You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I downloaded your ACO_resnet34.ckpt and try to test a single image to predict the steering value.
You had mentioned in your manuscript that the L1 error of steering prediction on NuScenes dataset are normalized to [0, 1].
When I fed an image [1, 3, 224, 224] into the model as:
net = resnet34()
net.load_state_dict(torch.load('ACO_resnet34.ckpt'), strict=False)
output = net(image)
The output is [1, 1000] which is not a steering value.
Did I miss something or misunderstood?
Best
Hugh
The text was updated successfully, but these errors were encountered:
Hi,
I downloaded your ACO_resnet34.ckpt and try to test a single image to predict the steering value.
You had mentioned in your manuscript that the L1 error of steering prediction on NuScenes dataset are normalized to [0, 1].
When I fed an image [1, 3, 224, 224] into the model as:
The output is [1, 1000] which is not a steering value.
Did I miss something or misunderstood?
Best
Hugh
The text was updated successfully, but these errors were encountered: