FastDeploy
FastDeploy copied to clipboard
PaddlePaddle
RuntimeError: tensor type 0 is not supported
[ERROR] csrc/fastdeploy/backends/ort/utils.cc(30)::GetOrtDtype Unrecognized fastdeply data type:FDDataType::UINT8.
Traceback (most recent call last):
File "main.py", line 9, in
测试如下：
1 import cv2
2 import fastdeploy.vision as vision
3
4
5 model = vision.detection.PPYOLOE("m/model.pdmodel",
6 "m/model.pdiparams",
7 "m/infer_cfg.yml")
8 im = cv2.imread("Images/1.jpg")
9 result = model.predict(im.copy())
10 print(result)
11
12 vis_im = vision.vis_detection(im, result, score_threshold=0.5)
13 cv2.imwrite("vis_image.jpg", vis_im)
十分抱歉，模型涉及到公司项目，是不能提供的， 模型是使用PaddleDetection_YOLOSeries/configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml这个配置训练的
python tools/export_model.py -c configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml -o weights=output/ppyoloe_crn_l_300e_coco/best_model.pdparams
这个是我导出模型时，运行的指令
@ZhangLe-fighting 试下先卸载现有的fastdeploy
安装下0.2.1版本的fastdeploy，下载链接如下
- CPU版本： https://bj.bcebos.com/fastdeploy/dev/wheels/fastdeploy_python-0.2.1-cp36-cp36m-manylinux1_x86_64.whl
- GPU版本： https://bj.bcebos.com/fastdeploy/dev/wheels/fastdeploy_gpu_python-0.2.1-cp36-cp36m-manylinux1_x86_64.whl
麻烦测试后反馈下哈
具体可以查看每个模型对应的api说明， example/vision/detection/paddledetection/python和cpp目录下的README分别有其python/api说明
您好： python tools/export_model.py -c configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml -o weights=output/ppyoloe_crn_l_300e_coco/best_model.pdparams
RuntimeError: /onnxruntime_src/onnxruntime/core/providers/cuda/cuda_call.cc:122 bool onnxruntime::CudaCall(ERRTYPE, const char*, const char*, ERRTYPE, const char*) [with ERRTYPE = cudaError; bool THRW = true] /onnxruntime_src/onnxruntime/core/providers/cuda/cuda_call.cc:116 bool onnxruntime::CudaCall(ERRTYPE, const char*, const char*, ERRTYPE, const char*) [with ERRTYPE = cudaError; bool THRW = true] CUDA failure 3: initialization error ; GPU=490585848 ; hostname=Thor ; expr=cudaSetDevice(info_.device_id);
将export的模型部署在英伟达T4卡上时，出现了如上的错误，不太明白是什么原因导致的，如果您了解的话，麻烦解答一下。
如果你只是需要TensorRT部署，可以自行编译FastDeploy。 当前Cuda 10.1上只能支持TensorRT
我去TensorRT官网看了一下，没有找到CUDA10.1对应的发行版本，请问需要下载哪个版本呢？
如果你只是需要TensorRT部署，可以自行编译FastDeploy。 当前Cuda 10.1上只能支持TensorRT
我去TensorRT官网看了一下，没有找到CUDA10.1对应的发行版本，请问需要下载哪个版本呢？
下载10.2对应的即可
麻烦您提供一下python编译选项的设置...
git clone https://github.com/PaddlePaddle/FastDeploy.git
cd FastDeploy/python
export ENABLE_TRT_BACKEND=ON
export ENABLE_VISION=ON
export ENABLE_OPENVINO_BACKEND=ON
export WITH_GPU=ON
export TRT_DIRECTORY=/trt/xxxx
export CUDA_DIRECTORY=/usr/loca/cuda-10.2
python setup.py build
python setup.py bdist_wheel
环境搞了一天没搞好，555555555555555 打算明天用 paddleinference 部署试一下...
方便的话，可以加我们微信群，了解下你具体遇到啥问题
InvalidArgumentError: The type of data we are trying to retrieve does not match the type of data currently contained in the container. [Hint: Expected dtype() == paddle::experimental::CppTypeToDataType<T>::Type(), but received dtype():10 != paddle::experimental::CppTypeToDataType<T>::Type():12.] (at /paddle/paddle/phi/core/dense_tensor.cc:137) [operator < elementwise_mul > error]
使用paddleinference推理，没用到infer_cfg.yml这个文件，然后按照教程将图片使用cv2读取，resize\transpose之后输入出现如上错误，请问有没有具体的教程呀，就是图片输入如何处理，然后加载pdmodel和pdiparams文件包含infer_cfg.yml的信息嘛，比如target_size之类的信息，上面错误好像是输入处理的不对...
环境搞了一天没搞好，555555555555555 打算明天用 paddleinference 部署试一下...
方便的话，可以加我们微信群，了解下你具体遇到啥问题
好滴
CMake Error at CMakeLists.txt:281 (find_package): By not providing "FindOpenCV.cmake" in CMAKE_MODULE_PATH this project has asked CMake to find a package configuration file provided by "OpenCV", but CMake did not find one.
Could not find a package configuration file provided by "OpenCV" with any of the following names:
OpenCVConfig.cmake
opencv-config.cmake
Add the installation prefix of "OpenCV" to CMAKE_PREFIX_PATH or set "OpenCV_DIR" to a directory containing one of the above files. If "OpenCV" provides a separate development package or SDK, be sure it has been installed.
进行编译的时候出现这个错误，但是我并没有找到这个文件