ёi'SSKJr SSKJrJr SSKJrJr SSKrSSK J s J r SSKJ r SSK Jr SSKJrJrJrJr SSKJr \(aSS KJr SS KJr "S S \5r/rS SS.r"SS\ R85r"SS\ R85r"SS\ R85rSSjr SSSjjr!SSSjjr"g)) annotations) TYPE_CHECKING TypedDict) NotRequiredUnpackN)nn) ParamAttr)AdaptiveAvgPool2DConv2DDropout MaxPool2D)get_weights_path_from_url)Tensor)Size2c*\rSrSr%S\S'S\S'Srg)_SqueezeNetOptions#zNotRequired[int] num_classeszNotRequired[bool] with_poolN)__name__ __module__ __qualname____firstlineno____annotations____static_attributes__r_/var/www/html/banglarbhumi/venv/lib/python3.13/site-packages/paddle/vision/models/squeezenet.pyrr#s%%$$rr)z[https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SqueezeNet1_0_pretrained.pdparams 30b95af60a2178f03cf9b66cd77e1db1)z[https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/SqueezeNet1_1_pretrained.pdparams a11250d3a1f91d7131fd095ebbf09eee) squeezenet1_0 squeezenet1_1cN^\rSrSrSSU4SjjjrSSjrSrU=r$) MakeFireConv6c h>[TU]5 [UUUU[5[5S9Ulg)N)padding weight_attr bias_attr)super__init__r r _conv)selfinput_channelsoutput_channels filter_sizer' __class__s rr+MakeFireConv.__init__7s2    ! k   rcTURU5n[R"U5nU$)N)r,Frelu)r-xs rforwardMakeFireConv.forwardHs! JJqM FF1Ir)r,)r) r.intr/r9r0rr'rreturnNone)r6rr:rrrrrr+r7r __classcell__r1s@rr$r$6sL            "rr$cJ^\rSrSrSU4SjjrSSjrSrU=r$)MakeFireNc>[TU]5 [XS5Ul[X#S5Ul[X$SSS9Ulg)N)r')r*r+r$r, _conv_path1 _conv_path2)r-r.squeeze_channelsexpand1x1_channelsexpand3x3_channelsr1s rr+MakeFire.__init__OsB !.AF '(8aP' !Q rcURU5nURU5nURU5n[R"X4/SS9$)NrCaxis)r,rErFpaddleconcat)r-inputsr6x1x2s rr7MakeFire.forward]sC JJv    a    a }}bXA..r)r,rErF) r.r9rGr9rHr9rIr9r:r;rPrr:rr<r>s@rr@r@NsB            //rr@cn^\rSrSr%SrS\S'S\S'S\S'S S U4S jjjrSS jrS rU=r $) SqueezeNetdaSqueezeNet model from `"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size" `_. Args: version (str): Version of SqueezeNet, which can be "1.0" or "1.1". num_classes (int, optional): Output dim of last fc layer. If num_classes <= 0, last fc layer will not be defined. Default: 1000. with_pool (bool, optional): Use pool before the last fc layer or not. Default: True. Returns: :ref:`api_paddle_nn_Layer`. An instance of SqueezeNet model. Examples: .. code-block:: pycon >>> import paddle >>> from paddle.vision.models import SqueezeNet >>> # build v1.0 model >>> model = SqueezeNet(version='1.0') >>> # build v1.1 model >>> # model = SqueezeNet(version='1.1') >>> x = paddle.rand([1, 3, 224, 224]) >>> out = model(x) >>> print(out.shape) paddle.Size([1, 1000]) strversionr9rboolrc D>[TU]5 XlX lX0lSS/nX;d SUSU35eURS:Xa[ SSSS[ 5[ 5S 9Ul[SSS S 9Ul [SS S S 5Ul [SS S S 5Ul [SSSS5Ul [SSSS5Ul[SSSS5Ul[SSSS5Ul[SS SS5Ul[SS SS5UlO[ SS SSS[ 5[ 5S9Ul[SSS S 9Ul [S S S S 5Ul [SS S S 5Ul [SSSS5Ul [SSSS5Ul[SSSS5Ul[SSSS5Ul[SS SS5Ul[SS SS5Ul['SSS9Ul[ SUS[ 5[ 5S9Ul[-S5Ulg)N1.01.1zsupported versions are z but input version is rD`)strider(r)r) kernel_sizerar'@ 0iirC)rar'r(r)g?downscale_in_infer)pmode)r(r))r*r+rYrrr r r,r _poolr@_conv1_conv2_conv3_conv4_conv5_conv6_conv7_conv8r _drop_conv9r _avg_pool)r-rYrrsupported_versionsr1s rr+SqueezeNet.