tensorflow-upstream: hip_code_object.cpp:92: guarantee(false && "hipErrorNoBinaryForGpu: Coudn't find binary for current devices!")

GPU: 5700xt

When using the following Docker image:

rocm/tensorflow     latest              d83f8c9d5c96        2 weeks ago         10.3GB

with ROCm installed on the Docker host as explained here: https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html

I get the following error when executing TensorFlow ops:

root@apoehlmann:/root# python3
Python 3.6.9 (default, Jul 17 2020, 12:50:27) 
[GCC 8.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> tf.add(1,2)
2020-09-06 20:14:03.889728: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libamdhip64.so
/src/external/hip-on-vdi/rocclr/hip_code_object.cpp:92: guarantee(false && "hipErrorNoBinaryForGpu: Coudn't find binary for current devices!")
Aborted (core dumped)

and the Python console dies. I started the container with the alias mentioned in the corresponding Docker registry: https://hub.docker.com/r/rocm/tensorflow

I get the same error when I try to run tensorflow ops on the host.

Googling this issue yields only a handful of results so I feel like I might have some misconfiguration but I cannot figure out what it is.