导出模型后的推理失败

[edenatoniuma]

问题确认 Search before asking

• 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer.

请提出你的问题 Please ask your question

训练完的模型用来推理是没有问题的，具体命令如下：

python .\infer.py -c 
..\configs\ppyolo\ppyolo_r50vd_dcn_1x_coco.yml

-o weights=G:\projects\GraduationProject\PaddleDetection-release-2.8\batch_size_1_model\model.pdparams 

--infer_img=G:\projects\GraduationProject\PaddleDetection-release-2.8\dataset\area_coco\images\Video1_0005.jpeg

使用tools/export.py导出模型后，也成功生成了对应的文件，然后用命令推理：

文件：
目录: G:\projects\GraduationProject\PaddleDetection-release-2.8\inference_model\ppyolo_r50vd_dcn_1x_coco


Mode                 LastWriteTime         Length Name
----                 -------------         ------ ----
-a----         2025/1/27     16:49            371 infer_cfg.yml
-a----         2025/1/27     16:49      186775554 model.pdiparams
-a----         2025/1/27     16:49          42202 model.pdiparams.info
-a----         2025/1/27     16:49         530112 model.pdmodel


推理：
python .\deploy\python\infer.py 

--model_dir=G:/projects/GraduationProject/PaddleDetection-release-2.8/inference_model/ppyolo_r50vd_dcn_1x_coco 

--image_file=G:/projects/GraduationProject/PaddleDetection-release-2.8/dataset/area_coco/images/Video1_0005.jpeg 

--device GPU

出现报错原因如下：

-----------  Running Arguments -----------
action_file: None
batch_size: 1
camera_id: -1
collect_trt_shape_info: False
combine_method: nms
cpu_threads: 1
device: GPU
enable_mkldnn: False
enable_mkldnn_bfloat16: False
image_dir: None
image_file: G:/projects/GraduationProject/PaddleDetection-release-2.8/dataset/area_coco/images/Video1_0005.jpeg
match_metric: ios
match_threshold: 0.6
model_dir: G:/projects/GraduationProject/PaddleDetection-release-2.8/inference_model/ppyolo_r50vd_dcn_1x_coco
output_dir: output
overlap_ratio: [0.25, 0.25]
random_pad: False
reid_batch_size: 50
reid_model_dir: None
run_benchmark: False
run_mode: paddle
save_images: True
save_mot_txt_per_img: False
save_mot_txts: False
save_results: False
scaled: False
slice_infer: False
slice_size: [640, 640]
task_type: Detection
threshold: 0.5
tracker_config: None
trt_calib_mode: False
trt_max_shape: 1280
trt_min_shape: 1
trt_opt_shape: 640
tuned_trt_shape_file: shape_range_info.pbtxt
use_coco_category: False
use_dark: True
use_fd_format: False
use_gpu: False
video_file: None
window_size: 50
------------------------------------------
-----------  Model Configuration -----------
Model Arch: YOLO
Transform Order:
--transform op: Resize
--transform op: NormalizeImage
--transform op: Permute
--------------------------------------------
Traceback (most recent call last):
  File ".\deploy\python\infer.py", line 1266, in <module>
    main()
  File ".\deploy\python\infer.py", line 1199, in main
    detector = eval(detector_func)(
  File ".\deploy\python\infer.py", line 108, in __init__
    self.predictor, self.config = load_predictor(
  File ".\deploy\python\infer.py", line 1098, in load_predictor
    predictor = create_predictor(config)
RuntimeError: (NotFound) Cannot open file G:/projects/GraduationProject/PaddleDetection-release-2.8/inference_model/ppyolo_r50vd_dcn_1x_coco, please confirm whether the file is normal.
  [Hint: Expected static_cast<bool>(fin.is_open()) == true, but received static_cast<bool>(fin.is_open()):0 != true:1.] (at C:\home\workspace\Paddle\paddle\fluid\inference\api\analysis_predictor.cc:2808)

我调试了很久也没有找到原因，不知道为什么报错消息是打开一个文件夹报错。

环境信息（python3.8）

(paddle) PS G:\projects\GraduationProject\PaddleDetection-release-2.8> pip list | findstr /b /i "paddle"
paddle2onnx              1.3.0
paddledet                0.0.0
paddlefsl                1.1.0
paddlenlp                2.6.1
paddleocr                2.9.1
paddlepaddle-gpu         3.0.0b1

[cuicheng01]

请问，导出的模型的扩展名是啥样子的呢？

[edenatoniuma]

请问，导出的模型的扩展名是啥样子的呢？

这样子的

[edenatoniuma]

infer脚本是可以运行的。

[HHHLLL1]

同样的问题，用的ppyoloe模型，导出的模型，paddleDection和paddle inference api都用不了