j UddlmZddlmZddlmZddlmZddlm Z dd gZ gd Z e e ed <d d gZe e ed< ddedede dedede f dZ ddedededede f dZ ddedededede f dZdS) )Any)_CompiledModule)classification) PreProcessor)OrientationPredictorcrop_orientation_predictorpage_orientation_predictor) magc_resnet31mobilenet_v3_smallmobilenet_v3_small_rmobilenet_v3_largemobilenet_v3_large_rresnet18resnet31resnet34resnet50 resnet34_wide textnet_tiny textnet_small textnet_base vgg16_bn_rvit_svit_bvip_tinyvip_baseARCHS#mobilenet_v3_small_crop_orientation#mobilenet_v3_small_page_orientationORIENTATION_ARCHSFarch pretrained model_typedisabledkwargsreturnc |rtddSt|tr8|tvrt d|dt j||}nQt jtg}t|t|st dt||}| d|j d|d<| d|j d|d<| d|dkrd nd |d<|j d d d}tt|fd d d||}|S)Nzunknown architecture '')r#zunknown architecture: meanstd batch_sizecrop input_shaperT)preserve_aspect_ratio symmetric_pad)r isinstancestrr! ValueErrorr__dict__ MobileNetV3rtupletypegetcfgr) r"r#r$r%r&_model allowed_archsr0 predictors b/var/www/html/Carbon-Document/venv/lib/python3.11/site-packages/doctr/models/classification/zoo.py_orientation_predictorr@(si0#D$///$  ( ( (=d===>> > (.*EEE(3_E $m 4 455 DBd4jjBBCC CZZ 6(:;;F6NJJufj&788F5M!::l:;O;OCCUVWWF<*]+ABB/K$[[D[[TZ[[]cI r.r,c $td|||dd|S)aCrop orientation classification architecture. >>> import numpy as np >>> from doctr.models import crop_orientation_predictor >>> model = crop_orientation_predictor(arch='mobilenet_v3_small_crop_orientation', pretrained=True) >>> input_crop = (255 * np.random.rand(256, 256, 3)).astype(np.uint8) >>> out = model([input_crop]) Args: arch: name of the architecture to use (e.g. 'mobilenet_v3_small_crop_orientation') pretrained: If True, returns a model pre-trained on our recognition crops dataset batch_size: number of samples the model processes in parallel **kwargs: keyword arguments to be passed to the OrientationPredictor Returns: OrientationPredictor r-r"r#r,r$r@r"r#r,r&s r?r r G%( " wt zfl w wpv w wwrAr/c $td|||dd|S)aPage orientation classification architecture. >>> import numpy as np >>> from doctr.models import page_orientation_predictor >>> model = page_orientation_predictor(arch='mobilenet_v3_small_page_orientation', pretrained=True) >>> input_page = (255 * np.random.rand(512, 512, 3)).astype(np.uint8) >>> out = model([input_page]) Args: arch: name of the architecture to use (e.g. 'mobilenet_v3_small_page_orientation') pretrained: If True, returns a model pre-trained on our recognition crops dataset batch_size: number of samples the model processes in parallel **kwargs: keyword arguments to be passed to the OrientationPredictor Returns: OrientationPredictor pagerCrDrErFs r?r r ^rGrAN)F)rFr.)r Fr/)typingrdoctr.models.utilsrr preprocessorrr>r __all__rlistr4__annotations__r!boolr@intr r rDrAr?rSs ......''''''++++++ ')E FtCy*!FGlm49mmmDI -0<@TW@dgxx xCGx]`xruxxxxx0dexx xCGx]`xpsxxxxxxxrA