jb& xUddlZddlmZddlmZddlmZddlmZddl m Z ddl m Z gd Z d d d ee d ddd d d ee d ddd d d ee d ddd d d ee d ddd d dgdddd d dgddddZeeeeeffed< d(dededeedzdeedzdedejf d Zd)dededejfd"Zd)dededejfd#Zd)dededejfd$Zd)dededejfd%Zd)dededejfd&Zd)dededejfd'ZdS)*N)deepcopy)Any) mobilenetv3) MobileNetV3)VOCABSload_pretrained_params)rmobilenet_v3_smallmobilenet_v3_small_rmobilenet_v3_largemobilenet_v3_large_r#mobilenet_v3_small_crop_orientation#mobilenet_v3_small_page_orientation)gh|?5?g= ףp=?gV-?)gA`"?gl?g$C?)r rfrenchzUhttps://doctr-static.mindee.com/models?id=v0.4.1/mobilenet_v3_large-11fc8cb9.pt&src=0)meanstd input_shapeclassesurlzWhttps://doctr-static.mindee.com/models?id=v0.4.1/mobilenet_v3_large_r-74a22066.pt&src=0zUhttps://doctr-static.mindee.com/models?id=v0.4.1/mobilenet_v3_small-6a4bfa6b.pt&src=0zWhttps://doctr-static.mindee.com/models?id=v0.4.1/mobilenet_v3_small_r-1a8a3530.pt&src=0)rr)riZzfhttps://doctr-static.mindee.com/models?id=v0.8.1/mobilenet_v3_small_crop_orientation-f0847a18.pt&src=0)rrzfhttps://doctr-static.mindee.com/models?id=v0.8.1/mobilenet_v3_small_page_orientation-8e60325c.pt&src=0)r rr r rr default_cfgsarch pretrained rect_strides ignore_keyskwargsreturnc |dtt|d|d<|dt|d|d<tt|}|d|d<|d|d<|d|drt jd i|ddi}nt jd i|ddi}t|tr6|D]3}|}| dD]} t|| }d|_ 4dtdtd dfd } t!j| ||_|rV|dtt|dkr|nd} |t|d | ||_|S)N num_classesrr weights.) path_or_urlr!r"c"t||fi|dS)zLoad pretrained parameters onto the model Args: path_or_url: the path or URL to the model parameters (checkpoint) **kwargs: additional arguments to be passed to `doctr.models.utils.load_pretrained_params` Nr )selfr)r!s p/var/www/html/Carbon-Document/venv/lib/python3.11/site-packages/doctr/models/classification/mobilenet/pytorch.pyfrom_pretrainedz&_mobilenet_v3..from_pretrainedhs" t[;;F;;;;;r)r )getlenrrpop startswithrr r isinstancelistsplitgetattrstridestrrtypes MethodTyper-cfg) rrrr r!_cfgmodel layer_namemchildr- _ignore_keyss r, _mobilenet_v3rCKs#JJ}c,t:LY:W6X6XYYF= 9l4.@.KLLF9 L& ' 'D /DY'DO JJy +,,G.FFFFFFF.FFFFFFF,%%&  JA#))#.. & &Au%%AHH<3<#<$<<<<",_eDDES'-]&;s`_. >>> import torch >>> from doctr.models import mobilenet_v3_small >>> model = mobilenetv3_small(pretrained=False) >>> input_tensor = torch.rand((1, 3, 512, 512), dtype=torch.float32) >>> out = model(input_tensor) Args: pretrained: boolean, True if model is pretrained **kwargs: keyword arguments of the MobileNetV3 architecture Returns: a torch.nn.Module r r classifier.3.weightclassifier.3.biasrCrr!s r,r r s6$ j  7LNa6b fl  r.c ,td|gdfdddgi|S)a;MobileNetV3-Small architecture as described in `"Searching for MobileNetV3", `_, with rectangular pooling. >>> import torch >>> from doctr.models import mobilenet_v3_small_r >>> model = mobilenet_v3_small_r(pretrained=False) >>> input_tensor = torch.rand((1, 3, 512, 512), dtype=torch.float32) >>> out = model(input_tensor) Args: pretrained: boolean, True if model is pretrained **kwargs: keyword arguments of the MobileNetV3 architecture Returns: a torch.nn.Module r )zfeatures.2.block.1.0features.4.block.1.0zfeatures.9.block.1.0r rErFrGrHs r,r r sD$ PPP  +,?@      r.c &td|fdddgi|S)aMobileNetV3-Large architecture as described in `"Searching for MobileNetV3", `_. >>> import torch >>> from doctr.models import mobilenet_v3_large >>> model = mobilenet_v3_large(pretrained=False) >>> input_tensor = torch.rand((1, 3, 512, 512), dtype=torch.float32) >>> out = model(input_tensor) Args: pretrained: boolean, True if model is pretrained **kwargs: keyword arguments of the MobileNetV3 architecture Returns: a torch.nn.Module r r rErFrGrHs r,r r s;$   +,?@     r.c ,td|gdfdddgi|S)a;MobileNetV3-Large architecture as described in `"Searching for MobileNetV3", `_, with rectangular pooling. >>> import torch >>> from doctr.models import mobilenet_v3_large_r >>> model = mobilenet_v3_large_r(pretrained=False) >>> input_tensor = torch.rand((1, 3, 512, 512), dtype=torch.float32) >>> out = model(input_tensor) Args: pretrained: boolean, True if model is pretrained **kwargs: keyword arguments of the MobileNetV3 architecture Returns: a torch.nn.Module r)rJzfeatures.7.block.1.0zfeatures.13.block.1.0r rErFrGrHs r,rrsD$ QQQ  +,?@      r.c &td|fdddgi|S)a?MobileNetV3-Small architecture as described in `"Searching for MobileNetV3", `_. >>> import torch >>> from doctr.models import mobilenet_v3_small_crop_orientation >>> model = mobilenet_v3_small_crop_orientation(pretrained=False) >>> input_tensor = torch.rand((1, 3, 512, 512), dtype=torch.float32) >>> out = model(input_tensor) Args: pretrained: boolean, True if model is pretrained **kwargs: keyword arguments of the MobileNetV3 architecture Returns: a torch.nn.Module rr rErFrGrHs r,rr;$ -  +,?@     r.c &td|fdddgi|S)a?MobileNetV3-Small architecture as described in `"Searching for MobileNetV3", `_. >>> import torch >>> from doctr.models import mobilenet_v3_small_page_orientation >>> model = mobilenet_v3_small_page_orientation(pretrained=False) >>> input_tensor = torch.rand((1, 3, 512, 512), dtype=torch.float32) >>> out = model(input_tensor) Args: pretrained: boolean, True if model is pretrained **kwargs: keyword arguments of the MobileNetV3 architecture Returns: a torch.nn.Module rr rErFrGrHs r,rrrNr.)NN)F)r:copyrtypingrtorchvision.modelsrtorchvision.models.mobilenetv3rdoctr.datasetsrutilsr __all__r5rdictr9__annotations__boolrCr r r rrrr/r.r,rZsK ******666666!!!!!!++++++   &$"4x())f &$"4x())h &$"4x())f &$"4x())h &$$$$$w ,,&$$$$$w ,,I++++ d3S#X&'+++b&*$( 22 22s)d"2cT! 2  2  2222j43;CZ.TS[E\643;CZ4TS[E\6DCT_Tk4DCT_Tkr.