MemoryError: std::bad_alloc

[idiot499]

Hello, I am running:
2) python export_detections.py configs/magic-point_coco_export.yaml magic-point_synth --pred_only --batch_size=5 --export_name=magic-point_coco-export1 reported an error terminate called after throwing an instance of 'std::bad_alloc'

1. Specific error message:

2. The graphics card model:

3. The Library:
   tensorflow-gpu==1.12.0
   cudatoolkit==9.0
   cudnn==7.3.1

4. The Environment:
   ubuntu 18.04

Can you please provide me some help? I can run successfully: 1) python experiment.py train configs/magic-point_shapes.yaml magic-point_synth ，but I can't run: 2) python export_detections.py configs/magic-point_coco_export.yaml magic-point_synth --pred_only --batch_size=5 --export_name=magic-point_coco-export1

[how-way-wang]

Excuse me, has the problem been solved? My problem is quite similar to yours. My environment is Ubuntu 22.04, using WSL2.