private async Task LoadModelAsync()
{
Debug.Write("LoadModelBegin | ");
Debug.Write("LoadModel Lock | ");
_binding?.Clear();
_session?.Dispose();
StorageFile modelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///Assets/{_appModel.ModelSource}.onnx"));
_learningModel = await LearningModel.LoadFromStorageFileAsync(modelFile);
_inferenceDeviceSelected = UseGpu ? LearningModelDeviceKind.DirectX : LearningModelDeviceKind.Cpu;
// Lock so can't create a new session or binding while also being disposed
lock (_processLock)
{
_session = new LearningModelSession(_learningModel, new LearningModelDevice(_inferenceDeviceSelected));
_binding = new LearningModelBinding(_session);
}
debugModelIO();
_inputImageDescription = _learningModel.InputFeatures.ToList().First().Name;
_outputImageDescription = _learningModel.OutputFeatures.ToList().First().Name;
Debug.Write("LoadModel Unlock\n");
}
// Loads the ML model file and creates LearningModel
private async Task LoadModelAsync()
{
try
{
Debug.WriteLine($"*************** LoadModelAsync [START] ***************");
Debug.WriteLine($" - Locating ONYX file");
var modelFile = await StorageFile.GetFileFromApplicationUriAsync(ModelUri);
Debug.WriteLine($" - {modelFile.DisplayName} file located, attempting to create LearningModel");
Debug.WriteLine($" - Attempting to create LearningModel...");
model = await LearningModel.LoadFromStorageFileAsync(modelFile);
Debug.WriteLine($" - {model.Name} LearningModel successfully instantiated.");
modelCreated = true;
}
catch (Exception ex)
{
Debug.WriteLine($" - Error: Problem loading model file or creating LearningModel: {ex.Message}");
model = null;
modelCreated = false;
}
finally
{
Debug.WriteLine($"*************** LoadModelAsync [END] ***************");
}
}
// Create a model from an ONNX 1.2 file
public static async Task <CustomVisionModel> CreateONNXModel(StorageFile file)
{
LearningModel learningModel = null;
try
{
learningModel = await LearningModel.LoadFromStorageFileAsync(file);
}
catch (Exception ex)
{
throw ex;
}
var inputFeatures = learningModel.InputFeatures;
ImageFeatureDescriptor inputImageDescription = inputFeatures?.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Image) as ImageFeatureDescriptor;
uint inputImageWidth = 0, inputImageHeight = 0;
if (inputImageDescription != null)
{
inputImageHeight = inputImageDescription.Height;
inputImageWidth = inputImageDescription.Width;
}
return(new CustomVisionModel()
{
_learningModel = learningModel,
_session = new LearningModelSession(learningModel),
InputImageWidth = (int)inputImageWidth,
InputImageHeight = (int)inputImageHeight
});
}
public static async Task <OnnxModel> CreateFromStreamAsync(StorageFile modelFile)
{
OnnxModel learningModel = new OnnxModel();
learningModel.model = await LearningModel.LoadFromStorageFileAsync(modelFile);
learningModel.session = new LearningModelSession(learningModel.model);
learningModel.binding = new LearningModelBinding(learningModel.session);
var inputFeature = learningModel.model.InputFeatures[0];
learningModel.inName = inputFeature.Name;
var inputTensorFeature = inputFeature as TensorFeatureDescriptor;
learningModel.InWidth = (int)inputTensorFeature.Shape[3];
learningModel.InHeight = (int)inputTensorFeature.Shape[2];
var outputFeature = learningModel.model.OutputFeatures[0];
learningModel.outName = outputFeature.Name;
var outputTensorFeature = outputFeature as TensorFeatureDescriptor;
learningModel.OutWidth = (int)outputTensorFeature.Shape[3];
learningModel.OutHeight = (int)outputTensorFeature.Shape[2];
Debug.Log(string.Format("Input Feature Name: {0}", learningModel.inName));
Debug.Log(string.Format("Output Feature Name: {0}", learningModel.outName));
Debug.Log(string.Format("Input Width, Height: {0}, {1}", learningModel.InWidth, learningModel.InHeight));
Debug.Log(string.Format("Output Width, Height: {0}, {1}", learningModel.OutWidth, learningModel.OutHeight));
return(learningModel);
}
public static async Task <Model> CreateModelAsync(string filename, bool gpu)
{
Log.WriteLine("creating model");
var file = await AsyncHelper.AsAsync(StorageFile.GetFileFromPathAsync(filename));
Log.WriteLine("have file");
var learningModel = await AsyncHelper.AsAsync(LearningModel.LoadFromStorageFileAsync(file));
Log.WriteLine("loaded model");
Model model = new Model();
model._model = learningModel;
LearningModelDeviceKind kind = LearningModelDeviceKind.Cpu;
if (gpu)
{
Log.WriteLine("using GPU");
kind = LearningModelDeviceKind.DirectXHighPerformance;
}
else
{
Log.WriteLine("using CPU");
}
model._device = new LearningModelDevice(kind);
model._session = new LearningModelSession(model._model, model._device);
Log.WriteLine("returning model now");
return(model);
}
/// <summary>
/// Load the label and model files
/// </summary>
/// <returns></returns>
private async Task LoadModelAsync()
{
// just load the model one time.
