解决了问题才来记录一下的，现在的心情是好点儿的，但是之前，昨天，真是无厘头，被折腾的一点脾气都没有。
本来就是一个softmax嘛，很简单的嘛，就是按照公式e的指数再相加求和，官方有api实现，比如我找的例子，
// Add activation layer using the ReLU algorithm.
IActivationLayer* relu1 = network->addActivation(*ip1->getOutput(0), ActivationType::kRELU);
assert(relu1);
// Add second fully connected layer with 20 outputs.
IFullyConnectedLayer* ip2 = network->addFullyConnected(
*relu1->getOutput(0), mParams.outputSize, mWeightMap["ip2filter"], mWeightMap["ip2bias"]);
assert(ip2);
// Add softmax layer to determine the probability.
ISoftMaxLayer* prob = network->addSoftMax(*ip2->getOutput(0));
assert(prob);
prob->getOutput(0)->setName(mParams.outputTensorNames[0].c_str());
network->markOutput(*prob->getOutput(0));
这是官方例子简单分类的例子，
https://github.com/NVIDIA/TensorRT/blob/master/samples/opensource/sampleMNISTAPI/sampleMNISTAPI.cpp
里面就有ISoftMaxLayer* prob = network->addSoftMax(*ip2->getOutput(0));
我这里要实现的稍微有点儿不同，在于我需要按照某个维度来softmax，上面这个是全局的。我需要实现的对应pytorch代码如下：
#[1,6375,2] # arm_conf[1,12750]
arm_conf_view = arm_conf.view(arm_conf.size(0), -1,2)
softmax_1 = nn.Softmax(dim=-1)
m2 = softmax_1(arm_conf_view) # #[1,6375,2]
对应的m2部分数值如下：
tensor([[[0.9575, 0.0425],
[0.9326, 0.0674],
[0.9131, 0.0869],
...,
[0.8707, 0.1293],
[0.8746, 0.1254],
[0.8783, 0.1217]]], grad_fn=<SoftmaxBackward>)
可以看到，是按照第二维进行softmax的，每行加起来之和为1.
在tensorrt端，我现在也得到了一个一维数组arm_conf[12750]，再把这个数组也一样softmax就可以了。但是我这个不是全局的softmax。然后我就去看
https://github.com/wang-xinyu/tensorrtx这个仓库有没有人用，一搜索有个人用。
https://github.com/wang-xinyu/tensorrtx/blob/18fa419ae35bfcbd27248b3eb9329f415f604366/retinafaceAntiCov/retinafaceAntiCov.cpp
ILayer* reshapeSoftmax(INetworkDefinition *network, ITensor& input, int c) {
auto re1 = network->addShuffle(input);
assert(re1);
re1->setReshapeDimensions(Dims3(c / 2, -1, 0));
auto sm = network->addSoftMax(*re1->getOutput(0));
assert(sm);
auto re2 = network->addShuffle(*sm->getOutput(0));
assert(re2);
re2->setReshapeDimensions(Dims3(c, -1, 0));
return re2;
}
好像和我需要的功能一样，因为我是12750数据，也需要先变成[6375,2]形状的数据，最后传出去的时候再变成[12750]. 很快，我也写类似代码：
ILayer* reshapeSoftmax_ok(INetworkDefinition *network, ITensor& input, int ch) {
//输入进来是一维的[12750]
//先变成[XX,ch]
auto re1 = network->addShuffle(input);
assert(re1);
re1->setReshapeDimensions(Dims3(1, -1, ch)); //[1,6375,2];
// re1->setReshapeDimensions(Dims2(-1, ch)); //[6375,2];
Dims dim0 = re1->getOutput(0)->getDimensions();
std::cout <<"debug re1 dim==" << dim0.d[0] << " " << dim0.d[1] << " " << dim0.d[2] << " " << dim0.d[3] << std::endl;
auto sm = network->addSoftMax(*re1->getOutput(0));
sm->setAxes(2);
assert(sm);
//再变成一维的,保持和传进来的形状一样
Dims dim_;
dim_.nbDims=1;
dim_.d[0]=-1;
auto re2 = network->addShuffle(*sm->getOutput(0));
assert(re2);
re2->setReshapeDimensions(dim_);
return re2;
}
我这里多了sm->setAxes(2);因为我需要在第二维度softmax，但是！结果不对啊。怎么整结果都不对，三维，二维都不对！
sm->setAxes(0);,sm->setAxes(1);sm->setAxes(2);都试了都不对，有的出来的结果好像是全局softmax的结果，有的都是1.
