DarkNet 시리즈 - LSTM Layer
lstm_layer LSTM Layer란? LSTM은 Long Short Term Memory networks의 약자입니다. RNN과 같이 자연어처리, 음성처리 등 Sequential 데이터를 처리하기 위해 많이 사용되는 layer입니다. 기존의 RNN은 학습하면서 점점 과거 정보를 잊어버리는(Gradient Vanishing) 문제가 발생하고 ...
lstm_layer LSTM Layer란? LSTM은 Long Short Term Memory networks의 약자입니다. RNN과 같이 자연어처리, 음성처리 등 Sequential 데이터를 처리하기 위해 많이 사용되는 layer입니다. 기존의 RNN은 학습하면서 점점 과거 정보를 잊어버리는(Gradient Vanishing) 문제가 발생하고 ...
logistic_layer forward_logistic_layer void forward_logistic_layer(const layer l, network net) { copy_cpu(l.outputs*l.batch, net.input, 1, l.output, 1); activate_array(l.output, l.outputs*...
local_layer local_out_height int local_out_height(local_layer l) { int h = l.h; if (!l.pad) h -= l.size; else h -= 1; return h/l.stride + 1; } 함수 이름: local_out_height 입력: ...
list // darknet.h typedef struct list{ int size; node *front; node *back; } list; 연결 리스트(list)의 구조체를 정의하는 코드입니다. 구조체 이름: list 구조체 멤버: size: 리스트에 저장된 노드의 수 fr...
layer // darknet.h typedef enum { CONVOLUTIONAL, DECONVOLUTIONAL, CONNECTED, MAXPOOL, SOFTMAX, DETECTION, DROPOUT, CROP, ROUTE, COST, NORMALIZATION, ...
iseg_layer instance segmentation을 위한 layer입니다. forward_iseg_layer void forward_iseg_layer(const layer l, network net) { double time = what_time_is_it_now(); int i,b,j,k; int ids = l...
image get_color float colors[6][3] = { {1,0,1}, {0,0,1},{0,1,1},{0,1,0},{1,1,0},{1,0,0} }; // BGR순서 float get_color(int c, int x, int max) { float ratio = ((float)x/max)*5; int i = floo...
image_opencv image_to_ipl #ifdef OPENCV using namespace cv; extern "C" { IplImage *image_to_ipl(image im) { int x,y,c; IplImage *disp = cvCreateImage(cvSize(im.w,im.h), IPL_DEPTH_8U, i...
im2col 이미지를 columns으로 변환해주는 것을 말합니다. im2col_get_pixel float im2col_get_pixel(float *im, int height, int width, int channels, int row, int col, int channel, int pad) {...
gru_layer GRU layer 란? GRU (Gated Recurrent Unit) 레이어는 반복 신경망 (Recurrent Neural Network, RNN)의 한 종류로, 긴 시퀀스를 처리하는 데에 사용됩니다. GRU는 기본적으로 LSTM (Long Short-Term Memory)과 유사한 아이디어를 기반으로 하고 있습니다. LSTM...