abhi1kumar / DEVIANT

[ECCV 2022] Official PyTorch Code of DEVIANT: Depth Equivariant Network for Monocular 3D Object Detection
https://arxiv.org/abs/2207.10758
MIT License
203 stars 29 forks source link

Waymo Dataset Training Epochs #9

Closed Yutong-gannis closed 1 year ago

Yutong-gannis commented 1 year ago

I'm trying to train deviant on waymo, and I read 1051.yaml. Is it ok to train 30 epochs on waymo? Or is it just a pretrained model config, and I should train more epochs.

abhi1kumar commented 1 year ago

Hi @Yutong-gannis Thank you for your interest in DEVIANT.

Is it ok to train 30 epochs on waymo?

Yes, training for 30 epochs should be fine since Waymo is a huge dataset. We also trained our Waymo model for 30 epochs.

Other paper such as CaDDN train Waymo model for 10 epochs. See their Section 4 Training and Inference Details.

Also, please consider starring our repo and close your older issues.

Yutong-gannis commented 1 year ago

OK, thanks