{-jDddlmZddlZddlmZmZddlmZmZddl Z ddl m cm Z ddl m Z ddlmZddlmZmZmZmZmZddlmZdd lmZd d iZgZerdd l mZdd lmZGddeZGdde j Z!Gdde j Z"ddZ# dd dZ$dS)!) annotationsN) TYPE_CHECKING TypedDict) NotRequiredUnpack)nn) ParamAttr)Conv2DDropoutLinear MaxPool2DReLU)Uniform)get_weights_path_from_urlalexnet)zUhttps://paddle-imagenet-models-name.bj.bcebos.com/dygraph/AlexNet_pretrained.pdparams 7f0f9f737132e02732d75a1459d98a43)Tensor)Size2ceZdZUded<dS)_AlexNetOptionszNotRequired[int] num_classesN)__name__ __module__ __qualname____annotations__\/var/www/html/banglarbhumi/venv/lib/python3.11/site-packages/paddle/vision/models/alexnet.pyrr.s%%%%%%rrc.eZdZ ddfd ZddZxZS) ConvPoolLayerNinput_channelsintoutput_channels filter_sizerstridepaddingstdvfloatgroupsact str | NonereturnNonec Xt|dkrtnd|_t ||||||t t | |t t | ||_tddd|_ dS)Nrelu initializer) in_channels out_channels kernel_sizer&r'r* weight_attr bias_attrr)r5r&r') super__init__rr0r r r_convr _pool) selfr"r$r%r&r'r(r*r+ __class__s rr;zConvPoolLayer.__init__3s !VmmDFFF &(#!gteT.B.BCCCGTE4,@,@AAA     1QBBB rinputsrc||}|j||}||}|S)N)r<r0r=r>r@xs rforwardzConvPoolLayer.forwardNs> JJv   9  ! A JJqMMr)r!N)r"r#r$r#r%rr&rr'rr(r)r*r#r+r,r-r.r@rr-r)rrrr;rD __classcell__r?s@rr r 2seCCCCCCC6rr c:eZdZUdZded<d d fd Zd d ZxZS)AlexNetaAlexNet model from `"ImageNet Classification with Deep Convolutional Neural Networks" `_. Args: num_classes (int, optional): Output dim of last fc layer. If num_classes <= 0, last fc layer will not be defined. Default: 1000. Returns: :ref:`api_paddle_nn_Layer`. An instance of AlexNet model. Examples: .. code-block:: pycon >>> import paddle >>> from paddle.vision.models import AlexNet >>> alexnet = AlexNet() >>> x = paddle.rand([1, 3, 224, 224]) >>> out = alexnet(x) >>> print(out.shape) paddle.Size([1, 1000]) r#rr-r.ct||_dtjdz }t ddddd|d |_dtjd z }t dd d d d|d |_dtjdz }td ddd d tt| |tt| ||_ dtjdz }tdddd d tt| |tt| ||_ dtjdz }t dddd d |d |_ |jdkr;dtjdz }tdd|_t!ddtt| |tt| ||_tdd|_t!ddtt| |tt| ||_t!d|tt| |tt| ||_dSdS)Ng?ikr8@ r9r0)r+i@r!iir1)r&r'r6r7i i ri$g?downscale_in_infer)pmodei) in_features out_featuresr6r7)r:r;rmathsqrtr _conv1_conv2r r r_conv3_conv4_conv5r _drop1r _fc6_drop2_fc7_fc8)r>rr(r?s rr;zAlexNet.__init__qs &TY{+++#Ar2q!TvFFF TYz***#BQ1dGGG TY{+++   !gteT.B.BCCCGTE4,@,@AAA    TY{+++   !gteT.B.BCCCGTE4,@,@AAA    TY{+++#CaAtHHH  a  ;///D!C.BCCCDK'!%'4%2F2FGGG#t0D0DEEE DI"C.BCCCDK !%'4%2F2FGGG#t0D0DEEE DI  (%'4%2F2FGGG#t0D0DEEE DIII# rr@rc||}||}||}tj|}||}tj|}||}|jdkrtj |dd}| |}| |}tj|}| |}| |}tj|}||}|S)Nrr!) start_axis stop_axis)rYrZr[Fr0r\r]rpaddleflattenr^r_r`rarbrBs rrDzAlexNet.forwards KK   KKNN KKNN F1II KKNN F1II KKNN  a  qQ"===A AA ! Aq A AA ! Aq A ! Ar)rJ)rr#r-r.rE)rrr__doc__rr;rDrFrGs@rrIrIVsq04444444lrrIarchstr pretrainedboolkwargsUnpack[_AlexNetOptions]r-c tdi|}|rq|tvs J|dtt|dt|d}tj|}|||S)NzJ model do not have a pretrained model now, you should set pretrained=Falserr!r)rI model_urlsrrhload load_dict)rkrmromodel weight_pathparams r_alexnetrxs   f  E z!!! _ _ _"!!0 t Q D!1!!4   K((  LrFc td|fi|S)a AlexNet model from `"ImageNet Classification with Deep Convolutional Neural Networks" `_. Args: pretrained (bool, optional): Whether to load pre-trained weights. If True, returns a model pre-trained on ImageNet. Default: False. **kwargs (optional): Additional keyword arguments. For details, please refer to :ref:`AlexNet `. Returns: :ref:`api_paddle_nn_Layer`. An instance of AlexNet model. Examples: .. code-block:: pycon >>> import paddle >>> from paddle.vision.models import alexnet >>> # Build model >>> model = alexnet() >>> # Build model and load imagenet pretrained weight >>> # model = alexnet(pretrained=True) >>> x = paddle.rand([1, 3, 224, 224]) >>> out = model(x) >>> print(out.shape) paddle.Size([1, 1000]) r)rx)rmros rrrsB Iz 4 4V 4 44r)rkrlrmrnrorpr-rI)F)rmrnrorpr-rI)% __future__rrWtypingrrtyping_extensionsrrrhpaddle.nn.functionalr functionalrgpaddle.base.param_attrr paddle.nnr r r r rpaddle.nn.initializerrpaddle.utils.downloadrrr__all__rpaddle._typingrrLayerr rIrxrrrrrs#"""""  21111111 ,,,,,,>>>>>>>>>>>>>>))))));;;;;;  &$$$$$$&&&&&)&&&!!!!!BH!!!HdddddbhdddN(!5!5!5!5!5!5!5r