关于yolov8的output0

最新推荐文章于 2024-07-05 00:13:54 发布

lindsayshuo

最新推荐文章于 2024-07-05 00:13:54 发布

阅读量351

点赞数 1

文章标签： YOLO

本文链接：https://blog.csdn.net/weixin_43269994/article/details/136485019

版权

本文深入剖析了Yolov8中的output0生成过程，涉及nvinfer1库中的convBnSiLU、卷积和元素级运算层，以及如何通过ONNX图连接这些层以进行特征处理。

摘要由CSDN通过智能技术生成

关于yolov8的output0

// output0
    nvinfer1::IElementWiseLayer* conv22_cv2_0_0 = convBnSiLU(network, weightMap, *conv15->getOutput(0), base_in_channel, 3, 1, 1, "model.22.cv2.0.0");
    nvinfer1::IElementWiseLayer* conv22_cv2_0_1 = convBnSiLU(network, weightMap, *conv22_cv2_0_0->getOutput(0), base_in_channel, 3, 1, 1, "model.22.cv2.0.1");
    nvinfer1::IConvolutionLayer* conv22_cv2_0_2 = network->addConvolutionNd(*conv22_cv2_0_1->getOutput(0), 64, nvinfer1::DimsHW{1, 1}, weightMap["model.22.cv2.0.2.weight"], weightMap["model.22.cv2.0.2.bias"]);
    conv22_cv2_0_2->setStrideNd(nvinfer1::DimsHW{1, 1});
    conv22_cv2_0_2->setPaddingNd(nvinfer1::DimsHW{0, 0});
    nvinfer1::IElementWiseLayer* conv22_cv3_0_0 = convBnSiLU(network, weightMap, *conv15->getOutput(0),base_out_channel, 3, 1, 1, "model.22.cv3.0.0");
    nvinfer1::IElementWiseLayer* conv22_cv3_0_1 = convBnSiLU(network, weightMap, *conv22_cv3_0_0->getOutput(0), base_out_channel, 3, 1, 1, "model.22.cv3.0.1");
    nvinfer1::IConvolutionLayer* conv22_cv3_0_2 = network->addConvolutionNd(*conv22_cv3_0_1->getOutput(0), kNumClass, nvinfer1::DimsHW{1, 1}, weightMap["model.22.cv3.0.2.weight"], weightMap["model.22.cv3.0.2.bias"]);
    conv22_cv3_0_2->setStride(nvinfer1::DimsHW{1, 1});
    conv22_cv3_0_2->setPadding(nvinfer1::DimsHW{0, 0});
    nvinfer1::ITensor* inputTensor22_0[] = {conv22_cv2_0_2->getOutput(0), conv22_cv3_0_2->getOutput(0)};
    nvinfer1::IConcatenationLayer* cat22_0 = network->addConcatenation(inputTensor22_0, 2);

onnx图为

请添加图片描述