+ j"^RIHt^RIHtHt^RIt^RIHt^RIH t ^RI H t ^RI H t HtHt^RIHt^RIHtHt.ROt!R R ]P,4t!R R ]4t] !4]!R ]P23R7RRRR/RRll44tR#))partial)AnyOptionalN)ImageClassification)_log_api_usage_once)register_modelWeights WeightsEnum)_IMAGENET_CATEGORIES)_ovewrite_named_paramhandle_legacy_interfaceAlexNetAlexNet_WeightscTaa]tRt^toRV3RlV3RllltV3RlRltRtVtV;t#)rc*<V^8dQhRS[RS[RR/#) num_classesdropoutreturnN)intfloat)format __classdict__s"[/home/luca/screen-checking/.venv/lib/python3.14/site-packages/torchvision/models/alexnet.py __annotate__AlexNet.__annotate__s"  C   cn<\SV`4\V4\P!\P !^^@^ ^^R7\P !RR7\P!^^R7\P !^@^^^R7\P !RR7\P!^^R7\P !^R^^R7\P !RR7\P !RR^^R7\P !RR7\P !RR^^R7\P !RR7\P!^^R74 Vn\P!R 4Vn \P!\P!VR7\P!R R 4\P !RR7\P!VR7\P!R R 4\P !RR7\P!R V44Vn R #) ) kernel_sizestridepaddingT)inplace)r r!)r r"i)piN)r&i$)super__init__rnn SequentialConv2dReLU MaxPool2dfeaturesAdaptiveAvgPool2davgpoolDropoutLinear classifier)selfrr __class__s&&&rr(AlexNet.__init__sf D! IIaAq A GGD ! LLQq 1 IIb#1a 8 GGD ! LLQq 1 IIc3Aq 9 GGD ! IIc3Aq 9 GGD ! IIc3Aq 9 GGD ! LLQq 1  ++F3 -- JJ ! IIk4 ( GGD ! JJ ! IIdD ! GGD ! IIdK ( rcN<V^8dQhRS[PRS[P/#)rxr)torchTensor)rrs"rrr/s#%,,rcVPV4pVPV4p\P!V^4pVP V4pV#))r.r0r9flattenr3)r4r8s&&rforwardAlexNet.forward/s@ MM!  LLO MM!Q  OOA r)r0r3r.)ig?) __name__ __module__ __qualname____firstlineno__r(r>__static_attributes____classdictcell__ __classcell__)r5rs@@rrrs  :rcl]tRt^7t]!R]!]^R7RRRRR]RRR R R R R R//RRRRRR/R7t]t Rt R#)rz (  . m VV  E '  M*Gr pretrained)weightsrYprogressTc^V^8dQhR\\,R\R\R\/#)rrYrZkwargsr)rrboolrr)rs"rrrRs/%%1%D%[^%cj%rc \PV4pVe(\VR\VPR,44\ R/VBpVe#VP VPVRR74V#)aAlexNet model architecture from `One weird trick for parallelizing convolutional neural networks `__. .. note:: AlexNet was originally introduced in the `ImageNet Classification with Deep Convolutional Neural Networks `__ paper. Our implementation is based instead on the "One weird trick" paper above. Args: weights (:class:`~torchvision.models.AlexNet_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.AlexNet_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.squeezenet.AlexNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.AlexNet_Weights :members: rrKT)rZ check_hashrT)rverifyr lenrSrload_state_dictget_state_dict)rYrZr\models$$, ralexnetrePsk:$$W-GfmSl9S5TU  f E g44hSW4XY Lr)rrre) functoolsrtypingrrr9torch.nnr)transforms._presetsrutilsr_apirr r _metar _utilsr r __all__ModulerrrVrerTrrrps 5'66'B 4#bii#Lk2,0M0M!NO%D%4%P%r