9id ddlZddlZddlmZddlmZmZmZmZm Z m Z m Z ddl Z ddlZddlmZddlmZmZmZmZmZmZmZmZddlmZmZmZmZmZm Z ertddl!Z"ddl#Z"ejHejHe"jJjLejHdk\re"jNjPZ)n e"jNZ)er erddl*Z*e d e jVd ed ee jVed fZ,Gd d eZ-d Z.dZ/dZ0dZ1de jVde2fdZ3d(de4dee,fdZ5de jVfdZ6 d)de jVde e e4e e4dffde-fdZ7 d)de jVde e e-e8fde4fdZ9d)de jVde-de e4e4ffdZ:dee8e ee ffde2fdZ;dee8e ee ffde2fd Zd)de e8d fd$e e?dd fd%Z@Gd&d'ZAy)*N)BytesIO) TYPE_CHECKINGDictIterableListOptionalTupleUnion)version) ExplicitEnum is_jax_tensor is_tf_tensoris_torch_availableis_torch_tensoris_vision_availablerequires_backendsto_numpy)IMAGENET_DEFAULT_MEANIMAGENET_DEFAULT_STDIMAGENET_STANDARD_MEANIMAGENET_STANDARD_STDOPENAI_CLIP_MEANOPENAI_CLIP_STDz9.1.0zPIL.Image.Imagez torch.TensorceZdZdZdZy)ChannelDimensionchannels_first channels_lastN)__name__ __module__ __qualname__FIRSTLASTb/var/www/html/backtest/airagagent/rag_env/lib/python3.12/site-packages/transformers/image_utils.pyrr@s E Dr%rcbtxr$t|tjjSN)r isinstancePILImageimgs r& is_pil_imager.Es  EZSYY__%EEr%ctxr$t|tjjxsAt|tj xs%t |xst|xs t|Sr() rr)r*r+npndarrayrrrr,s r&is_valid_imager2Is_   C:c399??#C  c2:: &  3        r%crt|ttfr|D]}t|ryyt |syy)NFT)r)listtuple valid_imagesr2)imgsr-s r&r6r6Ss?$u & C$  D ! r%cLt|ttfrt|dSy)NrF)r)r4r5r2r,s r& is_batchedr9_s"#e}%c!f%% r%imagereturnc|jtjk(rytj|dk\xrtj|dkS)zV Checks to see whether the pixel values have already been rescaled to [0, 1]. Frr )dtyper0uint8minmax)r:s r&is_scaled_imagerAes> {{bhh 66%=A  4"&&-1"44r%expected_ndimsc Zt|r|St|tjjr|gSt |rU|j |dzk(r t |}|S|j |k(r|g}|Std|dzd|d|j dtdt|d)a Ensure that the input is a list of images. If the input is a single image, it is converted to a list of length 1. If the input is a batch of images, it is converted to a list of images. Args: images (`ImageInput`): Image of images to turn into a list of images. expected_ndims (`int`, *optional*, defaults to 3): Expected number of dimensions for a single input image. If the input image has a different number of dimensions, an error is raised. r z%Invalid image shape. Expected either z or z dimensions, but got z dimensions.ztInvalid image type. Expected either PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray, but got .) r9r)r*r+r2ndimr4 ValueErrortype)imagesrBs r&make_list_of_imagesrIps& &#))//*xf ;;.1, ,&\F [[N *XF  78J7K4P^O_`KK= .   $V ~Q 0 r%ct|stdt|tr9t |t j j rtj|St|S)NzInvalid image type: ) r2rFrGrr)r*r+r0arrayrr,s r&to_numpy_arrayrLsP # /S {;<<C!Axx} C=r% num_channels.cV||nd}t|tr|fn|}|jdk(rd\}}n-|jdk(rd\}}ntd|j|j||vrt j S|j||vrt jStd)a[ Infers the channel dimension format of `image`. Args: image (`np.ndarray`): The image to infer the channel dimension of. num_channels (`int` or `Tuple[int, ...]`, *optional*, defaults to `(1, 3)`): The number of channels of the image. Returns: The channel dimension of the image. r rP)rz(Unsupported number of image dimensions: z(Unable to infer channel dimension format)r)intrErFshaperr"r#)r:rM first_dimlast_dims r&infer_channel_dimension_formatrWs$0#;<L&0s&CL?L zzQ" 8 q" 8CEJJ<PQQ {{9-%%% X , .$$$ ? @@r%input_data_formatc| t|}|tjk(r|jdz S|tjk(r|jdz St d|)a Returns the channel dimension axis of the image. Args: image (`np.ndarray`): The image to get the channel dimension axis of. