Hello, I am trying to run the non-distributed training for refcoco+ on my local machine using the below command:
./scripts/nondist_run.sh refcoco/train_end2end.py ./cfgs/refcoco/base_gt_boxes_4x16G.yaml ./checkpoints/
I have been receiving the below error:
[Partial Load] non pretrain keys: ['final_mlp.2.weight', 'final_mlp.2.bias']
PROGRESS: 0.00%
./scripts/nondist_run.sh: line 3: 13529 Segmentation fault (core dumped) python "$1" --cfg "$2" --model-dir "$3"
I have tried reducing both the BATCH_SIZE and NUM_WORKERS_PER_GPU to 1, but still get the same error. I have a GeForce GTX 1070. Do I need more resources? Or is there another possible problem with my setup?
Hello, I am trying to run the non-distributed training for refcoco+ on my local machine using the below command: ./scripts/nondist_run.sh refcoco/train_end2end.py ./cfgs/refcoco/base_gt_boxes_4x16G.yaml ./checkpoints/
I have been receiving the below error: [Partial Load] non pretrain keys: ['final_mlp.2.weight', 'final_mlp.2.bias'] PROGRESS: 0.00% ./scripts/nondist_run.sh: line 3: 13529 Segmentation fault (core dumped) python "$1" --cfg "$2" --model-dir "$3"
I have tried reducing both the BATCH_SIZE and NUM_WORKERS_PER_GPU to 1, but still get the same error. I have a GeForce GTX 1070. Do I need more resources? Or is there another possible problem with my setup?
Thank you.