o
    + iž  ã                   @  s\   d dl mZ d dlZd dlmZ ddlmZ er$d dlmZ ddl	m
Z
 G dd	„ d	eƒZdS )
é    )ÚannotationsN)ÚTYPE_CHECKINGé   )ÚQuantization)ÚLayer)ÚQuantConfigc                      s.   e Zd ZdZd‡ fdd„Zdddd„Z‡  ZS )ÚQATa  
    Tools used to prepare model for quantization-aware training.
    Args:
        config(QuantConfig): Quantization configuration

    Examples:
        .. code-block:: python

            >>> from paddle.quantization import QAT, QuantConfig
            >>> from paddle.quantization.quanters import FakeQuanterWithAbsMaxObserver
            >>> quanter = FakeQuanterWithAbsMaxObserver(moving_rate=0.9)
            >>> q_config = QuantConfig(activation=quanter, weight=quanter)
            >>> qat = QAT(q_config)
    Úconfigr   ÚreturnÚNonec                   s   t ƒ  |¡ d S )N)ÚsuperÚ__init__)Úselfr	   ©Ú	__class__© úc/home/app/PaddleOCR-VL-test/.venv_paddleocr/lib/python3.10/site-packages/paddle/quantization/qat.pyr   +   s   zQAT.__init__FÚmodelr   ÚinplaceÚboolc                 C  sL   |j sJ dƒ‚|r|nt |¡}| j |¡ |  || j¡ |  || j¡ |S )aÿ  
        Create a model for quantization-aware training.

        The quantization configuration will be propagated in the model.
        And it will insert fake quanters into the model to simulate the quantization.

        Args:
            model(Layer): The model to be quantized.
            inplace(bool): Whether to modify the model in-place.

        Return: The prepared model for quantization-aware training.

        Examples:
            .. code-block:: python

                >>> from paddle.quantization import QAT, QuantConfig
                >>> from paddle.quantization.quanters import FakeQuanterWithAbsMaxObserver
                >>> from paddle.vision.models import LeNet

                >>> quanter = FakeQuanterWithAbsMaxObserver(moving_rate=0.9)
                >>> q_config = QuantConfig(activation=quanter, weight=quanter)
                >>> qat = QAT(q_config)
                >>> model = LeNet()
                >>> quant_model = qat.quantize(model)
                >>> print(quant_model)
                LeNet(
                  (features): Sequential(
                    (0): QuantedConv2D(
                      (weight_quanter): FakeQuanterWithAbsMaxObserverLayer()
                      (activation_quanter): FakeQuanterWithAbsMaxObserverLayer()
                    )
                    (1): ObserveWrapper(
                      (_observer): FakeQuanterWithAbsMaxObserverLayer()
                      (_observed): ReLU()
                    )
                    (2): ObserveWrapper(
                      (_observer): FakeQuanterWithAbsMaxObserverLayer()
                      (_observed): MaxPool2D(kernel_size=2, stride=2, padding=0)
                    )
                    (3): QuantedConv2D(
                      (weight_quanter): FakeQuanterWithAbsMaxObserverLayer()
                      (activation_quanter): FakeQuanterWithAbsMaxObserverLayer()
                    )
                    (4): ObserveWrapper(
                      (_observer): FakeQuanterWithAbsMaxObserverLayer()
                      (_observed): ReLU()
                    )
                    (5): ObserveWrapper(
                      (_observer): FakeQuanterWithAbsMaxObserverLayer()
                      (_observed): MaxPool2D(kernel_size=2, stride=2, padding=0)
                    )
                  )
                  (fc): Sequential(
                    (0): QuantedLinear(
                      (weight_quanter): FakeQuanterWithAbsMaxObserverLayer()
                      (activation_quanter): FakeQuanterWithAbsMaxObserverLayer()
                    )
                    (1): QuantedLinear(
                      (weight_quanter): FakeQuanterWithAbsMaxObserverLayer()
                      (activation_quanter): FakeQuanterWithAbsMaxObserverLayer()
                    )
                    (2): QuantedLinear(
                      (weight_quanter): FakeQuanterWithAbsMaxObserverLayer()
                      (activation_quanter): FakeQuanterWithAbsMaxObserverLayer()
                    )
                  )
                )
        zfQuantization-Aware Training should work on training models. Please set training mode by model.train().)ZtrainingÚcopyÚdeepcopyÚ_configZ_specifyZ_convert_to_quant_layersZ_insert_activation_observers)r   r   r   Ú_modelr   r   r   Úquantize.   s   EÿzQAT.quantize)r	   r   r
   r   )F)r   r   r   r   r
   r   )Ú__name__Ú
__module__Ú__qualname__Ú__doc__r   r   Ú__classcell__r   r   r   r   r      s    r   )Ú
__future__r   r   Útypingr   r   r   Z	paddle.nnr   r	   r   r   r   r   r   r   Ú<module>   s   