Closed zeakey closed 6 years ago
I report two errors in the official train faster-rcnn on voc tutorial.
Errors are produced by directly run the train_faster_rcnn.py script.
ValueError: Model faster_rcnn_resnet50_v1b_voc is not supported.
It seems that the default model, resnet50_v1b, is nolonger supported, using --network resnet50_v2a resolves the error;
resnet50_v1b
--network resnet50_v2a
AttributeError: 'FasterRCNN' object has no attribute 'short'.
AttributeError: 'FasterRCNN' object has no attribute 'short'
I don't know what's happening with this error.
OS
uname -a Linux kai-PC 4.15.0-34-generic #37~16.04.1-Ubuntu SMP Tue Aug 28 10:44:06 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
Python
python --version Python 3.5.2
MxNet & Gluoncv
python -c "import mxnet, gluoncv; print(mxnet.__version__);print(gluoncv.__version__)" 1.3.0 0.2.0
Please use nightly build. https://github.com/dmlc/gluon-cv#installation
Closing due to lack of activity, ping me if not resolved.
I report two errors in the official train faster-rcnn on voc tutorial.
Errors are produced by directly run the train_faster_rcnn.py script.
ValueError: Model faster_rcnn_resnet50_v1b_voc is not supported.
It seems that the default model,
resnet50_v1b
, is nolonger supported, using--network resnet50_v2a
resolves the error;AttributeError: 'FasterRCNN' object has no attribute 'short'
.I don't know what's happening with this error.
Environments:
OS
Python
MxNet & Gluoncv