Updated README.md for #364

See #364 for a deep description of the problem

examples/image_classifier/README.md

@@ -45,6 +45,7 @@
    from torchvision import models
    import torch
    model = models.densenet161(pretrained=True)
+   model.eval()
    example_input = torch.rand(1, 3, 224, 224)
    traced_script_module = torch.jit.trace(model, example_input)
    traced_script_module.save("dense161.pt")