if (_model != null)
{
return;
}
StatusBlock.Text = $"Loading {_kModelFileName} ... patience ";
try
{
// Parse labels from label json file. We know the file's
// entries are already sorted in order.
var fileString = File.ReadAllText($"Assets/{_kLabelsFileName}");
var fileDict = JsonConvert.DeserializeObject <Dictionary <string, string> >(fileString);
foreach (var kvp in fileDict)
{
_labels.Add(kvp.Value);
}
// Load and create the model
var modelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///Assets/{_kModelFileName}"));
_model = await LearningModel.LoadFromStorageFileAsync(modelFile);
// Create the evaluation session with the model and device
_session = new LearningModelSession(_model, new LearningModelDevice(GetDeviceKind()));
}
catch (Exception ex)
{
StatusBlock.Text = $"error: {ex.Message}";
_model = null;
}
}
private async Task LoadAndEvaluateModelAsync(VideoFrame _inputFrame, string _modelFileName)
{
LearningModelBinding _binding = null;
VideoFrame _outputFrame = null;
LearningModelSession _session;
try
{
//Load and create the model
if (_model == null)
{
var modelFile =
await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///{_modelFileName}"));
_model = await LearningModel.LoadFromStorageFileAsync(modelFile);
}
// Create the evaluation session with the model
_session = new LearningModelSession(_model);
// Get input and output features of the model
var inputFeatures = _model.InputFeatures.ToList();
var outputFeatures = _model.OutputFeatures.ToList();
// Create binding and then bind input/ output features
_binding = new LearningModelBinding(_session);
_inputImageDescriptor =
inputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor) as TensorFeatureDescriptor;
_outputTensorDescriptor =
outputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor) as TensorFeatureDescriptor;
TensorFloat outputTensor = TensorFloat.Create(_outputTensorDescriptor.Shape);
ImageFeatureValue imageTensor = ImageFeatureValue.CreateFromVideoFrame(_inputFrame);
// Bind inputs +outputs
_binding.Bind(_inputImageDescriptor.Name, imageTensor);
_binding.Bind(_outputTensorDescriptor.Name, outputTensor);
// Evaluate and get the results
var results = await _session.EvaluateAsync(_binding, "test");
Debug.WriteLine("ResultsEvaluated: " + results.ToString());
var outputTensorList = outputTensor.GetAsVectorView();
var resultsList = new List <float>(outputTensorList.Count);
for (int i = 0; i < outputTensorList.Count; i++)
{
resultsList.Add(outputTensorList[i]);
}
}
catch (Exception ex)
{
Debug.WriteLine($"error: {ex.Message}");
_model = null;
}
}
public async Task Initialize()
{
// Load and create the model and session
var modelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///Assets//Network.onnx"));
_learning_model = await LearningModel.LoadFromStorageFileAsync(modelFile);
_session = new LearningModelSession(_learning_model);
}
/// <summary>
/// Initialize
/// </summary>
/// <param name="file">The ONNX file</param>
public async Task Init(StorageFile file)
{
this.model = await LearningModel.LoadFromStorageFileAsync(file);
this.session = new LearningModelSession(this.model);
Debug.Assert(this.model.InputFeatures.Count == 1, "The number of input must be 1");
Debug.Assert(this.model.OutputFeatures.Count == 1, "The number of output must be 1");
}
public async Task <bool> IniciarModelo()
{
var modelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///Assets/{_kModelFileName}"));
_model = await LearningModel.LoadFromStorageFileAsync(modelFile);
_session = new LearningModelSession(_model, new LearningModelDevice(LearningModelDeviceKindSelected));
return(true);
}
public static async Task <MyCustomVisionModel> CreateFromStreamAsync(StorageFile stream)
{
MyCustomVisionModel learningModel = new MyCustomVisionModel();
learningModel.model = await LearningModel.LoadFromStorageFileAsync(stream);
learningModel.session = new LearningModelSession(learningModel.model);
learningModel.binding = new LearningModelBinding(learningModel.session);