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the image. If `None`, will infer the channel dimension from the image. Returns: The channel dimension axis of the image. rPr Unsupported data format: )rWrr"rEr#rF)r:rXs r&get_channel_dimension_axisr[sd :5A,222zzA~ .33 3zzA~ 01B0CD EEr% channel_dimc| t|}|tjk(r|jd|jdfS|tjk(r|jd|jdfSt d|)a Returns the (height, width) dimensions of the image. Args: image (`np.ndarray`): The image to get the dimensions of. channel_dim (`ChannelDimension`, *optional*): Which dimension the channel dimension is in. If `None`, will infer the channel dimension from the image. Returns: A tuple of the image's height and width. rZ)rWrr"rTr#rF)r:r\s r&get_image_sizeras{4U; &,,,{{2 B// (-- -{{2 B//4[MBCCr% annotationct|trId|vrEd|vrAt|dttfr(t |ddk(st|ddtryy)Nimage_id annotationsrTFr)dictr4r5lenrbs r&"is_valid_annotation_coco_detectionrjs_:t$ * $ Z ' z-04- @  =) *a /*]3A6= r%ct|trMd|vrId|vrEd|vrAt|dttfr(t |ddk(st|ddtryy)Nrd segments_info file_namerTFrfris r&!is_valid_annotation_coco_panopticrnsg:t$ * $ z ) : % z/2T5M B  ?+ , 1*_5a8$? r%rec&td|DS)Nc32K|]}t|ywr()rj.0anns r& z3valid_coco_detection_annotations..sN31#6Nallres r& valid_coco_detection_annotationsrys N+N NNr%c&td|DS)Nc32K|]}t|ywr()rnrqs r&rtz2valid_coco_panoptic_annotations..sM#05Mrurvrxs r&valid_coco_panoptic_annotationsr|s MM MMr%timeoutc`ttdgt|tr|j ds|j dr@t j jtj|d|j}ntjj|r t j j|}n|j dr|jdd} tj |d }t j jt#|}n2t|t j j r|}n t'd t j(j+|}|j-d}|S#t$$r}t'd |d |d }~wwxYw)a3 Loads `image` to a PIL Image. Args: image (`str` or `PIL.Image.Image`): The image to convert to the PIL Image format. timeout (`float`, *optional*): The timeout value in seconds for the URL request. Returns: `PIL.Image.Image`: A PIL Image. visionzhttp://zhttps://T)streamr}z data:image/,r )validatezIncorrect image source. Must be a valid URL starting with `http://` or `https://`, a valid path to an image file, or a base64 encoded string. Got z. Failed with NzuIncorrect format used for image. Should be an url linking to an image, a base64 string, a local path, or a PIL image.RGB)r load_imager)str startswithr*r+openrequestsgetrawospathisfilesplitbase64 b64decoder ExceptionrFImageOpsexif_transposeconvert)r:r}b64es r&rrsoj8*-%   I &%*:*::*FIINN8<<dG#T#X#XYE WW^^E "IINN5)E . C(+ &&ut< ws|4 E399?? + D   LL ' ' .E MM% E L  ijoipp~@~AB s*?F F-F((F-ceZdZdZdZddZdZdejde e e fdejfd Z dd Z d Zdd Zdd ZdZdZddZy)ImageFeatureExtractionMixinzD Mixin that contain utilities for preparing image features. ct|tjjtjfs$t |st dt|dyy)Nz Got type zS which is not supported, only `PIL.Image.Image`, `np.array` and `torch.Tensor` are.)r)r*r+r0r1rrFrGselfr:s r&_ensure_format_supportedz4ImageFeatureExtractionMixin._ensure_format_supportedKsQ%#))//2::!>?X]H^DK=)&& I_?r%Nc|j|t|r|j}t|tj r|'t|j dtj}|jdk(r$|jddvr|jddd}|r|dz}|jtj}tjj|S|S)a" Converts `image` to a PIL Image. Optionally rescales it and puts the channel dimension back as the last axis if needed. Args: image (`PIL.Image.Image` or `numpy.ndarray` or `torch.Tensor`): The image to convert to the PIL Image format. rescale (`bool`, *optional*): Whether or not to apply the scaling factor (to make pixel values integers between 0 and 255). Will default to `True` if the image type is a floating type, `False` otherwise. rrPrOr rQ)rrnumpyr)r0r1flatfloatingrErT transposeastyper>r*r+ fromarray)rr:rescales r& to_pil_imagez(ImageFeatureExtractionMixin.to_pil_imageRs %%e, 5 !KKME eRZZ ($UZZ]BKK@zzQ5;;q>V#;1a0 LL*E99&&u- - r%c|j|t|tjjs|S|j dS)z Converts `PIL.Image.Image` to RGB format. Args: image (`PIL.Image.Image`): The image to convert. r)rr)r*r+rrs r& convert_rgbz'ImageFeatureExtractionMixin.convert_rgbps8 %%e,%1L}}U##r%r:scaler;c.