search

MarSaKi / VLN-BEVBert

[ICCV 2023} Official repo of "BEVBert: Multimodal Map Pre-training for Language-guided Navigation"
184 stars 5 forks source link

No found /img_features/depth_14x14.hdf5 #10

Open weimingz996 opened 8 months ago

weimingz996 commented 8 months ago

image Hi， Can you upload this file？thanks! By the way, can you show me your all file hierarchical structures for me（It can be screenshots）？

yueluhhxx commented 8 months ago

Thank you for sharing such an interesting job. When I tried to execute the 'CUDA_VISIBLE_DEVICES=0,1,2,3 bash run_pt/run_r2r.bash 2333' command, I found that I was also missing the file named 'depth_14x14.hdf5'. Could you please tell me where I can get this file? Look forward to your reply~

yueluhhxx commented 8 months ago

Thank you for sharing such an interesting job. When I tried to execute the 'CUDA_VISIBLE_DEVICES=0,1,2,3 bash run_pt/run_r2r.bash 2333' command, I found that I was also missing the file named 'depth_14x14.hdf5'. Could you please tell me where I can get this file? Look forward to your reply~

I guess this is because I didn't execute the "python precompute_features/grid_depth.py" command before performing the training on R2R-CE.

MarSaKi commented 8 months ago

Yes, you are right, just run precompute_features/grid_depth.py to get this file. Thx,