return(learningModel);
}
public static async Task <CustomVisionModel> CreateOnnxModel(StorageFile file)
{
CustomVisionModel learningModel = new CustomVisionModel();
learningModel.model = await LearningModel.LoadFromStorageFileAsync(file);
learningModel.session = new LearningModelSession(learningModel.model);
learningModel.binding = new LearningModelBinding(learningModel.session);
return(learningModel);
}
internal async Task InitModelAsync()
{
var model_file = await StorageFile.GetFileFromApplicationUriAsync(new Uri("ms-appx:///Assets//Yolo.onnx"));
_model = await LearningModel.LoadFromStorageFileAsync(model_file);
var device = new LearningModelDevice(LearningModelDeviceKind.Cpu);
_session = new LearningModelSession(_model, device);
_binding = new LearningModelBinding(_session);
}
/// <summary>
/// Initialize
/// </summary>
/// <param name="file">The ONNX file</param>
///
public async Task Init()
{
StorageFile ModelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///Assets/" + "model.onnx"));
this.model = await LearningModel.LoadFromStorageFileAsync(ModelFile);
this.session = new LearningModelSession(this.model);
this.binding = new LearningModelBinding(this.session);
Debug.Assert(this.model.InputFeatures.Count == 1, "The number of input must be 1");
Debug.Assert(this.model.OutputFeatures.Count == 1, "The number of output must be 1");
}
public static async Task <LearningModelYolo> CreateFromStreamAsync()
{
var modelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri("ms-appx:///OnnxModel/yolov2.onnx"));
var learningModel = new LearningModelYolo
{
Model = await LearningModel.LoadFromStorageFileAsync(modelFile)
};
learningModel.Session = new LearningModelSession(learningModel.Model);
learningModel.Binding = new LearningModelBinding(learningModel.Session);
return(learningModel);
}
private async Task EvaluateImageAsync(string imagePath, string modelPath)
{
var selectedStorageFile = await StorageFile.GetFileFromPathAsync(imagePath);
SoftwareBitmap softwareBitmap;
using (var stream = await selectedStorageFile.OpenAsync(FileAccessMode.Read)) {
// Create the decoder from the stream
var decoder = await BitmapDecoder.CreateAsync(stream);
// Get the SoftwareBitmap representation of the file in BGRA8 format
softwareBitmap = await decoder.GetSoftwareBitmapAsync();
softwareBitmap = SoftwareBitmap.Convert(softwareBitmap, BitmapPixelFormat.Bgra8,
BitmapAlphaMode.Premultiplied);
}
// Encapsulate the image within a VideoFrame to be bound and evaluated
var inputImage = VideoFrame.CreateWithSoftwareBitmap(softwareBitmap);
if (_model == null)
{
var modelFile = await StorageFile.GetFileFromPathAsync(modelPath);
_model = new SqueezeNetModel {
LearningModel = await LearningModel.LoadFromStorageFileAsync(modelFile)
};
_model.Session = new LearningModelSession(_model.LearningModel,
new LearningModelDevice(LearningModelDeviceKind.Default));
_model.Binding = new LearningModelBinding(_model.Session);
}
if (_model == null)
{
return;
}
var input = new SqueezeNetInput {
Image = ImageFeatureValue.CreateFromVideoFrame(inputImage)
};
try {
var output = (SqueezeNetOutput)await _model.EvaluateAsync(input);
var(label, probability) = output.classLabelProbs.FirstOrDefault();
DispatchEvent(Result, probability + ", " + label);
}
catch (Exception ex) {
Trace(ex.Message, ex.StackTrace);
}
}
public static async Task <SmartInkModel> CreateModelAsync(IStorageFile file, IList <string> tags)
{
try
{
SmartInkModel learningModel = new SmartInkModel();
learningModel._model = await LearningModel.LoadFromStorageFileAsync(file);
learningModel._session = new LearningModelSession(learningModel._model);
learningModel._binding = new LearningModelBinding(learningModel._session);
return(learningModel);
}
catch (Exception ex)
{
throw;
}
}
private async Task <bool> LoadModelAsync()
{
var modelStorageFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri(Constants.MODEL_PATH));
try
{
_model = await LearningModel.LoadFromStorageFileAsync(modelStorageFile);
}
