exit(ret) # init runtime environment print("--> Init runtime environment") ret = rknn.init_runtime() sdk_version = rknn.get_sdk_version() print(sdk_version) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
librknn_runtime version 1.3.2 (736df77 build: 2020-04-23 09:26:03 base: 1112)通过 RKNN ...
opset_version=11 ) print(f'Successfully exported ONNX model: {output_file}') 导出模型过程中可能出现RuntimeError: Failed to export an ONNX attribute 'onnx::Gather', since it's not constant, please try to make things (e.g., kernel size) static if possible问题。官方代码中使用了自适应的...
done --> Init runtime environment I RKNN: [22:59:21.100] RKNN Runtime Information: librknnrt version: 1.5.0 (e6fe0c678@2023-05-25T08:09:20) I RKNN: [22:59:21.112] RKNN Driver Information: version: 0.8.2 I RKNN: [22:59:21.127] RKNN Model Information: version: 4, toolkit versi...
librknn_runtime version 1.5.1 版本的 跟 rknn toolkit 1.4对不上 strings librknn_runtime.so | grep build vsi_nn_kernel_add_build_option librknn_runtime version 1.5.1 (161f53f build: 2020-11-05 15:12:30 base: 1126)编译rknpu-1.5.0/rknn/rknn_api/examples/rknn_mobilenet_demo ...
"rknpu runtime lib and server" config BR2_PACKAGE_RKNPU2_ARCH string depends on BR2_ARCH = "arm" || BR2_ARCH = "aarch64" default "armhf" if BR2_ARCH="arm" default BR2_ARCH config BR2_PACKAGE_RKNPU2_SOC string depends on BR2_PACKAGE_RK3566_RK3568 || BR2_PACKAGE_RK3562 || \ ...
onnx_session = onnxruntime.InferenceSession(onnx_path) input_name = get_in_out_name('inputs', onnx_session) #得到所有的输入节点名字 dummy_input = { input_name[0]: np.float32(xdata), #假设这里只有一个输入节点 } result = onnx_session.run(None, input_feed=dummy_input) ...
-->Initruntime environment INPUTransfer:StartingNPUTransferClient,Transferversion2.1.0(b5861e7@2020-11-23T11:50:36) DRKNNAPI: === DRKNNAPI:RKNNVERSION: DRKNNAPI:API:1.6.0(79320debuild:2020-12-2910:56:36) DRKNNAPI:DRV:1.7.0(7880361build:2021-08-1614:05:08) DRKNNAPI...
板端还有一部分，就是rknn runtime环境，包含一组C API库以及与NPU进行通信的驱动模块，可执行程序等。RKNN软件栈可以帮助用户快速将AI模型部署到瑞芯微芯片上。整体框架如下：为了使用 RKNPU，用户需要首先在计算机上运行 RKNN-Toolkit2 工具，将训练好的模型转换为 RKNN 格式的模型，然后使用 RKNN C API 或 ...
librknn_runtimeversion 1.7.3 (5047ff8 build: 2022-08-13 12:11:22 base: 1131) done --> Running model done --> Evaluate model performance W When performing performance evaluation, inputs can be set to None to use fake inputs.