1、TensorRT的需要的文件
需要的基本文件(不是必须的)
1>网络结构文件(deploy.prototxt)
2>训练的权重模型(net.caffemodel)
TensorRT 2.0 EA版中的sampleMNISTAPI和TensorRT 1.0中的sampleMNISTGIE 几乎没有变化,就是不使用caffemodel 文件构建network 的例子。
2、TensorRT支持的层
Convolution: 2D
Activation: ReLU, tanh and sigmoid
Pooling: max and average
ElementWise: sum, product or max of two tensors
LRN: cross-channel only
Fully-connected: with or without bias
SoftMax: cross-channel only
Deconvolution
对于TensorRT 不支持的层,可以先将支持的层跑完,然后将输出作为caffe的输入,用caffe再跑,V1不支持TensorRT 和caffe同时工作,V2支持。(例子NVIDIA正在做,后期可能会上传github)
3、TensorRT 处理流程
基本处理过程:1>caffe model 转化 gie的model,或者从磁盘或者网络加载gie可用的model;2>运行GIE引擎(数据提前copy到GPU中);3>提取结果
https://developer.nvidia.com/tensorrt
https://docs.nvidia.com/deeplearning/sdk/tensorrt-developer-guide/index.html
https://docs.nvidia.com/deeplearning/sdk/index.html
https://docs.nvidia.com/deeplearning/sdk/tensorrt-developer-guide/index.html#samples
https://www.cnblogs.com/bonelee/p/8311445.html
https://blog.csdn.net/xh_hit/article/details/82917948
https://blog.csdn.net/xh_hit/article/details/82918162
https://github.com/haoxurt/tensorRT_save_serialization