrapper.THIntTensor_data(a);
var B = THWrapper.THIntTensor_data(b);
var C = THWrapper.THFloatTensor_data(c);
var D = THWrapper.THFloatTensor_data(alphas);
var aa = InternalArray.FromTHIntTensor(a);
var bb = InternalArray.FromTHIntTensor(b);
var cc = InternalArray.FromTHFloatTensor(c);
var dd = InternalArray.FromTHFloatTensor(alphas);
int n = 1 + (nn - 1) / matmul.ENCODE_BIT;
int brow = transb != 0 ? 1 : k;
int bcol = transb != 0 ? n : 1;
//matmul.dgemm_nn(m, k, nn, A, n, 1, B, brow, bcol, C, k, 1, beta, alpha, D);
//matmul.dgemm_nn(m, k, nn, A, n, 1, B, brow, bcol, C, k, 1, beta, alpha, D);
matmul.fpdgemm_nn(m, k, nn, A, n, 1, B, brow, bcol, C, k, 1, beta);
if (alpha != 0)
{
for (int i = 0; i < m; i++)
{
for (int j = 0; j < k; j++)
{
//C[i * n + j] *= alphas[i];
var aa1 = matmul.GetFloat(C, i * k + j);
short aq1 = (short)(aa1 * 256);
var aa2 = matmul.GetFloat(D, i);
var aq2 = (short)(aa2 * 256);
var val4 = (short)((int)(aq1 * aq2) >> 8);
var quant_res = val4 / 256f;
var orig = aa1 * aa2;
//matmul.SetFloat(C, i * k + j, aa1 * aa2);
matmul.SetFloat(C, i * k + j, val4);
//C[i * n + j] = (float)(C[i * n + j] * alphas[i]);
}
}
}
}
