pipeline.py 3.3 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485
  1. # Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
  2. #
  3. # Licensed under the Apache License, Version 2.0 (the "License");
  4. # you may not use this file except in compliance with the License.
  5. # You may obtain a copy of the License at
  6. #
  7. # http://www.apache.org/licenses/LICENSE-2.0
  8. #
  9. # Unless required by applicable law or agreed to in writing, software
  10. # distributed under the License is distributed on an "AS IS" BASIS,
  11. # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  12. # See the License for the specific language governing permissions and
  13. # limitations under the License.
  14. from typing import Any, Dict, List, Optional, Union
  15. import numpy as np
  16. from ...models.image_multilabel_classification.result import MLClassResult
  17. from ...utils.hpi import HPIConfig
  18. from ...utils.pp_option import PaddlePredictorOption
  19. from ..base import BasePipeline
  20. class ImageMultiLabelClassificationPipeline(BasePipeline):
  21. """Image Multi Label Classification Pipeline"""
  22. entities = "image_multilabel_classification"
  23. def __init__(
  24. self,
  25. config: Dict,
  26. device: str = None,
  27. pp_option: PaddlePredictorOption = None,
  28. use_hpip: bool = False,
  29. hpi_config: Optional[Union[Dict[str, Any], HPIConfig]] = None,
  30. ) -> None:
  31. """
  32. Initializes the class with given configurations and options.
  33. Args:
  34. config (Dict): Configuration dictionary containing model and other parameters.
  35. device (str): The device to run the prediction on. Default is None.
  36. pp_option (PaddlePredictorOption): Options for PaddlePaddle predictor. Default is None.
  37. use_hpip (bool, optional): Whether to use the high-performance
  38. inference plugin (HPIP) by default. Defaults to False.
  39. hpi_config (Optional[Union[Dict[str, Any], HPIConfig]], optional):
  40. The default high-performance inference configuration dictionary.
  41. Defaults to None.
  42. """
  43. super().__init__(
  44. device=device, pp_option=pp_option, use_hpip=use_hpip, hpi_config=hpi_config
  45. )
  46. self.threshold = config["SubModules"]["ImageMultiLabelClassification"].get(
  47. "threshold", None
  48. )
  49. image_multilabel_classification_model_config = config["SubModules"][
  50. "ImageMultiLabelClassification"
  51. ]
  52. self.image_multilabel_classification_model = self.create_model(
  53. image_multilabel_classification_model_config
  54. )
  55. image_multilabel_classification_model_config["batch_size"]
  56. def predict(
  57. self,
  58. input: Union[str, List[str], np.ndarray, List[np.ndarray]],
  59. threshold: Union[float, dict, list, None] = None,
  60. **kwargs
  61. ) -> MLClassResult:
  62. """Predicts image classification results for the given input.
  63. Args:
  64. input (Union[str, list[str], np.ndarray, list[np.ndarray]]): The input image(s) or path(s) to the images.
  65. **kwargs: Additional keyword arguments that can be passed to the function.
  66. Returns:
  67. TopkResult: The predicted top k results.
  68. """
  69. yield from self.image_multilabel_classification_model(
  70. input=input,
  71. threshold=self.threshold if threshold is None else threshold,
  72. )