v2x-vit
v2x-vit copied to clipboard
[ECCV2022] Official Implementation of paper "V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer"
In https://github.com/DerrickXuNu/v2x-vit/blob/main/v2xvit/tools/train.py, I just noticed that train_loader = DataLoader(opencood_train_dataset, batch_size=hypes['train_params']['batch_size'], num_workers=8, collate_fn=opencood_train_dataset.collate_batch_train, shuffle=True, pin_memory=False, drop_last=True) **val_loader** = DataLoader(opencood_validate_dataset, batch_size=hypes['train_params']['batch_size'], num_workers=8, collate_fn=**opencood_train_dataset**.collate_batch_train, shuffle=False, pin_memory=False, drop_last=True) I'm not sure that it is...
HI Derrick, Thanks for providing the v2x-vit model and other baseline methods codes! The whole codes are very neat! Do you have plan to support multi-gpu training? Thanks,
Hi, could you please share the data on Baiduyun? The Google Drive is unstable, and the data downloading often interrupt
This is not an issue, but I just want to point out some useful tips for those who try to train better: 1) Multi-GPU training can easily increase some points....
These slight code adjustments aim to enhance conciseness of the code by employing optimized NumPy functions.
I am a postgraduate scholar from the Institute of Automation of the Chinese Academy of Sciences. Your research paper has been highly advantageous to me. I aspire to augment my...
Hi, runsheng, In early_fusion_dataset.py line 99 to 110, seems like ‘unique_indices’ should also be updated. unique_indices = unique_indices[:object_bbx_center_valid.shape[0]] Or it may cause error in BasePostprocessor.generate_gt_bbx() line 91 https://github.com/DerrickXuNu/v2x-vit/blob/main/v2xvit/data_utils/datasets/early_fusion_dataset.py#L99