期间，也是查看了文档帮助，
//!
//! rief Add a SoftMax layer to the network.
//!
//! see ISoftMaxLayer
//! warning Int32 tensors are not valid input tensors.
//!
//!
eturn The new SoftMax layer, or nullptr if it could not be created.
//!
virtual ISoftMaxLayer* addSoftMax(ITensor& input) TRTNOEXCEPT = 0;
//!
//! class ISoftMaxLayer
//!
//! rief A Softmax layer in a network definition.
//!
//! This layer applies a per-channel softmax to its input.
//!
//! The output size is the same as the input size.
//!
//! warning Do not inherit from this class, as doing so will break forward-compatibility of the API and ABI.
//!
class ISoftMaxLayer : public ILayer
{
protected:
virtual ~ISoftMaxLayer() {}
public:
//!
//! rief Set the axis along which softmax is computed. Currently, only one axis can be set.
//!
//! The axis is specified by setting the bit corresponding to the axis to 1.
//! Let's say we have an NCHW tensor as input (three non-batch dimensions).
//!
//! In implicit mode :
//! Bit 0 corresponds to the C dimension boolean.
//! Bit 1 corresponds to the H dimension boolean.
//! Bit 2 corresponds to the W dimension boolean.
//! By default, softmax is performed on the axis which is the number of axes minus three. It is 0 if
//! there are fewer than 3 non-batch axes. For example, if the input is NCHW, the default axis is C. If the input
//! is NHW, then the default axis is H.
//!
//! In explicit mode :
//! Bit 0 corresponds to the N dimension boolean.
//! Bit 1 corresponds to the C dimension boolean.
//! Bit 2 corresponds to the H dimension boolean.
//! Bit 3 corresponds to the W dimension boolean.
//! By default, softmax is performed on the axis which is the number of axes minus three. It is 0 if
//! there are fewer than 3 axes. For example, if the input is NCHW, the default axis is C. If the input
//! is NHW, then the default axis is N.
//!
//! For example, to perform softmax on axis R of a NPQRCHW input, set bit 2 with implicit batch mode,
//! set bit 3 with explicit batch mode.
//!
//! param axes The axis along which softmax is computed.
//! Here axes is a bitmap. For example, when doing softmax along axis 0, bit 0 is set to 1, axes = 1 << axis = 1.
//!
virtual void setAxes(uint32_t axes) TRTNOEXCEPT = 0;
//!
//! rief Get the axis along which softmax occurs.
//!
//! see setAxes()
//!
