EinBDSSKJr SSKJrJrJr SSKrSSKJr SSKJ r SSK J r SSK J r SS KJrJrJr SS KJr SS KJrJr SS K JrJrJrJr S SKJrJrJr /SQr"SS\ R@5r!"SS\ RD5r#S\$\%S\$\%S\\S\&S\&S\S\#4Sjr'S\SSSS .r("S!S"\5r)"S#S$\5r*"S%S&\5r+"S'S(\5r,\"S)S*9\"S+S,4S-9SS.S/S0.S\\\)\4S\&S\&S\S\#4 S1jj55r-\"S2S*9\"S+S34S-9SS.S/S0.S\\\*\4S\&S\&S\S\#4 S4jj55r.\"S5S*9\"S+S64S-9SS.S/S0.S\\\+\4S\&S\&S\S\#4 S7jj55r/\"S8S*9\"S+S94S-9SS.S/S0.S\\\,\4S\&S\&S\S\#4 S:jj55r0g);)partial)AnyOptionalUnionN)Tensor) shufflenetv2)ImageClassification)register_modelWeights WeightsEnum)_IMAGENET_CATEGORIES)_ovewrite_named_paramhandle_legacy_interface)ShuffleNet_V2_X0_5_WeightsShuffleNet_V2_X1_0_WeightsShuffleNet_V2_X1_5_WeightsShuffleNet_V2_X2_0_Weights) _fuse_modules _replace_reluquantize_model) QuantizableShuffleNetV2#ShuffleNet_V2_X0_5_QuantizedWeights#ShuffleNet_V2_X1_0_QuantizedWeights#ShuffleNet_V2_X1_5_QuantizedWeights#ShuffleNet_V2_X2_0_QuantizedWeightsshufflenet_v2_x0_5shufflenet_v2_x1_0shufflenet_v2_x1_5shufflenet_v2_x2_0cJ^\rSrSrS\S\SS4U4SjjrS\S\4SjrS rU=r $) QuantizableInvertedResidual#argskwargsreturnNcl>[TU]"U0UD6 [RR 5UlgN)super__init__nn quantizedFloatFunctionalcatselfr&r' __class__s i/home/mande/repo/quber/.venv/lib/python3.13/site-packages/torchvision/models/quantization/shufflenetv2.pyr,$QuantizableInvertedResidual.__init__$s) $)&)<<//1xcBURS:Xa=URSSS9up#URRX RU5/SS9nO:URRUR U5URU5/SS9n[ R "US5nU$)Nrr )dim)stridechunkr0branch2branch1rchannel_shuffle)r2r7x1x2outs r4forward#QuantizableInvertedResidual.forward(s ;;! WWQAW&FB((,,LL$451,=C((,, QaAq,IC**32 r6)r0) __name__ __module__ __qualname____firstlineno__rr,rrB__static_attributes__ __classcell__r3s@r4r$r$#s52c2S2T2  F  r6r$cf^\rSrSrS\S\SS4U4SjjrS\S\4SjrS S \\ SS4S jjr S r U=r $) r4r&r'r(Nc>[TU]"US[0UD6 [RR R 5Ul[RR R5Ul g)Ninverted_residual) r+r,r$torchao quantization QuantStubquant DeQuantStubdequantr1s r4r, QuantizableShuffleNetV2.__init__6sP $X2MXQWXXX**446 xx,,88: r6r7clURU5nURU5nURU5nU$r*)rS _forward_implrU)r2r7s r4rBQuantizableShuffleNetV2.forward;s1 JJqM   q ! LLOr6is_qatcURR5H!up#US;dM UcM[U/SQ/USS9 M# UR5Hn[ U5[ LdM[ URRR55S:a[URSS//S Q/USS9 [UR/SQS S //S Q/USS9 M g) aFuse conv/bn/relu modules in shufflenetv2 model Fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization. Model is modified in place. .. note:: Note that this operation does not change numerics and the model after modification is in floating point )conv1conv5N)012T)inplacerr^r_)r`34rbrc)567) _modulesitemsrmodulestyper$lenr=r<)r2rZnamems r4 fuse_model"QuantizableShuffleNetV2.fuse_modelAs}}**,GD))ama/!2FDI-AAw55qyy))//12Q6!!))sCj/-JF\`aII$sCj/B   r6)rUrSr*) rDrErFrGrr,rrBrboolrnrHrIrJs@r4rr4sL;c;S;T; F $4r6rstages_repeatsstages_out_channelsweightsprogressquantizer'r(c `UbM[US[URS55 SUR;a[USURS5 URSS5n[ X40UD6n[ U5 U(a [ Xv5 UbURURUSS95 U$)N num_classes categoriesbackendfbgemmT)rt check_hash) rrkmetapoprrrload_state_dictget_state_dict)rqrrrsrtrur'rymodels r4 _shufflenetv2rZsfmSl9S5TU  $ !&)W\\)5L MjjH-G #N R6 RE%u& g44hSW4XY Lr6)rrrzzdhttps://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-modelsz These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized weights listed below. )min_sizerxryrecipe_docsc d\rSrSr\"S\"\SS90\ES\RSSSS .0S S S .ES 9r \ r Sr g)rzShttps://download.pytorch.org/models/quantized/shufflenetv2_x0.5_fbgemm-00845098.pth crop_sizei ImageNet-1Kg#~jL@gRS@zacc@1zacc@5g{Gz?gjt? num_params unquantized_metrics_ops _file_sizeurl transformsr|N) rDrErFrGr rr _COMMON_METAr IMAGENET1K_V1IMAGENET1K_FBGEMM_V1DEFAULTrHrr6r4rrs[" a.#>    !5CC##    "#Gr6rc d\rSrSr\"S\"\SS90\ES\RSSSS .0S S S .ES 9r \ r Sr g)rzQhttps://download.pytorch.org/models/quantized/shufflenetv2_x1_fbgemm-1e62bb32.pthrri"rgףp= Q@gh|?U@rg(\?gy&1@rrrN) rDrErFrGr rr rrrrrrHrr6r4rrs[" _.#>    !5CC##    "#Gr6rc h\rSrSr\"S\"\SSS90\ESS\RSS S S .0S S S.ES9r \ r Sr g)rzShttps://download.pytorch.org/models/quantized/shufflenetv2_x1_5_fbgemm-d7401f05.pthrr resize_size+https://github.com/pytorch/vision/pull/5906iv5rgSR@g̬V@rgl?gK7A` @rrrrrrrrN) rDrErFrGr rr rrrrrrHrr6r4rr`" a.