sample is referenced before initialized in classification_dataset.py, many thanks!

loiccordone / object-detection-with-spiking-neural-networks

Repository code for the IJCNN 2022 paper "Object Detection with Spiking Neural Networks on Automotive Event Data"

MIT License

57 stars 12 forks source link

sample is referenced before initialized in classification_dataset.py, many thanks! #33

Open kwunhangwong opened 6 months ago

kwunhangwong commented 6 months ago

def __getitem__(self, index):
    sparse_tensor, target = self.samples[index]
    sample = sample.sparse_resize_(
        (self.T, sample.size(1), sample.size(2), self.C), 3, 1
    ).to_dense().permute(0,3,1,2)
    sample = T.Resize((128,128), T.InterpolationMode.NEAREST)(sample)
    return sample, target

kwunhangwong commented 6 months ago

Sorry I have another issue, but thanks for sharing your code really appreciate!! It seems there is a problem with NCARS dataset processing. When I plotted the events of NCARS, they were all gone to the top left corner, aggregated in a very small area.

108360215 commented 5 months ago

Hello, how do you solve the sample initialized problem? I also need to give it a value. Or you replaced from sparse_tensor? Thanks!

108360215 commented 5 months ago

@u3556440 do you implement this classification.py successfully? I got some troubles.. Could you help me plz! I will appreciate it.