virtual uint32_t getAxes() const TRTNOEXCEPT = 0;
};
就两个函数可以调用设置维度啊，就是不对，晚上加班各种实验，都是和之前一样的结果，然后又怀疑输入数据不一样，或者tensorrt reshape之后数据不一样，
//输入进来是一维的[12750]
//先变成[XX,ch]
auto re1 = network->addShuffle(input);
assert(re1);
re1->setReshapeDimensions(Dims3(1, -1, ch)); //[1,6375,2];
// re1->setReshapeDimensions(Dims2(-1, ch)); //[6375,2];
return re1;/////////////////////////////////////////
reshape之后提前return，然后看是否和pytorch一致，发现是一致的！
这里实验一次特别麻烦，要先序列化生成engine，然后再反序列化推理出结果才能完成一次实验。
晚上加班到10点多，没有解决，解决不了，郁郁闷闷的回家了。
期间还有个思路就是用cuda编程完成softmax操作，但是感觉太麻烦了。
群里问了大佬，没人回答，然后又私聊问了群里的球哥，他也表示代码是对的，说明天给我实验。
最后我总结了一下就是
“你明天做实验的时候就是比如输入是一个形状是[8]的tensor，然后进函数里面4️ 2，softmax，应该也得到4乘2的tensor，4行2列，每行加起来和为1，然后再变为形状为[8]的tensor输出来。”
然后第二天早上又来搞，检查了一下其他环节有没有问题，没有问题的。又实验了几次，现象还是和之前一样。无解。
中午球哥联系我，让我确认了输入和pytorch代码数据一致。我太困了，拿着手机趴会儿因为一有消息有就能醒，果真到13点的时候，
可以！！！
我意识到这是bit位操作的一些写法。球哥说是的。
给我发了一个图片：
注意图片上面的这段英文：
reduceAxes: the least significant bit corresponds to the first explicit dimension
reduceAxes：最低有效位对应于第一个显式维度
mask是0010，bitmax就是0100
比如我是[6375,2]，想在第1维度softmax，mask就是[0,1]对应的bitmap就是[1,0]，所以需要这么写1<<1
比如我是[1,6375,2]，想在第2维度softmax，mask就是[0,0,1]对应的bitmap就是[1,0,0]，所以需要这么写1<<2
bitmap和mask是倒着的啊！
现在反过来再去看官方文档，
//! In explicit mode :
//! Bit 0 corresponds to the N dimension boolean.
//! Bit 1 corresponds to the C dimension boolean.
//! Bit 2 corresponds to the H dimension boolean.
//! Bit 3 corresponds to the W dimension boolean.
//! By default, softmax is performed on the axis which is the number of axes minus three. It is 0 if
//! there are fewer than 3 axes. For example, if the input is NCHW, the default axis is C. If the input
//! is NHW, then the default axis is N.
//!
//! For example, to perform softmax on axis R of a NPQRCHW input, set bit 2 with implicit batch mode,
//! set bit 3 with explicit batch mode.
//!
//! param axes The axis along which softmax is computed.
//! Here axes is a bitmap. For example, when doing softmax along axis 0, bit 0 is set to 1, axes = 1 << axis = 1.
还是看不懂，axes = 1 << axis = 1.这什么鬼，就不能好好写个例子的吗？
感谢球哥！！
正确的代码如下
ILayer* reshapeSoftmax(INetworkDefinition *network, ITensor& input, int ch) {
//输入进来是一维的[12750]
//先变成[XX,ch]
auto re1 = network->addShuffle(input);
assert(re1);
re1->setReshapeDimensions(Dims3(1, -1, ch)); //[1,6375,2];
// re1->setReshapeDimensions(Dims2(-1, ch)); //[6375,2];
Dims dim0 = re1->getOutput(0)->getDimensions();
std::cout <<"debug re1 dim==" << dim0.d[0] << " " << dim0.d[1] << " " << dim0.d[2] << " " << dim0.d[3] << std::endl;
// return re1;/////////////////////////////////////////
auto sm = network->addSoftMax(*re1->getOutput(0));
sm->setAxes(1<<2);
assert(sm);
//再变成一维的,保持和传进来的形状一样
Dims dim_;
dim_.nbDims=1;
dim_.d[0]=-1;
auto re2 = network->addShuffle(*sm->getOutput(0));
assert(re2);
re2->setReshapeDimensions(dim_);
return re2;
}
和球哥讨论，不写setAxes，应该是默认第二个。nchw是c chw是h
然后回顾一下仓库里面的这段代码，人家之所以reshape，可能就是这个大佬发现了softmax操作只能按照某个维度来，需要按照自己指定的维度需要reshape一下才能得到正确结果。
可能他没有知道需要这么写->setAxes(1<<2);指定维度。
感觉球哥研究tensorrt比较深入，还用到了其他的高级api，