catch (Exception ex)
{
Debug.WriteLine(ex.Message);
}
// since we do not specify the device, we are using the default CPU option
_session = new LearningModelSession(_model);
List <ILearningModelFeatureDescriptor> inputFeatures;
List <ILearningModelFeatureDescriptor> outputFeatures;
if (_model.InputFeatures == null)
{
return(false);
}
else
{
inputFeatures = _model.InputFeatures.ToList();
}
if (_model.OutputFeatures == null)
{
return(false);
}
else
{
outputFeatures = _model.OutputFeatures.ToList();
}
_inputImageDescriptor =
inputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor) as TensorFeatureDescriptor;
_outputTensorDescriptor =
outputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor) as TensorFeatureDescriptor;
return(true);
}
// Create a model from an ONNX 1.2 file
public static async Task <CustomVisionModel> CreateONNXModel(StorageFile file)
{
LearningModel learningModel = null;
try
{
learningModel = await LearningModel.LoadFromStorageFileAsync(file);
}
catch (Exception ex)
{
throw ex;
}
return(new CustomVisionModel()
{
_learningModel = learningModel,
_session = new LearningModelSession(learningModel)
});
}
// Create a model from an ONNX 1.2 file
public static async Task <ONNXModel> CreateONNXModel(StorageFile file)
{
LearningModel learningModel = null;
try
{
learningModel = await LearningModel.LoadFromStorageFileAsync(file);
}
catch (Exception e)
{
System.Console.WriteLine(e.Message);
throw e;
}
return(new ONNXModel()
{
_learningModel = learningModel,
_session = new LearningModelSession(learningModel)
});
}
public static async Task <Model> CreateModelAsync(string filename)
{
Log.WriteLine("creating model");
var file = await AsyncHelper.SyncFromAsync(StorageFile.GetFileFromPathAsync(filename), "file");
Log.WriteLine("have file");
var learningModel = await AsyncHelper.SyncFromAsync(LearningModel.LoadFromStorageFileAsync(file), "learningModel");
Log.WriteLine("loaded model");
Model model = new Model();
model._model = learningModel;
//LearningModelDeviceKind kind = LearningModelDeviceKind.DirectXHighPerformance;
LearningModelDeviceKind kind = LearningModelDeviceKind.Cpu;
model._device = new LearningModelDevice(kind);
model._session = new LearningModelSession(model._model, model._device);
Log.WriteLine("returning model");
return(model);
}
public static async Task <OnnxModel> CreateOnnxModel(StorageFile file)
{
LearningModel learningModel = null;
try
{
learningModel = await LearningModel.LoadFromStorageFileAsync(file);
}
catch (Exception e)
{
var exceptionStr = e.ToString();
System.Console.WriteLine(exceptionStr);
throw e;
}
return(new OnnxModel()
{
_learningModel = learningModel,
_session = new LearningModelSession(learningModel)
});
}
public static async Task <CustomVisionModel> CreateFromStorageFile(StorageFile file)
{
CustomVisionModel learningModel = new CustomVisionModel();
learningModel.model = await LearningModel.LoadFromStorageFileAsync(file);
IReadOnlyList <ILearningModelFeatureDescriptor> input = learningModel.model.InputFeatures.ToList();
MapFeatureDescriptor imgDesc = input[0] as MapFeatureDescriptor;
TensorFeatureDescriptor tfDesc = input[0] as TensorFeatureDescriptor;
learningModel.inputParameterName = input[0].Name;
learningModel.inputWidth = (int)tfDesc.Shape[2];
learningModel.inputHeight = (int)tfDesc.Shape[3];
IReadOnlyList <ILearningModelFeatureDescriptor> output = learningModel.model.OutputFeatures.ToList();
MapFeatureDescriptor imgDesc1 = output[0] as MapFeatureDescriptor;
TensorFeatureDescriptor tfDesc1 = output[0] as TensorFeatureDescriptor;
learningModel.outputParameterName = output[0].Name;
learningModel.session = new LearningModelSession(learningModel.model);
learningModel.binding = new LearningModelBinding(learningModel.session);
return(learningModel);
}
public static async Task <TinyYoloV2Model> CreateModel(StorageFile file)
{
LearningModel model = await LearningModel.LoadFromStorageFileAsync(file);
// todo investigate the input image description and the outputDescription.
