Debugging-setup with a single training sample to check the lack of learning progress
This commit is contained in:
parent
1d343a079a
commit
44760d0402
1 changed files with 3 additions and 3 deletions
|
@ -26,10 +26,10 @@ void NetLearner::run()
|
|||
size_t numIterations = 100000;
|
||||
for (size_t iteration = 0; iteration < numIterations && cancel == false; ++iteration)
|
||||
{
|
||||
auto trainingSample = mnistLoader.getRandomSample();
|
||||
auto trainingSample = mnistLoader.getSample(0);
|
||||
|
||||
QImage trainingImage(trainingSample.data, 28, 28, QImage::Format_Grayscale8);
|
||||
emit sampleImageLoaded(trainingImage);
|
||||
emit logMessage(QString("training sample ") + QString::number(trainingSample.label));
|
||||
emit sampleImageLoaded(trainingSample.toQImage());
|
||||
|
||||
std::vector<double> targetValues =
|
||||
{
|
||||
|
|
Loading…
Reference in a new issue