EasyLocateTrain

Support
Required licenses	EasyLocate
Recommended images	n/a
Location	Deep Learning Inspection/EasyLocateTrain

Purpose

This sample program demonstrates how to:

• Train an EasyLocate tool in console mode.

Code highlights

By default, the sample program trains a tool for the dataset ElectronicComponentsBag in the Deep Learning Additional Resources.

Note

In the code, you need to replace DEEP_LEARNING_ADDITIONAL_RESOURCES by the path of the Deep Learning Additional Resources.

1. Select the first available GPU for the training.

   // initialize a segmenter that will be trained
   ELocator newLocator;
   
   // make sure that we have a GPU to use (note: GPU's can only be detected in x64 configuration)
   std::vector<EDeepLearningDevice> devices = newClassifier.GetAvailableDevices();
   
   bool foundGPU = false;
   for (size_t i = 0; i < devices.size(); i++)
   {
     if (devices[i].GetDeviceType() == EDeepLearningDeviceType_GPU)
     {
       newClassifier.SetActiveDevice(devices[i]);
       foundGPU = true;
       break;
     }
   }
   

2. Load a dataset from a project.

   const EClassificationDataset* dataset;
   EDeepLearningProject project;
   project.Load(PROJECT_PATH);
   dataset = &project.GetDataset();
   

3. Alternatively, load a dataset directly from a .edldataset file.

   const EClassificationDataset* dataset;
   EClassificationDataset datasetHolder;
   datasetHolder.Load(DATASET_PATH);
   dataset = &datasetHolder;
   

4. Split the dataset in a training part and a validation part.

Note

Be careful that the method for splitting a dataset is different for each type of tool.

```c++
// Put 80% of images into the training dataset and 20% in the validation dataset
EClassificationDataset trainingDataset;
EClassificationDataset validationDataset;
dataset->SplitDatasetForLocator(trainingDataset, validationDataset, 0.8f);
```

5. Configure and train the EasyLocate tool and wait for the training to end.

   newLocator.SetHeight(550);
   newLocator.SetWidth(345);
   newLocator.SetCapacity(ELocatorCapacity_Small);
   newLocator.SetChannels(1);
   newLocator.Train(trainingDataset, validationDataset, 1);
   newLocator.WaitForTrainingCompletion();
   

6. Check the validation F-Score of the tool we just trained.

   float newLocatorFScore = newLocator.GetValidationMetrics(newLocator.GetBestIteration()).GetBestWeightedFScore();
   std::cout << "New locator F-Score:" << newLocatorFScore << std::endl;