Programmer Think

1 Overview This document is mainly composed of Two classes, Darknet yolayer,Several functions create_modules，get_yolo_layers，load_darknet_weights，save_weights，convert，attempt_download form Darknet's constructor first calls parse_model_cfg，create_modules，get_yolo_layers are three functions, so let's start with them and then further ...

preface [personal learning notes, please correct any errors] YOLO-V5 code warehouse address: https://github.com/ultralytics/yolov5 1, Conv module Before introducing each module, you need to introduce the most basic Conv module in YOLOV5. class Conv(nn.Module): def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_i ...

preface The article is linked to: Detailed analysis of YOLO-V3-SPP This paper mainly explains the processing of YoloV3-SPP in the verification reasoning stage, which is divided into the following points: Preprocessing dataProcessing of validationReasoning NMS processingdrawbox for visualization of reasoning results For the calculation of map ...

1 data organization (1) coco128 dataset Here we use the coco dataset, but the coco dataset is too large. Here we use the coco128 dataset, because there are only 128 pictures, and the speed of copying and decompression can be fast. The download link of the coco128 dataset is: https://github.com/ultralytics/yolov5/releases/download/v1.0/coc ...