周彬

icon company DUT icon location dalian，China icon location Do what I can do.

Results 5 comments of 周彬

D_data_loss and G_discriminator_loss don't change

> l2_loss_weight change to what

D_data_loss and G_discriminator_loss don't change

> Even if I replace ReLU with LeakyReLU, the losses do not change basically. U can change the L2_loos_weight. It could be help.

Training loss does not change, and validation FDE error is super high

activation using relu or leakyrelu

KeyError: 'traj'

> ./tools/uniad_dist_eval.sh ./projects/configs/stage2_e2e/base_e2e.py ./ckpts/uniad_base_e2e.pth 1 && python ./tools/analysis_tools/visualize/run.py --predroot ./output/results.pkl --out_folder ./output_visualize --demo_video test_demo.avi --project_to_cam True If you want to get visualization results, before end-to-end training or eval, you can...

Difference between scene_level_offset_embedding and scene_level_ego_embedding.

> scene leve hi，why rotation_transform used in local frame.