rknn-toolkit

rknn-toolkit copied to clipboard

torch的EfficientNet转RK失败，是Identity算子不支持吗？

3

log如下，查了下OP手册torch没包括Identity算子，是这个原因吗？ I Start importing pytorch... D import clients finished W Pt model version is 1.6(same as you can check through ), but the installed pytorch is 1.8.0. This may cause...

MinYounZhang

tensorflow.dtypes' has no attribute 'int64

2

请问为什么在使用torch.argmax()时，转rknn报出如下错误： E Catch exception when building RKNN model! E Traceback (most recent call last): E File "rknn/api/rknn_base.py", line 616, in rknn.api.rknn_base.RKNNBase.build E File "rknn/api/rknn_base.py", line 1523, in rknn.api.rknn_base.RKNNBase._quantize2 E File...

Myspacel

E Try match layer: Pad_Pad_14:out0 failed! The OP of this layer may not yet support, please check document <RKNN_OP_Support.md>!

1

--> Config model done --> Loading model W Call onnx.optimizer.optimize fail, skip optimize /usr/local/lib/python3.6/dist-packages/rknn/api/rknn.py:150: RuntimeWarning: overflow encountered in exp ret = self.rknn_base.load_onnx(model, inputs, input_size_list, outputs) W The follow op is...

ttttttt12

jit-pt 转rknn失败

1

训练的yolov3+mobilenet v2模型, 先转onnx 再转rknn,可以正常实现,运行推断也正常. 但是将训练的模型,通过orch.jit.trace()接口转换后,再转rknn时候,在load模型的时候,rknn.load_pytorch()接口出错. 转换程序如下: D Output shape of layer slice_at_conv_raw_prob.2_345_299: [[1, 9, 64, 64, -2]] I Parse prim::Constant at 3859 I Parse aten::arange at 3860 I Parse aten::unsqueeze...

BaseJiang

RuntimeError: [enforce fail at inline_container.cc:209] . file not found: best_model_50_torchscript/constants.pkl E Please feedback the detailed log file <log_feedback_to_the_rknn_toolkit_dev_team.log> to the RKNN Toolkit development team. E You can also check github issues: https://gith

1

RuntimeError: [enforce fail at inline_container.cc:209] . file not found: best_model_50_torchscript/constants.pkl E Please feedback the detailed log file to the RKNN Toolkit development team. E You can also check github issues:...

ttttttt12

Cannot download anyfile from baidu

2

Hello, I couldn't download anything from the baidu storage. Is there any other mirror?

ngoclinhle

onnx模型转rknn时不支持topk算子

1

topk算子目前rknn的版本不支持吗，有什么可以替换的算子吗

xuan-xuan6

1.7.1版本 yolov5 onnx转rknn模型出错AttributeError: 'NoneType' object has no attribute 'get_input_layers'

4

========== parser_config ========== NPU_VERSION:1 RK_device_platform:RK1808 input_example:['/workspace/data/BackgroundIdentification/data_0331_1000/images/train/20210706141643_1_19.jpg'] pre_compile:off config:{'target_platform': 'RK1808', 'quantized_dtype': 'asymmetric_affine-u8', 'quantized_algorithm': 'normal', 'optimization_level': 3, 'mean_values': [[0, 0, 0]], 'std_values': [[255, 255, 255]], 'mmse_epoch': 3, 'do_sparse_network': False, 'output_optimize': 1, 'batch_size':...

colinChing

⭐ Some useful advice that may help you ⭐

4

**RKNN-Toolkit:** 1. For **yolov5**, please refer here : https://github.com/airockchip/rknn_model_zoo/tree/main/models/vision/object_detection/yolov5-pytorch 2. For loading **already quantized model**, refer here: https://github.com/rockchip-linux/rknn-toolkit/tree/master/examples/common_function_demos/load_quantized_model 3. For how to **export a .pt model**(torchscript format), refer here: https://github.com/rockchip-linux/rknn-toolkit/blob/master/examples/pytorch/resnet18/test.py...

zen-xingle

Error 量化模型无法使用！

3

Hi， 1 pytorch 量化后的pth 或者pt模型，可以直接在rk3399 ARM和rk3399pro NPU上运行吗？ 2 如果我用其他格式的量化模型，比如onnx ，怎样才能在rk3399和rk3399rpo NPU上跑通？ 3 rk3399pro NPU支持哪些格式的量化模型？还是说必须先转换为rockchip框架的模型，才能进一步量化，最后部署在板子上？ BR

2050airobert