o #j"@stddlZddlmZddlmZddlmZgZddiZGdd d ejZ Gd d d ejZ dd dZ dddZ dS)N)nn)get_weights_path_from_url)ConvNormActivationzmobilenetv1_1.0)zAhttps://paddle-hapi.bj.bcebos.com/models/mobilenetv1_1.0.pdparamsZ 3033ab1975b1670bef51545feb65fc45cs$eZdZfddZddZZS)DepthwiseSeparablec sXtt|t||d|dt||d|_tt||t||dddd|_dS)N) kernel_sizestridepaddinggroupsr)r r r )super__init__rint_depthwise_conv_pointwise_conv)self in_channels out_channels1 out_channels2 num_groupsr scale __class__a/var/www/html/Deteccion_Ine/venv/lib/python3.10/site-packages/paddle/vision/models/mobilenetv1.pyr s      zDepthwiseSeparable.__init__cCs||}||}|S)N)rr)rxrrrforward<s  zDepthwiseSeparable.forward)__name__ __module__ __qualname__rr __classcell__rrrrrs rcs*eZdZdZd fdd ZddZZS) MobileNetV1awMobileNetV1 model from `"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications" `_. Args: scale (float, optional): Scale of channels in each layer. Default: 1.0. num_classes (int, optional): Output dim of last fc layer. If num_classes <= 0, last fc layer will not be defined. Default: 1000. with_pool (bool, optional): Use pool before the last fc layer or not. Default: True. Returns: :ref:`api_paddle_nn_Layer`. An instance of MobileNetV1 model. Examples: .. code-block:: python >>> import paddle >>> from paddle.vision.models import MobileNetV1 >>> model = MobileNetV1() >>> x = paddle.rand([1, 3, 224, 224]) >>> out = model(x) >>> print(out.shape) [1, 1000] ?Tc sTt||_g|_||_||_tdtd|dddd|_|j t td|dddd|ddd }|j ||j t td|dd dd|dd d }|j ||j t td |d d d d|dd d }|j ||j t td |d d d d|ddd }|j ||j t td |d d d d|ddd }|j ||j t td |d dd d|ddd } |j | t dD]!} |j t td|dddd|ddt | dd } |j | q|j t td|dddd|ddd } |j | |j t td|dddd|ddd } |j | |rtd|_|dkr(ttd|||_dSdS)Nr rr)rZ out_channelsr r r @)rrrrr rZconv2_1)ZsublayernameZconv2_2Zconv3_1Zconv3_2Zconv4_1iZconv4_2Zconv5_iZconv5_6Zconv6r)r rrdwsl num_classes with_poolrrconv1Z add_sublayerrappendrangestrrZAdaptiveAvgPool2D pool2d_avgZLinearfc)rrr,r-Zdws21Zdws22Zdws31Zdws32Zdws41Zdws42itmpZdws56Zdws6rrrr^s                      zMobileNetV1.__init__cCsR||}|jD]}||}q|jr||}|jdkr't|d}||}|S)Nrr)r.r+r-r2r,paddleflattenr3)rrZdwsrrrrs       zMobileNetV1.forward)r#r$T)rrr __doc__rrr!rrrrr"Bs  r"FcKsZtdi|}|r+|tvsJ|dtt|dt|d}t|}|||S)NzJ model do not have a pretrained model now, you should set pretrained=Falserrr)r" model_urlsrr6load load_dict)arch pretrainedkwargsmodelZ weight_pathparamrrr _mobilenets   rAr#cKs"tdt||fd|i|}|S)aMobileNetV1 from `"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications" `_. Args: pretrained (bool, optional): Whether to load pre-trained weights. If True, returns a model pre-trained on ImageNet. Default: False. scale (float, optional): Scale of channels in each layer. Default: 1.0. **kwargs (optional): Additional keyword arguments. For details, please refer to :ref:`MobileNetV1 `. Returns: :ref:`api_paddle_nn_Layer`. An instance of MobileNetV1 model. Examples: .. code-block:: python >>> import paddle >>> from paddle.vision.models import mobilenet_v1 >>> # Build model >>> model = mobilenet_v1() >>> # Build model and load imagenet pretrained weight >>> # model = mobilenet_v1(pretrained=True) >>> # build mobilenet v1 with scale=0.5 >>> model_scale = mobilenet_v1(scale=0.5) >>> x = paddle.rand([1, 3, 224, 224]) >>> out = model(x) >>> print(out.shape) [1, 1000] Z mobilenetv1_r)rAr1)r=rr>r?rrr mobilenet_v1s" rB)F)Fr#) r6rZpaddle.utils.downloadropsr__all__r9ZLayerrr"rArBrrrrs   # 6