__init__s"  &"#U^, %&8%99OPWy Y , <<5 %K#+ DJ#qAFDJ"2r2r2DK"3B3DK"3C5DK"3C5DK"3C5DK"3C5DK"3C5DK"3C5DK%K#+DJ#qAFDJ"2r2r2DK"3B3DK"3C5DK"3C5DK"3C5DK"3C5DK"3C5DK"3C5DKs)=>  aY[IK  +1-rcBURU5n[R"U5nURU5nURS:XaUR U5nUR U5nURU5nURU5nURU5nURU5nURU5nURU5nURU5nURU5nOUR U5nUR U5nURU5nURU5nURU5nURU5nURU5nURU5nURU5nURU5nURS:a"URU5nURU5nUR (a>[R"U5nUR#U5n[$R&"USS/S9nU$)Nr\rr`rDrL)r,r4r5rmrYrnrorprqrrrsrtrurrvrwrrxrNsqueeze)r-rPr6s rr7SqueezeNet.forwards JJv  FF1I JJqM <<5  AA AA AA 1 A AA AA AA AA 1 A AA AA AA 1 A AA AA 1 A AA AA AA AA   a  1 A AA >>q Aq!Aq1v.Ar)rxr,rnrorprqrrrsrtrurwrvrmrrYr)iT)rYrXrr9rrZr:r;rT) rrrr__doc__rr+r7rr=r>s@rrVrVdsW@LOHL7.7.),7.@D7. 7.7.r""rrVc [U40UD6nU(a[U[;d US35e[[US[US5n[R"U5nUR U5 U$)NzJ model do not have a pretrained model now, you should set pretrained=FalserrC)rV model_urlsrrNloadset_dict)archrY pretrainedkwargsmodel weight_pathparams r _squeezenetrs| w )& )Ez! f^ _ !0 t Q D!1!!4  K( u Lrc [SSU40UD6$)aSqueezeNet v1.0 model from `"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size" `_. Args: pretrained (bool, optional): Whether to load pre-trained weights. If True, returns a model pre-trained on ImageNet. Default: False. **kwargs (optional): Additional keyword arguments. For details, please refer to :ref:`SqueezeNet `. Returns: :ref:`api_paddle_nn_Layer`. An instance of SqueezeNet v1.0 model. Examples: .. code-block:: pycon >>> import paddle >>> from paddle.vision.models import squeezenet1_0 >>> # build model >>> model = squeezenet1_0() >>> # build model and load imagenet pretrained weight >>> # model = squeezenet1_0(pretrained=True) >>> x = paddle.rand([1, 3, 224, 224]) >>> out = model(x) >>> print(out.shape) paddle.Size([1, 1000]) r!r\rrrs rr!r!B z DV DDrc [SSU40UD6$)aSqueezeNet v1.1 model from `"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size" `_. 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:`SqueezeNet `. Returns: :ref:`api_paddle_nn_Layer`. An instance of SqueezeNet v1.1 model. Examples: .. code-block:: pycon >>> import paddle >>> from paddle.vision.models import squeezenet1_1 >>> # build model >>> model = squeezenet1_1() >>> # build model and load imagenet pretrained weight >>> # model = squeezenet1_1(pretrained=True) >>> x = paddle.rand([1, 3, 224, 224]) >>> out = model(x) >>> print(out.shape) paddle.Size([1, 1000]) r"r]rrs rr"r"rr) rrXrYrXrrZrUnpack[_SqueezeNetOptions]r:rV)F)rrZrrr:rV)# __future__rtypingrrtyping_extensionsrrrNpaddle.nn.functionalr functionalr4paddle.base.param_attrr paddle.nnr r r r paddle.utils.downloadrrpaddle._typingrr__all__rLayerr$r@rVrr!r"rrrrs # 2 ,CC;$%Y%   2880/rxx/,@@F  )   *!E!E(B!E!EJ!E!E(B!E!Er