trouble to use latest docker image

[jwgim]

I have used docker image version of cdeep3m2 which works great. Recently, my server down with kernel panic because of another reason. cdeep3m doesn't work anymore even with fresh setup.

previous environment was
Ubuntu 16.04, nvidia driver& CUDA version - ? docker image tag-2020_04_17

current environment is
Ubuntu 20.04, Driver Version: 510.60.02 CUDA Version: 11.6, docker image tag-latest

Here are error logs:
I0406 05:20:27.934778 358 solver.cpp:60] Solver scaffolding done.
I0406 05:20:27.939950 358 caffe.cpp:214] Starting Optimization
I0406 05:20:27.939957 358 solver.cpp:288] Solving
I0406 05:20:27.939961 358 solver.cpp:289] Learning Rate Policy: fixed
I0406 05:20:27.945987 358 solver.cpp:341] Iteration 0, Testing net (#0)
F0406 05:20:27.961402 358 im2col.cu:231] Check failed: error == cudaSuccess (209 vs. 0) no kernel image is available for execution on the device
*** Check failure stack trace: ***
@ 0x7f71329eb5cd google::LogMessage::Fail()
@ 0x7f71329ed433 google::LogMessage::SendToLog()
@ 0x7f71329eb15b google::LogMessage::Flush()
@ 0x7f71329ede1e google::LogMessageFatal::~LogMessageFatal()
@ 0x7f713323e94e caffe::im2col_nd_gpu<>()
@ 0x7f7133076961 caffe::BaseConvolutionLayer<>::conv_im2col_gpu()
@ 0x7f7133076b66 caffe::BaseConvolutionLayer<>::forward_gpu_gemm()
@ 0x7f71331fd85c caffe::ConvolutionLayer<>::Forward_gpu()
@ 0x7f713316c8e2 caffe::Net<>::ForwardFromTo()
@ 0x7f713316c9f7 caffe::Net<>::ForwardPrefilled()
@ 0x7f713319ca07 caffe::Solver<>::Test()
@ 0x7f713319d40e caffe::Solver<>::TestAll()
@ 0x7f713319d55b caffe::Solver<>::Step()
@ 0x7f713319e185 caffe::Solver<>::Solve()
@ 0x40b9eb train()
@ 0x40768c main
@ 0x7f713149f840 __libc_start_main
@ 0x407e39 _start
@ (nil) (unknown)

When I use current environment with 2020_04_17 tagged docker image, works fine.