How to use opencl to speed up yolov3-tiny on rk3399


#1

WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 3, 416, 416, ‘float32’), (16, 3, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 16, 208, 208, ‘float32’), (32, 16, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 32, 104, 104, ‘float32’), (64, 32, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 64, 52, 52, ‘float32’), (128, 64, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 128, 26, 26, ‘float32’), (256, 128, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 256, 13, 13, ‘float32’), (512, 256, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 512, 13, 13, ‘float32’), (1024, 512, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 1024, 13, 13, ‘float32’), (256, 1024, 1, 1, ‘float32’), (1, 1), (0, 0), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 256, 13, 13, ‘float32’), (128, 256, 1, 1, ‘float32’), (1, 1), (0, 0), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 384, 26, 26, ‘float32’), (256, 384, 3, 3, ‘float32’), (1, 1), (1, 1), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 256, 26, 26, ‘float32’), (21, 256, 1, 1, ‘float32’), (1, 1), (0, 0), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression.
WARNING:autotvm:Cannot find config for target=opencl -device=mali, workload=(‘conv2d’, (1, 512, 13, 13, ‘float32’), (21, 512, 1, 1, ‘float32’), (1, 1), (0, 0), (1, 1), ‘NCHW’, ‘float32’). A fallback configuration is used, which may bring great performance regression


#2

Try following the ARM GPU autotuning tutorial example here:
https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_mobile_gpu.html

You will need to add your network to the get_network here.