//Get input and output features of the model
List <ILearningModelFeatureDescriptor> inputFeatures = model.InputFeatures.ToList();
List <ILearningModelFeatureDescriptor> outputFeatures = model.OutputFeatures.ToList();
// Retrieve the first input feature which is an image
var _inputImageDescription = inputFeatures.FirstOrDefault(
feature => feature.Kind == LearningModelFeatureKind.Image) as ImageFeatureDescriptor;
// Retrieve the first output feature which is a tensor
var _outputImageDescription = outputFeatures.FirstOrDefault(
feature => feature.Kind == LearningModelFeatureKind.Tensor) as TensorFeatureDescriptor;
return(new TinyYoloV2Model()
{
learningModel = model
});
}
public static async Task <RoadSignDetectionMLModel> Createmlmodel(StorageFile file)
{
LearningModel learningModel = null;
try
{
learningModel = await LearningModel.LoadFromStorageFileAsync(file);
}
catch (Exception e)
{
var exceptionStr = e.ToString();
System.Console.WriteLine(exceptionStr);
throw e;
}
var model = new RoadSignDetectionMLModel()
{
model = learningModel,
session = new LearningModelSession(learningModel),
};
model.binding = new LearningModelBinding(model.session);
return(model);
}
/// <summary>
/// Load the labels and model and initialize WinML
/// </summary>
/// <returns></returns>
private async Task LoadModelAsync(string modelFileName)
{
Debug.WriteLine("LoadModelAsync");
_evaluationLock.Wait();
{
m_binding = null;
m_model = null;
m_session = null;
_isReadyForEval = false;
try
{
// Start stopwatch
_perfStopwatch.Restart();
// Load Model
StorageFile modelFile = await StorageFile.GetFileFromApplicationUriAsync(new Uri($"ms-appx:///Assets/{modelFileName}.onnx"));
m_model = await LearningModel.LoadFromStorageFileAsync(modelFile);
// Stop stopwatch
_perfStopwatch.Stop();
// Setting preferred inference device given user's intent
m_inferenceDeviceSelected = _useGPU ? LearningModelDeviceKind.DirectXHighPerformance : LearningModelDeviceKind.Cpu;
m_session = new LearningModelSession(m_model, new LearningModelDevice(m_inferenceDeviceSelected));
// Debugging logic to see the input and output of ther model and retrieve dimensions of input/output variables
// ### DEBUG ###
foreach (var inputF in m_model.InputFeatures)
{
Debug.WriteLine($"input | kind:{inputF.Kind}, name:{inputF.Name}, type:{inputF.GetType()}");
int i = 0;
ImageFeatureDescriptor imgDesc = inputF as ImageFeatureDescriptor;
TensorFeatureDescriptor tfDesc = inputF as TensorFeatureDescriptor;
m_inWidth = (uint)(imgDesc == null ? tfDesc.Shape[3] : imgDesc.Width);
m_inHeight = (uint)(imgDesc == null ? tfDesc.Shape[2] : imgDesc.Height);
m_inName = inputF.Name;
Debug.WriteLine($"N: {(imgDesc == null ? tfDesc.Shape[0] : 1)}, " +
$"Channel: {(imgDesc == null ? tfDesc.Shape[1].ToString() : imgDesc.BitmapPixelFormat.ToString())}, " +
$"Height:{(imgDesc == null ? tfDesc.Shape[2] : imgDesc.Height)}, " +
$"Width: {(imgDesc == null ? tfDesc.Shape[3] : imgDesc.Width)}");
}
foreach (var outputF in m_model.OutputFeatures)
{
Debug.WriteLine($"output | kind:{outputF.Kind}, name:{outputF.Name}, type:{outputF.GetType()}");
int i = 0;
ImageFeatureDescriptor imgDesc = outputF as ImageFeatureDescriptor;
TensorFeatureDescriptor tfDesc = outputF as TensorFeatureDescriptor;
m_outWidth = (uint)(imgDesc == null ? tfDesc.Shape[3] : imgDesc.Width);
m_outHeight = (uint)(imgDesc == null ? tfDesc.Shape[2] : imgDesc.Height);
m_outName = outputF.Name;
Debug.WriteLine($"N: {(imgDesc == null ? tfDesc.Shape[0] : 1)}, " +
