Build error with Dockerfile.gpu

[uersoy]

I get the below error when I tried this on my Jetson Xavier NX. Anyone had a similar issue? My device has ARM v8 processor, OS Type 64-bit

~/gpt-2$ sudo docker build --tag gpt-2 -f Dockerfile.gpu .

Sending build context to Docker daemon 4.712MB Step 1/9 : FROM tensorflow/tensorflow:1.12.0-gpu-py3 ---> 413b9533f92a Step 2/9 : LABEL com.nvidia.volumes.needed="nvidia_driver" ---> Using cache ---> 05ba5810d177 Step 3/9 : LABEL com.nvidia.cuda.version="${CUDA_VERSION}" ---> Using cache ---> ebbf24a709ba Step 4/9 : ENV NVIDIA_VISIBLE_DEVICES=all NVIDIA_DRIVER_CAPABILITIES=compute,utility NVIDIA_REQUIRE_CUDA="cuda>=8.0" LANG=C.UTF-8 ---> Using cache ---> ca9f7af4b739 Step 5/9 : RUN pip3 install -r requirements.txt ---> Running in 7839cb2e5a88 standard_init_linux.go:211: exec user process caused "exec format error" The command '/bin/sh -c pip3 install -r requirements.txt' returned a non-zero code: 1

[mikolasan]

@uersoy , it seems for me that you have modified the Dockerfile, because pip3 install is not the 5th step. You are missing

RUN mkdir /gpt-2
WORKDIR /gpt-2
ADD . /gpt-2

which copies essential code and requirements.txt file to the container.