search

apache / mxnet

Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
https://mxnet.apache.org
Apache License 2.0
20.77k stars 6.79k forks source link

Mxnet errors while running gluoncv from mxnet docker image #21187

Open mikelibg opened 1 year ago

mikelibg commented 1 year ago

As I was not able to run the gluoncv with mxnet on my M1 mac, so I tried to use the docker file as python interpreter.

mxnet/python:1.9.1_aarch64_cpu_py3 which is running Python 3.7.13, mxnet==1.9.1b20220517 I installed gluoncv==0.10.5.post0 in that docker and getting the following two errors:

mxnet.base.MXNetError: MXNetError: Invalid Parameter format for std expect tuple of but value='[0.229, 0.224, 0.225]', in operator _image_normalize(name="", std="[0.229, 0.224, 0.225]", mean="[0.485, 0.456, 0.406]")

when trying to do gluoncv.data.transforms.presets.segmentation.test_transform on the image

and:

mxnet.base.MXNetError: MXNetError: Invalid Parameter format for layout expect int or None but value='', in operator Convolution(name="", stride="(2, 2)", no_bias="True", kernel="(3, 3)", layout="NCHW", num_group="1", dilate="(1, 1)", pad="(1, 1)", num_filter="64")

when trying to load model using model = gluoncv.model_zoo.get_model('fcn_resnet101_voc', pretrained=True)

github-actions[bot] commented 1 year ago

Welcome to Apache MXNet (incubating)! We are on a mission to democratize AI, and we are glad that you are contributing to it by opening this issue. Please make sure to include all the relevant context, and one of the @apache/mxnet-committers will be here shortly. If you are interested in contributing to our project, let us know! Also, be sure to check out our guide on contributing to MXNet and our development guides wiki.

ZacCarrico commented 1 year ago

Getting the same error using mxnet/python:1.9.1_aarch64_cpu_py3. Interestingly, the error only occurs when run in this docker container for this image on raspberrypi (arm64). It works when run in mxnet/python:1.9.1_cpu_py3 on a linux desktop (amd64).