I have been happily following along the tutorial for Faster RCNN in torchvision… and now trained multiple models with custom backbones (resnet101 and resnet101_wide_2)
However, when I go to call the evaluate function… my result is being rejected with an AssertionError
I ave augmented some of the printing outs myself so I can see where I crash…
dataset found to be instance of torchvision.datasets.CocoDetection
dataset found to be instance of torchvision.datasets.CocoDetection, returning dataset.coco
entering image target loop
.
attempting to sync
/envs/pytorch_150/lib/python3.7/site-packages/torch/nn/functional.py:2854: UserWarning: The default behavior for interpolate/upsample with float scale_factor will change in 1.6.0 to align with other frameworks/libraries, and use scale_factor directly, instead of relying on the computed output size. If you wish to keep the old behavior, please set recompute_scale_factor=True. See the documentation of nn.Upsample for details.
warnings.warn("The default behavior for interpolate/upsample with float scale_factor will change "
/opt/conda/conda-bld/pytorch_1587428266983/work/torch/csrc/utils/python_arg_parser.cpp:756: UserWarning: This overload of nonzero is deprecated:
nonzero(Tensor input, *, Tensor out)
Consider using one of the following signatures instead:
nonzero(Tensor input, *, bool as_tuple)
received outputs
getting result
coco_evaluator.update() calling...
Traceback (most recent call last):
File "/faster-rcnn-torchvision/model_components/scoring.py", line 149, in <module>
evaluate(data_conf=config_json, model_conf=model_conf)
File "/faster-rcnn-torchvision/model_components/scoring.py", line 86, in evaluate
result = exec_evaluate(model, testing_data_loader, device=device)
File "/anaconda3/envs/pytorch_150/lib/python3.7/site-packages/torch/autograd/grad_mode.py", line 15, in decorate_context
return func(*args, **kwargs)
File "/faster-rcnn-torchvision/model_components/scoring.py", line 122, in exec_evaluate
coco_evaluator.update(res)
File "~/faster-rcnn-torchvision/model_components/model/coco_eval.py", line 37, in update
coco_dt = loadRes(self.coco_gt, results) if results else COCO()
File "~/faster-rcnn-torchvision/model_components/model/coco_eval.py", line 256, in loadRes
'Results do not correspond to current coco set'
AssertionError: Results do not correspond to current coco set
seems the area is right here… line 122
