o h@sddlZddlZddlmZddlmZddlmZddlm Z m Z m Z m Z ddl mZddlmZGd d d eZGd d d eZGd ddeZdS)N) defaultdict) HTMLParser)Path)AnyCallableOptionalUnion)default_loader) VisionDatasetcs|eZdZdZdeeefddffdd Zdedee ee efddfd d Z deddfd d Z d eddfddZ ZS)Flickr8kParserzBParser for extracting captions from the Flickr8k dataset web page.rootreturnNcs,t||_i|_d|_d|_d|_dS)NF)super__init__r annotationsin_table current_tag current_img)selfr  __class__a/home/www/facesmatcher.com/frenv_anti/lib/python3.10/site-packages/torchvision/datasets/flickr.pyrs  zFlickr8kParser.__init__tagattrscCs||_|dkr d|_dSdS)NtableTrr)rrrrrrhandle_starttag zFlickr8kParser.handle_starttagcCsd|_|dkr d|_dSdS)NrFr)rrrrr handle_endtag"rzFlickr8kParser.handle_endtagdatacCs|jrJ|dkr d|_dS|jdkr3|dd}tj|j|d}t|d}||_g|j |<dS|jdkrL|jrN|j}|j | | dSdSdSdS)NzImage Not Founda/z_*.jpgrZli) rrrsplitospathjoinr globrappendstrip)rr!img_idrrr handle_data(s   zFlickr8kParser.handle_data)__name__ __module__ __qualname____doc__rstrrrlisttuplerrr r- __classcell__rrrrr s & r c seZdZdZddefdeeefdedee dee de ege fddf fd d Z d e de e e ffd d Zde fddZZS)Flickr8ka?`Flickr8k Entities `_ Dataset. Args: root (str or ``pathlib.Path``): Root directory where images are downloaded to. ann_file (string): Path to annotation file. transform (callable, optional): A function/transform that takes in a PIL image or torch.Tensor, depends on the given loader, and returns a transformed version. E.g, ``transforms.RandomCrop`` target_transform (callable, optional): A function/transform that takes in the target and transforms it. loader (callable, optional): A function to load an image given its path. By default, it uses PIL as its image loader, but users could also pass in ``torchvision.io.decode_image`` for decoding image data into tensors directly. Nr ann_file transformtarget_transformloaderrcstj|||dtj||_t|j}t|j}| | Wdn1s,wY|j |_ t t |j |_||_dS)Nr8r9)rrr&r' expanduserr7r r openfeedreadrr3sortedkeysidsr:)rr r7r8r9r:parserfhrrrrFs   zFlickr8k.__init__indexcCsN|j|}||}|jdur||}|j|}|jdur#||}||fSz Args: index (int): Index Returns: tuple: Tuple (image, target). target is a list of captions for the image. N)rBr:r8rr9)rrEr,imgtargetrrr __getitem__Zs       zFlickr8k.__getitem__cC t|jSNlenrBrrrr__len__p zFlickr8k.__len__)r.r/r0r1r rr2rrrrrintr4rIrOr5rrrrr67s(  r6c s|eZdZdZddefdededeedeedeegefddf fd d Z d e de eeffd d Z de fddZ ZS) Flickr30ka8`Flickr30k Entities `_ Dataset. Args: root (str or ``pathlib.Path``): Root directory where images are downloaded to. ann_file (string): Path to annotation file. transform (callable, optional): A function/transform that takes in a PIL image or torch.Tensor, depends on the given loader, and returns a transformed version. E.g, ``transforms.RandomCrop`` target_transform (callable, optional): A function/transform that takes in the target and transforms it. loader (callable, optional): A function to load an image given its path. By default, it uses PIL as its image loader, but users could also pass in ``torchvision.io.decode_image`` for decoding image data into tensors directly. Nr r7r8r9r:rc stj|||dtj||_tt|_t |j"}|D]}| d\}} |j|dd | qWdn1s?wYtt |j|_||_dS)Nr; r$)rrr&r'r<r7rr3rr=r+r%r*r@rArBr:) rr r7r8r9r:rDliner,captionrrrrs   zFlickr30k.__init__rEcCs^|j|}tj|j|}||}|jdur||}|j|}|jdur+||}||fSrF) rBr&r'r(r r:r8rr9)rrEr,filenamerGrHrrrrIs       zFlickr30k.__getitem__cCrJrKrLrNrrrrOrPzFlickr30k.__len__)r.r/r0r1r r2rrrrrQr4rIrOr5rrrrrRts( rR)r)r& collectionsr html.parserrpathlibrtypingrrrrfolderr Zvisionr r r6rRrrrrs     +=