How to run the benchmark obeject detection task on VOC dataset?

facebookresearch / vissl

VISSL is FAIR's library of extensible, modular and scalable components for SOTA Self-Supervised Learning with images.

https://vissl.ai

MIT License

3.25k stars 332 forks source link

How to run the benchmark obeject detection task on VOC dataset? #557

Open Yoooss opened 2 years ago

Yoooss commented 2 years ago

I want to run the benchmark object detection task on VOC0712 dataset. And I found the file "vissl/configs/config/benchmark/object_detection/voc0712/mocoV2/rn50_transfer_voc0712_detectron2_e2e.yaml" and the file "vissl/tools/object_detection_benchmark.py". And I noticed that the object detection task need the dependency "Detectron". However I didn't find the tutorial about how to install the compatible version of Detectron and using what command to run a object detection task ？May I get some help?

Yoooss commented 2 years ago

I used the command "python tools/object_detection_benchmark.py --config-file configs/config/benchmark/object_detection/voc0712/mocoV2/rn50_transfer_voc0712_detectron2_e2e.yaml --num-gpus 1 MODEL.WEIGHTS content/resnet_simclr_d2.torch" , to train the object detection task on VOC dataset on 1 GPU.

But It raised the error that "RuntimeError: radix_sort: failed on 1st step: cudaErrorInvalidDevice: invalid device ordinal". May I get some help about how to fix it?