jefferyzhang 发表于 2019-11-26 16:31
本地仿真正常么?
jefferyzhang 发表于 2019-11-26 16:57
是不是转换模型的rknn toolkit和本地api版本不匹配。
如果你换了新固件,toolkit也要用新的来转rknn ...
jefferyzhang 发表于 2019-11-26 17:23
toolkit 1.2.1 必须搭配 0.9.9或者1.2.0的Drv
如果0.9.9不能用,你应该是踩到坑了,可以把模型和转换代码一 ...
- from rknn.api import RKNN
- rknn = RKNN(verbose=True)
- #
- TF_model = rknn.load_tensorflow(
- tf_pb="./tf_models/model.pb",
- inputs= ['Placeholder'],
- outputs= ['output_node'],
- input_size_list=[[1,256,1]]
- )
- ret = rknn.build(do_quantization=False, pre_compile=False)
- ret = rknn.export_rknn(export_path = './rknn/test_one.rknn')
复制代码
kaishi 发表于 2019-12-1 15:05
My platform has the same issue sometimes. It doesn't happen all the time.
Below is the printout duri ...
jefferyzhang 发表于 2019-11-26 17:23
toolkit 1.2.1 必须搭配 0.9.9或者1.2.0的Drv
如果0.9.9不能用,你应该是踩到坑了,可以把模型和转换代码一 ...
- /root/anaconda3/envs/tensorflow/lib/python3.6/site-packages/onnx_tf/common/__init__.py:87: UserWarning: FrontendHandler.get_outputs_names is deprecated. It will be removed in future release.. Use node.outputs instead.
- warnings.warn(message)
- W Verbose file path is invalid, debug info will not dump to file.
- -->loading model
- load model done:0
- --> init runtime environment
- I When we run model on simulator, we do not need set perf_debug to get layer performance.
- #productname=VSI SIMULATOR, pid=0x82
- Created VX Thread: 50aea700
- Segmentation fault (core dumped)
复制代码
欢迎光临 Toybrick (https://t.rock-chips.com/) | Powered by Discuz! X3.3 |