2prolems occured to train my custom datasets

[VictorLee0214]

sry sir to interrupting u here！ when i use kpconv to train my custom datasets which is following ur data format to make done，it occurs these mishaps below with starting training：

1. raise ValueError("Expected more than 1 value per channel when training, got input size {}".format(size)) ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 8, 1])
2. ValueError: Expected more than 1 value per channel when training, got input size torch. Size([1, 8, 1])

i tried to debug this problem for 2days. i fonud problms are related to BN layer， but still confused now and i dont know hw to fix it up， hope u can help me

[LucHayward]

I would make sure that your data is actually being treated the same as the benchmark datasets and you end up with the same shape going into the model.

[VictorLee0214]

really thanks for ur reply，and ive already fixed up this problem. but uknow this work is a little bit hard and no one is knowing relevant algorithms knowledge around me. so could i follow ur ins or probably other contact way？

[WesamOM]

@VictorLee0214 I am trying to train my custom data on KPConv. Is there anyway I can contact you to get some help from you?

[VictorLee0214]

my wechat id is 1072412273 or my ig is ispw214 anyway email is K

Sent from my iPhone

------------------ Original ------------------ From: WesamOM @.> Date: Thu, Jun 22, 2023 1:34 AM To: HuguesTHOMAS/KPConv-PyTorch @.> Cc: VictorLee0214 @.>, Mention @.> Subject: Re: [HuguesTHOMAS/KPConv-PyTorch] 2prolems occured to train my customdatasets (Issue #207)