#3O    C!5CC##    $#Gr6rc h\rSrSr\"S\"\SSS90\ESS\RSS S S .0S S S.ES9r \ r Sr g)rzShttps://download.pytorch.org/models/quantized/shufflenetv2_x2_0_fbgemm-5cac526c.pthrrrriprg-R@gZd;W@rg-?g|?5@rrrN) rDrErFrGr rr rrrrrrHrr6r4rrrr6rquantized_shufflenet_v2_x0_5)rl pretrainedcpURSS5(a[R$[R$NruF)getrrrrr's r4r1zz*e,, 0 D D :+99 :r6)rsTFrsrtruc nU(a[O[RU5n[/SQ/SQ4XUS.UD6$)a Constructs a ShuffleNetV2 with 0.5x output channels, as described in `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design `__. .. note:: Note that ``quantize = True`` returns a quantized model with 8 bit weights. Quantized models only support inference and run on CPUs. GPU inference is not yet supported. Args: weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X0_5_Weights`, optional): The pretrained weights for the model. See :class:`~torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights` 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. quantize (bool, optional): If True, return a quantized version of the model. Default is False. **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights :members: .. autoclass:: torchvision.models.ShuffleNet_V2_X0_5_Weights :members: :noindex: r)0`rr)rrverifyrrsrtrur's r4rrsBd7?2D^ffgnoG * 4;Ya ek r6quantized_shufflenet_v2_x1_0cpURSS5(a[R$[R$r)rrrrrrs r4rrrr6c nU(a[O[RU5n[/SQ/SQ4XUS.UD6$)a Constructs a ShuffleNetV2 with 1.0x output channels, as described in `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design `__. .. note:: Note that ``quantize = True`` returns a quantized model with 8 bit weights. Quantized models only support inference and run on CPUs. GPU inference is not yet supported. Args: weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X1_0_Weights`, optional): The pretrained weights for the model. See :class:`~torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. quantize (bool, optional): If True, return a quantized version of the model. Default is False. **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights :members: .. autoclass:: torchvision.models.ShuffleNet_V2_X1_0_Weights :members: :noindex: r)rtrirr)rrrrrs r4r r Bd7?2D^ffgnoG , 6=[c gm r6quantized_shufflenet_v2_x1_5cpURSS5(a[R$[R$r)rrrrrrs r4rrJrr6c nU(a[O[RU5n[/SQ/SQ4XUS.UD6$)a Constructs a ShuffleNetV2 with 1.5x output channels, as described in `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design `__. .. note:: Note that ``quantize = True`` returns a quantized model with 8 bit weights. Quantized models only support inference and run on CPUs. GPU inference is not yet supported. Args: weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X1_5_Weights`, optional): The pretrained weights for the model. See :class:`~torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights` 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. quantize (bool, optional): If True, return a quantized version of the model. Default is False. **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights :members: .. autoclass:: torchvision.models.ShuffleNet_V2_X1_5_Weights :members: :noindex: r)ri`irr)rrrrrs r4r!r!Frr6quantized_shufflenet_v2_x2_0cpURSS5(a[R$[R$r)rrrrrrs r4rrrr6c nU(a[O[RU5n[/SQ/SQ4XUS.UD6$)a Constructs a ShuffleNetV2 with 2.0x output channels, as described in `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design `__. .. note:: Note that ``quantize = True`` returns a quantized model with 8 bit weights. Quantized models only support inference and run on CPUs. GPU inference is not yet supported. Args: weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X2_0_Weights`, optional): The pretrained weights for the model. See :class:`~torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. quantize (bool, optional): If True, return a quantized version of the model. Default is False. **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights :members: .. autoclass:: torchvision.models.ShuffleNet_V2_X2_0_Weights :members: :noindex: r)riiir)rrrrrs r4r"r"~rr6)1 functoolsrtypingrrrrOtorch.nnr-rtorchvision.modelsrtransforms._presetsr _apir r r_metar_utilsrrrrrrutilsrrr__all__InvertedResidualr$ ShuffleNetV2rlistintrprrrrrrrr r!r"rr6r4rs'' +677(C @? ,"?"?"#l77#LIck "     6&t  #+#*#+#*#+#,#+#,34   ae * e?A[[\ ]** *  *  * 5*Z34   ae * e?A[[\ ]** *  *  * 5*Z34   ae * e?A[[\ ]** *  *  * 5*Z34   ae * e?A[[\ ]** *  *  * 5*r6