/* eslint-disable */ /** * @file 主函数 * @author chenhaoze */ export default ` // start函数 void main(void) { ivec4 oPos = getOutputTensorPosLIMIT_OUT(); int x = oPos.a; int c = oPos.g; int y = oPos.b; int b = oPos.r; float res = 0.0; int temp_x = 0; int temp_y = 0; float o = 0.0; float f = 0.0; if (x % 2 == 1) x = x - 2; if (y % 2 == 1) y = y - 2; // 重排遍历顺序 //int sumVal = oPos.g + oPos.a * channel_out + oPos.b * channel_out * width_shape_out; //int new_a = sumVal % width_shape_out; //int new_b = int((sumVal - new_a) / width_shape_out) % height_shape_out; //int new_g = int((((sumVal - new_a) / width_shape_out) - new_b) / height_shape_out); //int x = new_a; //int c = new_g; //int y = new_b; // 获取output的坐标 int oTensorChannel = int(c * groups / channel_out) * channel_origin; int oy = y; for (int fy = 0; fy < height_shape_filter; fy++) { if (oy < 0) { oy += dilation_v; continue; } int ox = x; for (int fx = 0; fx < width_shape_filter; fx++) { if (ox < 0) { ox += dilation_h; continue; } // channel计算 for (int j = 0; j < channel_origin; j++) { if (ox % stride_h == 0 && oy % stride_v == 0) { temp_x = int(floor(float(ox) / float(stride_h))); temp_y = int(floor(float(oy) / float(stride_v))); if (temp_x < width_shape_origin && temp_y < height_shape_origin){ o = getValueFromTensorPosLIMIT_ORIGIN_origin(b, j, temp_y, temp_x); f = getValueFromTensorPosLIMIT_FILTER_filter(j, c, fy, fx); res += f * o; } } } ox += dilation_h; } oy += dilation_v; } setOutput(float(res)); } `;