$"Channel: {(imgDesc == null ? tfDesc.Shape[1].ToString() : imgDesc.BitmapPixelFormat.ToString())}, " +
$"Height:{(imgDesc == null ? tfDesc.Shape[2] : imgDesc.Height)}, " +
$"Width: {(imgDesc == null ? tfDesc.Shape[3] : imgDesc.Width)}");
}
// ### END OF DEBUG ###
// Create output frame
_outputFrame?.Dispose();
_outputFrame = new VideoFrame(BitmapPixelFormat.Bgra8, (int)m_outWidth, (int)m_outHeight);
Debug.WriteLine($"Elapsed time: {_perfStopwatch.ElapsedMilliseconds} ms");
_isReadyForEval = true;
}
catch (Exception ex)
{
NotifyUser($"error: {ex.Message}", NotifyType.ErrorMessage);
Debug.WriteLine($"error: {ex.Message}");
}
}
_evaluationLock.Release();
}
/// <summary>
/// Initialize
/// </summary>
/// <param name="file">The ONNX file</param>
public async Task Init(StorageFile file)
{
this.model = await LearningModel.LoadFromStorageFileAsync(file);
this.session = new LearningModelSession(this.model);
}
public async Task LoadModelAsync(bool shouldUseGpu, bool resourceLoad)
{
try
{
// Parse labels from label file
var labelsTextAsset = Resources.Load(LabelsFileName) as TextAsset;
using (var streamReader = new StringReader(labelsTextAsset.text))
{
string line = "";
char[] charToTrim = { '\"', ' ' };
while (streamReader.Peek() >= 0)
{
line = streamReader.ReadLine();
line.Trim(charToTrim);
var indexAndLabel = line.Split(':');
if (indexAndLabel.Count() == 2)
{
_labels.Add(indexAndLabel[1]);
}
}
}
#if ENABLE_WINMD_SUPPORT
if (resourceLoad)
{
// Load from Unity Resources via awkward UWP streams and initialize model
using (var modelStream = new InMemoryRandomAccessStream())
{
var dataWriter = new DataWriter(modelStream);
var modelResource = Resources.Load(ModelFileName) as TextAsset;
dataWriter.WriteBytes(modelResource.bytes);
await dataWriter.StoreAsync();
var randomAccessStream = RandomAccessStreamReference.CreateFromStream(modelStream);
_model = await LearningModel.LoadFromStreamAsync(randomAccessStream);
var deviceKind = shouldUseGpu ? LearningModelDeviceKind.DirectXHighPerformance : LearningModelDeviceKind.Cpu;
_session = new LearningModelSession(_model, new LearningModelDevice(deviceKind));
}
}
else
{
try
{
var modelFile = await StorageFile.GetFileFromApplicationUriAsync(
new Uri($"ms-appx:///Data/StreamingAssets/SqueezeNet.onnx"));
_model = await LearningModel.LoadFromStorageFileAsync(modelFile);
var deviceKind = shouldUseGpu ? LearningModelDeviceKind.DirectXHighPerformance : LearningModelDeviceKind.Cpu;
_session = new LearningModelSession(_model, new LearningModelDevice(deviceKind));
}
catch (Exception e)
{
var exceptionStr = e.ToString();
//StatusBlock.text = exceptionStr;
}
}
// Get model input and output descriptions
var inputImageDescription =
_model.InputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Image)
as ImageFeatureDescriptor;
// Check if input is passed as image if not try to interpret it as generic tensor
if (inputImageDescription != null)
{
InputWidth = inputImageDescription.Width;
InputHeight = inputImageDescription.Height;
}
else
{
var inputTensorDescription =
_model.InputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor)
as TensorFeatureDescriptor;
InputWidth = (uint)inputTensorDescription.Shape[3];
InputHeight = (uint)inputTensorDescription.Shape[2];
}
_outputDescription =
_model.OutputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor)
as TensorFeatureDescriptor;
#endif
}
catch
{
#if ENABLE_WINMD_SUPPORT
_model = null;
#endif
throw;
}
}