`AttributeError: 'Tensor' object has no attribute 'astype'. Did you mean: 'dtype'?` when prediction in image classification

[PeerlessMonster]

python stm32ai_main.py --config-path ./config_file_examples_pt --config-name prediction_n6_config.yaml

throw an exception

Traceback (most recent call last):
  File "D:\Softwares\miniconda3\envs\st_zoo\Lib\site-packages\clearml\binding\hydra_bind.py", line 230, in _patched_tas
k_function
    return task_function(a_config, *a_args, **a_kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\workspace\stm32ai-modelzoo-services\image_classification\stm32ai_main.py", line 337, in main
    _process_mode(cfg=cfg)
  File "D:\workspace\stm32ai-modelzoo-services\image_classification\stm32ai_main.py", line 118, in _process_mode
    predictor.predict()
  File "D:\workspace\stm32ai-modelzoo-services\image_classification\tf\src\prediction\onnx_predictor.py", line 196, in 
predict
    scores = self._get_scores(image_processed)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\workspace\stm32ai-modelzoo-services\image_classification\tf\src\prediction\onnx_predictor.py", line 119, in 
_get_scores
    imagee = ai_interp_input_quant(self.ai_runner_interpreter, image_processed, '.onnx')
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\workspace\stm32ai-modelzoo-services\common\utils\models_utils.py", line 35, in ai_interp_input_quant
    out_data = resc_data.astype(np.uint8)
               ^^^^^^^^^^^^^^^^
AttributeError: 'Tensor' object has no attribute 'astype'. Did you mean: 'dtype'?

complete output from terminal: 2026_03_16_23_28_39.log

content of config file:

operation_mode: prediction
general:
  display_figures: false
model:
  framework: torch
  model_name: mobilenetv2_a100_pt
  pretrained: true
  pretrained_dataset: imagenet
  input_shape:
  - 3
  - 224
  - 224
  model_path: ../../stm32ai-modelzoo/image_classification/mobilenetv2_pt/Public_pretrainedmodel_public_dataset/Imagenet/mobilenetv2_a100_pt_224/mobilenetv2_a100_pt_224_qdq_int8.onnx
dataset:
  dataset_name: imagenet
  class_names: ''
  classes_file_path: ./datasets/deployment_labels_imagenet.txt
  num_classes: 1000
  prediction_path: ./datasets/mini-imagenet/test/n01532829/
  workers: 4
  no_prefetcher: true
preprocessing:
  rescaling:
    scale: None
    offset: None
  resizing:
    interpolation: random
    aspect_ratio: fit
  color_mode: rgb
  normalization:
    mean:
    - 0.485
    - 0.456
    - 0.406
    std:
    - 0.229
    - 0.224
    - 0.225
prediction:
  profile: profile_O3
  input_type: uint8
  output_type: int8
  input_chpos: chlast
  output_chpos: chlast
  target: stedgeai_n6
tools:
  stedgeai:
    path_to_stedgeai: D:/Softwares/ST/STEdgeAI/4.0/Utilities/windows/stedgeai.exe
mlflow:
  uri: ./pt/src/experiments_outputs/mlruns

version:
STEdgeAI Core = 4.0
STM32CubeIDE = 1.17.0