a hA@svddlZddlmZddlmZddlmZmZddlZddl m Z ddl m m Z ddlmmZddlmZddlmZddlmZd d lmZmZmZd d lmZd d lmZmZgd ZGddde j Z!Gddde j"Z#Gddde j$Z%Gddde j Z&e j ee'ddddZ(e)e*e)e)e)e)fe)eee'ee&dddZ+dedddZ,Gd d!d!eZ-Gd"d#d#eZ.Gd$d%d%eZ/Gd&d'd'eZ0eed(e-j1fd)dd*d+ee-e'ee&d,d-d.Z2eed(e.j1fd)dd*d+ee.e'ee&d,d/d0Z3eed(e/j1fd)dd*d+ee/e'ee&d,d1d2Z4eed(e0j1fd)dd*d+ee0e'ee&d,d3d4Z5dS)5N) OrderedDict)partial)AnyOptional)Tensor)ImageClassification)_log_api_usage_once)register_modelWeights WeightsEnum)_IMAGENET_CATEGORIES)_ovewrite_named_paramhandle_legacy_interface) DenseNetDenseNet121_WeightsDenseNet161_WeightsDenseNet169_WeightsDenseNet201_Weights densenet121 densenet161 densenet169 densenet201cseZdZdeeeeeddfdd ZeeedddZ eeed d d Z e j j eeed d d Ze j jeeed ddZe j jeed ddZeed ddZZS) _DenseLayerFN)num_input_features growth_ratebn_size drop_ratememory_efficientreturncstt||_tjdd|_tj|||dddd|_t|||_ tjdd|_ tj|||ddddd|_ t ||_ ||_dS)NTZinplacer F kernel_sizestridebiasr#r$paddingr%)super__init__nn BatchNorm2dnorm1ReLUrelu1Conv2dconv1norm2relu2conv2floatrr)selfrrrrr __class__I/var/www/auris/lib/python3.9/site-packages/torchvision/models/densenet.pyr* s   z_DenseLayer.__init__)inputsr cCs&t|d}||||}|SNr )torchcatr1r/r-)r6r;Zconcated_featuresbottleneck_outputr9r9r: bn_function/s z_DenseLayer.bn_function)inputr cCs|D]}|jrdSqdS)NTF)Z requires_grad)r6rAZtensorr9r9r:any_requires_grad5sz_DenseLayer.any_requires_gradcs$fdd}tj|g|RddiS)Ncs |SN)r@)r;r6r9r:closure=sz7_DenseLayer.call_checkpoint_bottleneck..closureZ use_reentrantF)cp checkpoint)r6rArEr9rDr:call_checkpoint_bottleneck;s z&_DenseLayer.call_checkpoint_bottleneckcCsdSrCr9r6rAr9r9r:forwardBsz_DenseLayer.forwardcCsdSrCr9rIr9r9r:rJFscCst|tr|g}n|}|jrD||rDtjr8td||}n | |}| | | |}|j dkrtj||j |jd}|S)Nz%Memory Efficient not supported in JITr)ptraining) isinstancerrrBr=jitZ is_scripting ExceptionrHr@r4r3r2rFZdropoutrL)r6rAZ prev_featuresr? new_featuresr9r9r:rJLs     )F)__name__ __module__ __qualname__intr5boolr*listrr@rBr=rNZunusedrHZ_overload_methodrJ __classcell__r9r9r7r:rs rc sDeZdZdZd eeeeeeddfdd Zeeddd Z Z S) _DenseBlockrFN) num_layersrrrrrr c sJtt|D]2}t|||||||d}|d|d|qdS)N)rrrrz denselayer%dr )r)r*ranger add_module) r6rZrrrrrilayerr7r9r:r*cs   z_DenseBlock.__init__) init_featuresr cCs6|g}|D]\}}||}||qt|dSr<)itemsappendr=r>)r6r_featuresnamer^rQr9r9r:rJws  z_DenseBlock.forward)F) rRrSrT_versionrUr5rVr*rrJrXr9r9r7r:rY`s rYcs&eZdZeeddfdd ZZS) _TransitionN)rnum_output_featuresr csNtt||_tjdd|_tj||dddd|_tj ddd|_ dS)NTr!r Fr"r)r#r$) r)r*r+r,Znormr.relur0convZ AvgPool2dpool)r6rrfr7r9r:r*s   z_Transition.__init__)rRrSrTrUr*rXr9r9r7r:resrec sReZdZdZdeeeeeefeeeeed d fd d Ze e d ddZ Z S)raKDensenet-BC model class, based on `"Densely Connected Convolutional Networks" `_. Args: growth_rate (int) - how many filters to add each layer (`k` in paper) block_config (list of 4 ints) - how many layers in each pooling block num_init_features (int) - the number of filters to learn in the first convolution layer bn_size (int) - multiplicative factor for number of bottle neck layers (i.e. bn_size * k features in the bottleneck layer) drop_rate (float) - dropout rate after each dense layer num_classes (int) - number of classification classes memory_efficient (bool) - If True, uses checkpointing. Much more memory efficient, but slower. Default: *False*. See `"paper" `_.  @rFN)r block_confignum_init_featuresrr num_classesrr c stt|ttdtjd|dddddfdt|fdtjd d fd tj ddd d fg|_ |}t |D]|\} } t | |||||d} |j d| d | || |}| t|d krrt||dd} |j d| d | |d}qr|j dt|t|||_|D]r} t| tjr||}tj|dd}t|d}t|d}||}|S)NTr!)r r r )rbrPrgZadaptive_avg_pool2dr=flattenrz)r6rrboutr9r9r:rJs     zDenseNet.forward)rjrkrprqrrrF) rRrSrT__doc__rUtupler5rVr*rrJrXr9r9r7r:rs&@Z num_paramsZ_metricsZ_ops _file_sizeurlZ transformsrN rRrSrTr rr _COMMON_META IMAGENET1K_V1DEFAULTr9r9r9r:r s" rc @sBeZdZedeeddiedddddid d d d ZeZd S)rz`_. Args: weights (:class:`~torchvision.models.DenseNet121_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.DenseNet121_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.densenet.DenseNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.DenseNet121_Weights :members: rjrkrp)rverifyrrrrr9r9r:r[s rcKs"t|}tddd||fi|S)a{Densenet-161 model from `Densely Connected Convolutional Networks `_. Args: weights (:class:`~torchvision.models.DenseNet161_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.DenseNet161_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.densenet.DenseNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.DenseNet161_Weights :members: 0)rlrm$rn`)rrrrr9r9r:rus rcKs"t|}tddd||fi|S)a{Densenet-169 model from `Densely Connected Convolutional Networks `_. Args: weights (:class:`~torchvision.models.DenseNet169_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.DenseNet169_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.densenet.DenseNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.DenseNet169_Weights :members: rj)rlrmrjrjrp)rrrrr9r9r:rs rcKs"t|}tddd||fi|S)a{Densenet-201 model from `Densely Connected Convolutional Networks `_. Args: weights (:class:`~torchvision.models.DenseNet201_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.DenseNet201_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.densenet.DenseNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.DenseNet201_Weights :members: rj)rlrmrrjrp)rrrrr9r9r:rs r)6r collectionsr functoolsrtypingrrr=Ztorch.nnr+Ztorch.nn.functionalZ functionalrPZtorch.utils.checkpointutilsrGrFrZtransforms._presetsrr Z_apir r r Z_metar_utilsrr__all__ModulerZ ModuleDictrYrwrerrVrrUrrrrrrrrrrrrr9r9r9r:s`        A U $$$