training problem #154
Description
Hi, I would like to use my own images for your project. Thank you for providing an excellent training method. I reduced the number of training iterations and even tried using only 5 images, but after 6 to 7 minutes, it is still stuck at 0% without any improvement. Is this normal? Additionally, I noticed that both my GPU and CPU utilization are not increasing. I would like to know what method you use for training. In this situation, what steps should I take to address this issue?
Training T : 1.0
Number of trainable parameters of Point CNF: 1691649
loss: 4329673.50000: 0%| | 0/3 [06:53<?, ?it/s]