RuntimeError: 0INTERNAL ASSERT FAILED at "/opt/conda/conda-bld/pytorch_1639180594101/work/torch/csrc/jit/ir/alias_analysis.cpp":584, please report a bug to PyTorch. We don't have an op for aten::fill_ but it isn't a special case. Argument types: Tensor, bool,

[Chen Xiangqi]

RuntimeError: 0INTERNAL ASSERT FAILED at "/opt/conda/conda-bld/pytorch_1639180594101/work/torch/csrc/jit/ir/alias_analysis.cpp":584, please report a bug to PyTorch. We don't have an op for aten::fill_ but it isn't a special case. Argument types: Tensor, bool,

[Taewoo Jeong]

I had exactly same error when trying to train model with following command:

GPUS_PER_NODE=1 ./tools/run_dist_launch.sh 1 python main.py \
    --with_box_refine \
    --two_stage \
    --eff_query_init \
    --eff_specific_head \
    --rho 0.1 \
    --use_enc_aux_loss \
    --coco_path {MY_OWN_COCO_PATH} \
    --output_dir {MY_OWN_OUTPUT_DIR_PATH}

Is their anyone dealt with this?

Thanks.

[Chen Xiangqi]

尝试使用以下命令训练模型时出现完全相同的错误：

GPUS_PER_NODE=1 ./tools/run_dist_launch.sh 1 python main.py \
    --with_box_refine \
    --two_stage \
    --eff_query_init \
    --eff_specific_head \
    --rho 0.1 \
    --use_enc_aux_loss \
    --coco_path {MY_OWN_COCO_PATH} \
    --output_dir {MY_OWN_OUTPUT_DIR_PATH}

他们的任何人都处理过这个吗？

谢谢。

There may be a problem with the compute_GFLOP function, which I made the code run by removing the call to that function

[Taewoo Jeong]

Thanks. I would try out.