move backward debugging functions .h -> .cu

This commit is contained in:
Hieu Hoang 2016-09-21 11:03:35 +01:00
parent 828a0db8bc
commit 0fe591f64b
4 changed files with 115 additions and 96 deletions

View File

@ -3,12 +3,13 @@ include_directories(.)
cuda_add_library(marian_lib
cnpy/cnpy.cpp
dropout.cu
exception.cpp
expression_graph.cu
expression_operators.cu
node.cu
tensor.cu
tensor_operators.cu
expression_operators.cu
dropout.cu
vocab.cpp
)

102
src/node.cu Normal file
View File

@ -0,0 +1,102 @@
#include "node.h"
namespace marian {
// for backward_numeric
void Node::calc_numeric_grad(
Float delta,
Tensor input,
Tensor grad,
const std::vector<float> &prevCalcGrad
)
{
using namespace std;
size_t inputSize = GetTotalSize(input.shape());
size_t gradSize = GetTotalSize(grad.shape());
size_t adjSize = GetTotalSize(adj_.shape());
cerr << "sizes: "
<< Debug(input.shape())<< "=" << inputSize << " "
<< Debug(grad.shape()) << "=" << gradSize << " "
<< Debug(adj_.shape()) << "=" << adjSize
<< endl;
std::vector<float> diffGrad(gradSize);
thrust::copy(grad.begin(), grad.end(), diffGrad.begin());
cerr << "diffGrad=" << grad.Debug() << endl;
//output("diffGrad", diffGrad);
// reset grad
thrust::copy(prevCalcGrad.begin(), prevCalcGrad.end(), grad.begin());
//cerr << "reset a_->grad()=" << a_->grad().Debug() << endl;
// START CALC of numerical gradient
// new values
input.incr(delta);
forward();
//cerr << "input=" << input.Debug() << endl;
//cerr << "val_=" << val_.Debug() << endl;
std::vector<float> newVal(inputSize);
thrust::copy(val_.begin(), val_.end(), newVal.begin());
//output("newVal", newVal);
// old values
input.incr(-delta);
forward();
//cerr << "input=" << input.Debug() << endl;
//cerr << "val_=" << val_.Debug() << endl;
std::vector<float> origVal(inputSize);
thrust::copy(val_.begin(), val_.end(), origVal.begin());
//output("origVal", origVal);
// calc gradient
//cerr << "adj_=" << adj_.Debug() << endl;
std::vector<float> adjVec(adjSize);
thrust::copy(adj_.begin(), adj_.end(), adjVec.begin());
std::vector<float> numericalGrad(gradSize);
for (size_t i = 0; i < numericalGrad.size(); ++i) {
numericalGrad[i] = prevCalcGrad[i] + (adjVec[i] * (newVal[i] - origVal[i]) / delta);
}
// set grad results
thrust::copy(numericalGrad.begin(), numericalGrad.end(), grad.begin());
cerr << "numericalGrad=" << grad.Debug() << endl;
//output("numericalGrad", numericalGrad);
// print out diff between diffGrad and numericalGrad
std::vector<float> origGrad(gradSize);
std::vector<float> diff(gradSize);
thrust::copy(grad.begin(), grad.end(), origGrad.begin());
for (size_t i = 0; i < diff.size(); ++i) {
diff[i] = (diffGrad[i] - numericalGrad[i]) ;
}
output("diff", diff);
}
std::vector<float> Node::StoreTensorInVec(Tensor tensor)
{
size_t totSize = GetTotalSize(tensor.shape());
std::vector<float> vec(totSize);
thrust::copy(tensor.begin(), tensor.end(), vec.begin());
return vec;
}
void Node::output(const std::string &title, const std::vector<float> &vec)
{
std::cerr << title << "(" << vec.size() << "): ";
for (size_t i = 0; i < vec.size(); ++i) {
std::cerr << vec[i] << " ";
}
std::cerr << std::endl;
}
}

