search

airockchip / rknn_model_zoo

Apache License 2.0
965 stars 178 forks source link

Error convert yolov8 ! #25

Open lktinh2018 opened 1 year ago

lktinh2018 commented 1 year ago

Hello author,

I can convert to yolov5 but got following error with yolov8. Could you please help me?

E Catch exception when building RKNN model! E Traceback (most recent call last): E File "rknn/base/RKNNlib/RKNNnetbuilder.py", line 486, in rknn.base.RKNNlib.RKNNnetbuilder.RKNNNetBuilder.build_layer E File "rknn/base/RKNNlib/layer/RKNNlayer.py", line 293, in rknn.base.RKNNlib.layer.RKNNlayer.RKNNLayer.compute_tensor E File "rknn/base/RKNNlib/layer/divlayer.py", line 35, in rknn.base.RKNNlib.layer.divlayer.DivLayer.compute_out_tensor E File "/home/nexroot/Data/Virtual_Envs/env_rknn/lib/python3.8/site-packages/tensorflow/python/util/traceback_utils.py", line 153, in error_handler E raise e.with_traceback(filtered_tb) from None E File "/home/nexroot/Data/Virtual_Envs/env_rknn/lib/python3.8/site-packages/tensorflow/python/ops/math_ops.py", line 1517, in _truediv_python3 E raise TypeError(f"x and y must have the same dtype, " E TypeError: x and y must have the same dtype, got tf.float32 != tf.int32. E Please feedback the detailed log file to the RKNN Toolkit development team. E You can also check github issues: https://github.com/rockchip-linux/rknn-toolkit/issues ---> Export RKNN model E RKNN model data is None, please load model first! E RKNN model data is None, please load model first! WARNING: simulator does not support exporting pre_compiled model. Ignore! Convert Done!

marco0913 commented 1 year ago

Hi were you able to convert to yolov8 at the end?