Mathis/CUDANet
1
0
Fork 0
mirror of https://github.com/lordmathis/CUDANet.git synced 2025-11-05 17:34:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Implement Inception block A

Browse Source
This commit is contained in:
Matúš Námešný
2024-05-26 14:29:03 +02:00
parent cba177e417
commit 1e3b2f7d7b
1 changed files with 82 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

115
examples/inception_v3/inception_v3.cpp
View File

@@ -14,23 +14,23 @@ int main(int argc, const char *const argv[]) {
class BasicConv2d : public CUDANet::Module {
public:
BasicConv2d(
const int inputSize,
const dim2d inputSize,
const int inputChannels,
const int outputChannels,
const int kernelSize,
const int stride,
const int padding,
const dim2d kernelSize,
const dim2d stride,
const dim2d padding,
const std::string &prefix
) {
// Create the convolution layer
CUDANet::Layers::Conv2d *conv = new CUDANet::Layers::Conv2d(
conv = new CUDANet::Layers::Conv2d(
inputSize, inputChannels, kernelSize, stride, outputChannels,
padding, CUDANet::Layers::ActivationType::NONE
);
int batchNormSize = conv->getOutputSize();
dim2d batchNormSize = conv->getOutputDims();
CUDANet::Layers::BatchNorm2d *batchNorm = new CUDANet::Layers::BatchNorm2d(
batchNorm = new CUDANet::Layers::BatchNorm2d(
batchNormSize, outputChannels, 1e-3f,
CUDANet::Layers::ActivationType::RELU
);
@@ -39,20 +39,20 @@ class BasicConv2d : public CUDANet::Module {
addLayer(prefix + ".bn", batchNorm);
}
float* forward(const float* d_input) {
for (auto& layer : layers) {
d_input = layer.second->forward(d_input);
}
return d_input;
float *forward(const float *d_input) {
float *d_output = conv->forward(d_input);
return batchNorm->forward(d_output);
}
private:
CUDANet::Layers::Conv2d *conv;
CUDANet::Layers::BatchNorm2d *batchNorm;
};
class InceptionA : public CUDANet::Module {
public:
InceptionA(
const int inputSize,
const dim2d inputSize,
const int inputChannels,
const int poolFeatures,
const std::string &prefix
@@ -60,42 +60,47 @@ class InceptionA : public CUDANet::Module {
: inputSize(inputSize),
inputChannels(inputChannels),
poolFeatures(poolFeatures) {
// Branch 1x1
CUDANet::Module *branch1x1 = new BasicConv2d(
inputSize, inputChannels, 64, 1, 1, 0, prefix + ".branch1x1"
branch1x1 = new BasicConv2d(
inputSize, inputChannels, 64, {1, 1}, {1, 1}, {0, 0}, prefix + ".branch1x1"
);
addLayer("", branch1x1);
// Branch 5x5
CUDANet::Module *branch5x5_1 = new BasicConv2d(
inputSize, inputChannels, 48, 1, 1, 0, prefix + ".branch5x5_1"
branch5x5_1 = new BasicConv2d(
inputSize, inputChannels, 48, {1, 1}, {1, 1}, {0, 0}, prefix + ".branch5x5_1"
);
addLayer("", branch5x5_1);
CUDANet::Module *branch5x5_2 = new BasicConv2d(
inputSize, 48, 64, 5, 1, 2, prefix + ".branch5x5_2"
branch5x5_2 = new BasicConv2d(
inputSize, 48, 64, {5, 5}, {1, 1}, {2, 2}, prefix + ".branch5x5_2"
);
addLayer("", branch5x5_2);
// Branch 3x3
CUDANet::Module *branch3x3_1 = new BasicConv2d(
inputSize, inputChannels, 64, 1, 1, 0, prefix + ".branch3x3_1"
branch3x3_1 = new BasicConv2d(
inputSize, inputChannels, 64, {1, 1}, {1, 1}, {0, 0}, prefix + ".branch3x3_1"
);
addLayer("", branch3x3_1);
CUDANet::Module *branch3x3_2 = new BasicConv2d(
inputSize, 64, 96, 3, 1, 1, prefix + ".branch3x3_2"
branch3x3_2 = new BasicConv2d(
inputSize, 64, 96, {3, 3}, {1, 1}, {1, 1}, prefix + ".branch3x3_2"
);
addLayer("", branch3x3_2);
CUDANet::Module *branch3x3_3 = new BasicConv2d(
inputSize, 96, 96, 3, 1, 1, prefix + ".branch3x3_3"
branch3x3_3 = new BasicConv2d(
inputSize, 96, 96, {3, 3}, {1, 1}, {1, 1}, prefix + ".branch3x3_3"
);
addLayer("", branch3x3_3);
// Branch Pool
CUDANet::Module *branchPool = new BasicConv2d(
inputSize, inputChannels, poolFeatures, 1, 1, 0, prefix + ".branchPool"
branchPool_1 = new CUDANet::Layers::AvgPooling2d(
inputSize, inputChannels, {3, 3}, {1, 1},
CUDANet::Layers::ActivationType::NONE
);
addLayer("", branchPool);
addLayer("", branchPool_1);
branchPool_2 = new BasicConv2d(
branchPool_1->getOutputDims(), inputChannels, poolFeatures, {1, 1}, {1, 1}, {0, 0},
prefix + ".branchPool"
);
addLayer("", branchPool_2);
// Concat
concat_1 = new CUDANet::Layers::Concat(
@@ -104,14 +109,58 @@ class InceptionA : public CUDANet::Module {
concat_2 = new CUDANet::Layers::Concat(
concat_1->getOutputSize(), branch3x3_3->getOutputSize()
);
concat_3 = new CUDANet::Layers::Concat(
concat_2->getOutputSize(), branchPool_2->getOutputSize()
);
}
float *forward(const float *d_input) {
float *d_branch1x1_out = branch1x1->forward(d_input);
float *d_branch5x5_out = branch5x5_1->forward(d_input);
d_branch5x5_out = branch5x5_2->forward(d_branch5x5_out);
float *d_branch3x3_out = branch3x3_1->forward(d_input);
d_branch3x3_out = branch3x3_2->forward(d_branch3x3_out);
d_branch3x3_out = branch3x3_3->forward(d_branch3x3_out);
float *d_branchPool_out = branchPool_1->forward(d_input);
d_branchPool_out = branchPool_2->forward(d_branchPool_out);
float *d_output = concat_1->forward(d_branch1x1_out, d_branch5x5_out);
d_output = concat_2->forward(d_output, d_branch3x3_out);
d_output = concat_3->forward(d_output, d_branchPool_out);
return d_output;
}
private:
int inputSize;
dim2d inputSize;
int inputChannels;
int poolFeatures;
CUDANet::Module *branch1x1;
CUDANet::Module *branch5x5_1;
CUDANet::Module *branch5x5_2;
CUDANet::Module *branch3x3_1;
CUDANet::Module *branch3x3_2;
CUDANet::Module *branch3x3_3;
CUDANet::Layers::AvgPooling2d *branchPool_1;
CUDANet::Module *branchPool_2;
CUDANet::Layers::Concat *concat_1;
CUDANet::Layers::Concat *concat_2;
};
CUDANet::Layers::Concat *concat_3;
};
class InceptionB : public CUDANet::Module {
public:
InceptionB(const int inputSize, const int inputChannels)
: inputSize(inputSize), inputChannels(inputChannels) {
// Branch 1x1
branch1x1 = new BasicConv2d(
inputSize, inputChannels, 128, 1, 1, 0, "branch1x1"
);
addLayer("", branch1x1);
}
}