View File

@ -112,100 +112,15 @@ class Node : public Chainable<Tensor>,
Tensor val_;
Tensor adj_;
// for backward_numeric
void calc_numeric_grad(
Float delta,
Tensor input,
Tensor grad,
const std::vector<float> &prevCalcGrad
)
{
using namespace std;
void output(const std::string &title, const std::vector<float> &vec);
std::vector<float> StoreTensorInVec(Tensor tensor);
void calc_numeric_grad(
Float delta,
Tensor input,
Tensor grad,
const std::vector<float> &prevCalcGrad
);
size_t inputSize = GetTotalSize(input.shape());
size_t gradSize = GetTotalSize(grad.shape());
size_t adjSize = GetTotalSize(adj_.shape());
cerr << "sizes: "
<< Debug(input.shape())<< "=" << inputSize << " "
<< Debug(grad.shape()) << "=" << gradSize << " "
<< Debug(adj_.shape()) << "=" << adjSize
<< endl;
std::vector<float> diffGrad(gradSize);
thrust::copy(grad.begin(), grad.end(), diffGrad.begin());
cerr << "diffGrad=" << grad.Debug() << endl;
//output("diffGrad", diffGrad);
// reset grad
thrust::copy(prevCalcGrad.begin(), prevCalcGrad.end(), grad.begin());
//cerr << "reset a_->grad()=" << a_->grad().Debug() << endl;
// START CALC of numerical gradient
// new values
input.incr(delta);
forward();
//cerr << "input=" << input.Debug() << endl;
//cerr << "val_=" << val_.Debug() << endl;
std::vector<float> newVal(inputSize);
thrust::copy(val_.begin(), val_.end(), newVal.begin());
//output("newVal", newVal);
// old values
input.incr(-delta);
forward();
//cerr << "input=" << input.Debug() << endl;
//cerr << "val_=" << val_.Debug() << endl;
std::vector<float> origVal(inputSize);
thrust::copy(val_.begin(), val_.end(), origVal.begin());
//output("origVal", origVal);
// calc gradient
//cerr << "adj_=" << adj_.Debug() << endl;
std::vector<float> adjVec(adjSize);
thrust::copy(adj_.begin(), adj_.end(), adjVec.begin());
std::vector<float> numericalGrad(gradSize);
for (size_t i = 0; i < numericalGrad.size(); ++i) {
numericalGrad[i] = prevCalcGrad[i] + (adjVec[i] * (newVal[i] - origVal[i]) / delta);
}
// set grad results
thrust::copy(numericalGrad.begin(), numericalGrad.end(), grad.begin());
cerr << "numericalGrad=" << grad.Debug() << endl;
//output("numericalGrad", numericalGrad);
// print out diff between diffGrad and numericalGrad
std::vector<float> origGrad(gradSize);
std::vector<float> diff(gradSize);
thrust::copy(grad.begin(), grad.end(), origGrad.begin());
for (size_t i = 0; i < diff.size(); ++i) {
diff[i] = (diffGrad[i] - numericalGrad[i]) ;
}
output("diff", diff);
}
std::vector<float> StoreTensorInVec(Tensor tensor)
{
size_t totSize = GetTotalSize(tensor.shape());
std::vector<float> vec(totSize);
thrust::copy(tensor.begin(), tensor.end(), vec.begin());
return vec;
}
void output(const std::string &title, const std::vector<float> &vec)
{
std::cerr << title << "(" << vec.size() << "): ";
for (size_t i = 0; i < vec.size(); ++i) {
std::cerr << vec[i] << " ";
}
std::cerr << std::endl;
}
};

View File

@ -59,6 +59,7 @@ int main(int argc, char** argv)
std::cout << g.graphviz() << std::endl;
/*
std::cerr << "inTensor=" << inTensor.Debug() << std::endl;
Tensor outTensor = outExpr.val();
@ -66,6 +67,6 @@ int main(int argc, char** argv)
Tensor outGrad = outExpr.grad();
std::cerr << "outGrad=" << outGrad.Debug() << std::endl;
*/
}