Code samples contain errors and is missing a function

[krisbock]

Copying the code snippets into a file and executing doesn't work, as the testAccuracy function doesn't test for GPU. So if you are using a GPU and the train function loads the train dataset/model weights onto the GPU, you receive the following error:

RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same or input should be a MKLDNN tensor and weight is a dense tensor

Fixing this by adding the following:

images, labels = images.to(device), labels.to(device)

to the testAccuracy function after "images, labels = data"

In addition, the main method calls a testModelAccuracy function that doesn't exist. From the code sample referenced in GitHub (https://github.com/microsoft/Windows-Machine-Learning/blob/master/Samples/Tutorial%20Samples/PyTorch%20Image%20Classification/PyTorchTraining%20-%20Image%20Classification/PyTorchTraining.py), it appears a testClasses function is required.

---

Document Details

⚠ Do not edit this section. It is required for learn.microsoft.com ➟ GitHub issue linking.

• ID: b54afee7-1e3b-98f3-c25d-c0d71b82bf97
• Version Independent ID: b75b9ae6-de0e-acc3-b9cd-4918422bf7aa
• Content: Use PyTorch to train your image classification model
• Content Source: docs/windows-ml/tutorials/pytorch-train-model.md
• Product: windows
• GitHub Login: @QuinnRadich
• Microsoft Alias: quradic

[wizardpisces]

Problem still exist, any progress?