KPConv-PyTorch
KPConv-PyTorch copied to clipboard
About the C2 fast confusion
C1 is right. But for C2 fast_confusion error: ValueError: operands could not be broadcast together with shapes (135,) (2468,) I do not why self.val_prob is 2468 that leads to the error. Please check.
I get the same error. I'm running on my data with 8 classes instead of 40.
vec_conf = np.bincount(true * num_classes + pred) ValueError: operands could not be broadcast together with shapes (707,) (2468,)
@WanpengShao did you solve this error?
ok, found the issue, in the file ModelNet40.py
the number of test and train samples are hard coded:
# Number of models and models used per epoch
if self.train:
self.num_models = 9843
if config.epoch_steps and config.epoch_steps * config.batch_num < self.num_models:
self.epoch_n = config.epoch_steps * config.batch_num
else:
self.epoch_n = self.num_models
else:
self.num_models = 2468
self.epoch_n = min(self.num_models, config.validation_size * config.batch_num)
The number 2468 comes from here. If you change those to the actual number of split samples in your test.txt and train.txt files then the error should be resolved.