FastDeploy

FastDeploy copied to clipboard

使用use_trt_backend()报错

6

主要代码： option = fd.RuntimeOption() option.use_gpu() option.use_trt_backend() start = time.time() print("time:",start) model = vision.detection.YOLOv3("assets/model.pdmodel", "assets/model.pdiparams", "assets/infer_cfg.yml",option) 报错： [ERROR] fastdeploy/backends/tensorrt/trt_backend.cc(573)::CreateTrtEngineFromOnnx Failed to parse ONNX model by TensorRT. [ERROR] fastdeploy/backends/tensorrt/trt_backend.cc(265)::InitFromOnnx Failed to create...

fanyuanxiang

[Model] Integrate FlyCV image processing library

1

### PR types ### PR changes ### Describe

jiangjiajun

### PR types(PR类型) Backend ### Describe 通过PaddleLite的芯原 TIM-VX支持Rockchip RV1126 Soc的部署。 mobilenetv1全量化模型 NPU avg latency：6.12 ms CPU avg latency：331.93 ms

yeliang2258

### PR types(PR类型) Serving ### Describe - Add basepreprocess && basepostprocess class for serving

felixhjh

### PR types(PR类型) Other ### Describe - Support download pre-trained models for fastdeploy using one interface, i.e. fastdeploy.hub.download_model - Support record users' downloading action for statistics. - Support configuration for...

rainyfly

PaddleSeg Python部署示例：推理完成后，不输出可视化结果

4

Win10 64位，Python3.8，fastdeploy-python==0.4.0，按照下面链接的步骤做实例分割推理，不输出可视化结果 https://github.com/PaddlePaddle/FastDeploy/tree/develop/examples/vision/segmentation/paddleseg/python —————————————————————————————————— 测试模型：Unet_cityscapes_without_argmax_infer 运行指令：python infer.py --model Unet_cityscapes_without_argmax_infer --image 0.jpg --device cpu 使用图片：  —————————————————————————————————— 报错信息： result shape is: [480 640] Traceback (most recent call last): File "infer.py", line...

stq054188

[Model] yolo use external stream, avoid reallocating output tensors

1

### PR types(PR类型) Model ### Describe - Avoid reallocating output tensors - Use external stream when using cuda preprocessing, avoid reallocating cuda streams

wang-xinyu

[Backend] support ipu in paddle inference backend.

3

### PR types(PR类型) Backend ### Describe 增加paddle inference backend对IPU的支持。

czr-gc

[Model] add PFLD model

2

### PR types Model ### Describe - **[FaceAlignment]** Support new model [PFLD](https://github.com/Hsintao/pfld_106_face_landmarks) - Add new result struct ```FaceAlignmentResult```

wjj19950828

[Serving] Modify OCR related model functions to support Serving

1

### PR types(PR类型) [Serving] Modify OCR related model functions to support Serving ### Describe 1.Modify OCR related model functions to support Serving 2.Add new pybind interface for preprocess and postprocess

HexToString