hSSKJr SSKJr SSKJrJr SSKJrJ r SSK J r SSK Jr SSKJr S S KJrJrJr S S KJr S S KJrJrJr S S KJrJrJr /SQr"SS\R@5r!"SS\R@5r"S\S\#S\!4Sjr$"SS\5r%\"5\"S\%RL4S\RN4S9SSS\RNS.S\\%S \(S\\#S!\\S"\S\!4 S#jj55r)g)$) OrderedDict)partial)AnyOptional)nnTensor) functional)SemanticSegmentation)_log_api_usage_once)register_modelWeights WeightsEnum)_VOC_CATEGORIES)_ovewrite_value_paramhandle_legacy_interfaceIntermediateLayerGetter)mobilenet_v3_largeMobileNet_V3_Large_Weights MobileNetV3)LRASPP!LRASPP_MobileNet_V3_Large_Weightslraspp_mobilenet_v3_largec ~^\rSrSrSrSS\R S\S\S\S\SS 4 U4S jjjrS \ S\ \ \ 44S jr S r U=r$)ra Implements a Lite R-ASPP Network for semantic segmentation from `"Searching for MobileNetV3" `_. Args: backbone (nn.Module): the network used to compute the features for the model. The backbone should return an OrderedDict[Tensor], with the key being "high" for the high level feature map and "low" for the low level feature map. low_channels (int): the number of channels of the low level features. high_channels (int): the number of channels of the high level features. num_classes (int, optional): number of output classes of the model (including the background). inter_channels (int, optional): the number of channels for intermediate computations. backbone low_channels high_channels num_classesinter_channelsreturnNcf>[TU]5 [U5 Xl[ X#XE5Ulg)N)super__init__r r LRASPPHead classifier)selfrrrr r! __class__s ^/var/www/auris/envauris/lib/python3.13/site-packages/torchvision/models/segmentation/lraspp.pyr%LRASPP.__init__#s+ D! $\+^inputcURU5nURU5n[R"X1RSSSSS9n[ 5nX4S'U$)NbilinearFsizemode align_cornersout)rr'F interpolateshaper)r(r-featuresr5results r*forwardLRASPP.forward+sO=='ooh'mmCkk"#&6ZW\]u  r,)rr'))__name__ __module__ __qualname____firstlineno____doc__rModuleintr%rdictstrr;__static_attributes__ __classcell__r)s@r*rrsv  sv_ _14_EH_WZ_lo_ __VS&[(9r,rc \^\rSrSrS\S\S\S\SS4 U4SjjrS \\\4S\4S jr S r U=r $) r&6rrr r!r"Nc >[TU]5 [R"[R"X$SSS9[R "U5[R "SS95Ul[R"[R"S5[R"X$SSS9[R"55Ul [R"XS5Ul [R"XCS5Ul g)NF)biasT)inplace) r$r%r SequentialConv2d BatchNorm2dReLUcbrAdaptiveAvgPool2dSigmoidscalelow_classifierhigh_classifier)r(rrr r!r)s r*r%LRASPPHead.__init__7s == IImQU C NN> * GGD !  ]]   # IImQU C JJL  !ii 1E!yyaHr,r-cUSnUSnURU5nURU5nXE-n[R"XBRSSSSS9nUR U5UR U5-$)Nlowhighr/r0Fr1)rTrWr6r7r8rXrY)r(r-r\r]xss r*r;LRASPPHead.forwardFstElV} HHTN JJt  E MM!))BC.zQV W""3'$*>*>q*AAAr,)rTrYrXrW) r>r?r@rArDr%rErFrr;rGrHrIs@r*r&r&6sX IS I I3 I`c Ihl I BT#v+. B6 B Br,r&rr r"c ZURnS/[U5VVs/sHup#[USS5(dMUPM snn-[U5S- /-nUSnUSnXRnXRn[ U[ U5S[ U5S0S 9n[XX5$s snnf) Nr_is_cnFrMr\r]) return_layers)r9 enumerategetattrlen out_channelsrrFr) rr ib stage_indiceslow_poshigh_posrrs r*_lraspp_mobilenetv3roRs  HC8)<\)<8UZ@[1)<\\`cdl`mpq`q_rrMBGR H$11L&33M&xG eUXYaUbdj?klH (- EE]s B'B'c N\rSrSr\"S\"\SS9S\SSSS S S .0S S SS.S9r\r Sr g)r`zJhttps://download.pytorch.org/models/lraspp_mobilenet_v3_large-d234d4ea.pthi) resize_sizei"(1)rMrMz]https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_largezCOCO-val2017-VOC-labelsg33333L@gV@)miou pixel_accg㥛 @g{G(@z These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC dataset. ) num_params categoriesmin_sizerecipe_metrics_ops _file_size_docs)url transformsmetaN) r>r?r@rArrr rCOCO_WITH_VOC_LABELS_V1DEFAULTrGrr,r*rr`sS% X/SA!)u) !%,  ,&Gr,r pretrainedpretrained_backbone)weightsweights_backboneNT)rprogressr rrrrkwargsc jURSS5(a [S5e[RU5n[R"U5nUb&Sn[ SU[ URS55nOUcSn[USS 9n[XR5nUbURURUSS 95 U$) a(Constructs a Lite R-ASPP Network model with a MobileNetV3-Large backbone from `Searching for MobileNetV3 `_ paper. .. betastatus:: segmentation module Args: weights (:class:`~torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_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. num_classes (int, optional): number of output classes of the model (including the background). aux_loss (bool, optional): If True, it uses an auxiliary loss. weights_backbone (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The pretrained weights for the backbone. **kwargs: parameters passed to the ``torchvision.models.segmentation.LRASPP`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights :members: aux_lossFz&This model does not use auxiliary lossNr rvT)rdilated)r check_hash) popNotImplementedErrorrverifyrrrhrrroload_state_dictget_state_dict)rrr rrrmodels r*rrzsLzz*e$$!"JKK/66w?G1889IJ+M;GLLYeLfHgh   !*:DIH  6E g44hSW4XY Lr,)* collectionsr functoolsrtypingrrtorchrrtorch.nnr r6transforms._presetsr utilsr _apirrr_metar_utilsrrr mobilenetv3rrr__all__rCrr&rDrorr IMAGENET1K_V1boolrrr,r*rs%# $7(77#\\UU W RYY FBB8 F+ FC FF F& &4 <TT U+-G-U-UV <@!%=W=e=e 3 7 833# 3 9: 3  3 3  3r,