|j|||zS)z7 Rescale a numpy image by scale amount )r)rr:rs r&rz#ImageFeatureExtractionMixin.rescale~s %%e,u}r%c|j|t|tjjrt j |}t |r|j}|'t|jdtjn|}|r/|j|jtjd}|r"|jdk(r|jddd}|S)a Converts `image` to a numpy array. Optionally rescales it and puts the channel dimension as the first dimension. Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor`): The image to convert to a NumPy array. rescale (`bool`, *optional*): Whether or not to apply the scaling factor (to make pixel values floats between 0. and 1.). Will default to `True` if the image is a PIL Image or an array/tensor of integers, `False` otherwise. channel_first (`bool`, *optional*, defaults to `True`): Whether or not to permute the dimensions of the image to put the channel dimension first. rp?rPrQr )rr)r*r+r0rKrrrintegerrrfloat32rEr)rr:r channel_firsts r&rLz*ImageFeatureExtractionMixin.to_numpy_arrays %%e, eSYY__ -HHUOE 5 !KKME;B?*UZZ]BJJ7PW LLbjj!99EE UZZ1_OOAq!,E r%c|j|t|tjjr|St |r|j d}|St j|d}|S)z Expands 2-dimensional `image` to 3 dimensions. Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor`): The image to expand. r)axis)rr)r*r+r unsqueezer0 expand_dimsrs r&rz'ImageFeatureExtractionMixin.expand_dimss_ %%e, eSYY__ -L 5 !OOA&E NN5q1E r%c|j|t|tjjr|j |d}nw|rut|t j r0|j|jt jd}n+t|r |j|jd}t|t j rt|t j s.t j|j|j}t|t j st j|j|j}n_t|rTddl}t||js|j |}t||js|j |}|j"dk(r)|j$ddvr||ddddfz |ddddfz S||z |z S)a  Normalizes `image` with `mean` and `std`. Note that this will trigger a conversion of `image` to a NumPy array if it's a PIL Image. Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor`): The image to normalize. mean (`List[float]` or `np.ndarray` or `torch.Tensor`): The mean (per channel) to use for normalization. std (`List[float]` or `np.ndarray` or `torch.Tensor`): The standard deviation (per channel) to use for normalization. rescale (`bool`, *optional*, defaults to `False`): Whether or not to rescale the image to be between 0 and 1. If a PIL image is provided, scaling will happen automatically. T)rrrNrPrO)rr)r*r+rLr0r1rrrrfloatrKr=torchTensortensorrErT)rr:meanstdrrs r& normalizez%ImageFeatureExtractionMixin.normalizes %%e, eSYY__ -''t' eRZZ (dBJJ/xx~,,U[[9c2::.hhsm**5;;7 U # dELL1#u||D)c5<<0"ell3' ::?u{{1~7DD$//3q$}3EE EDLC' 'r%c||ntj}|j|t|tj j s|j |}t|tr t|}t|tst|dk(r|rt|tr||fn |d|df}n|j\}}||kr||fn||f\}} t|tr|n|d} || k(r|S| t| | z|z } } |.|| krtd|d|| |kDrt|| z| z |} } ||kr| | fn| | f}|j||S)a Resizes `image`. Enforces conversion of input to PIL.Image. Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor`): The image to resize. size (`int` or `Tuple[int, int]`): The size to use for resizing the image. If `size` is a sequence like (h, w), output size will be matched to this. If `size` is an int and `default_to_square` is `True`, then image will be resized to (size, size). If `size` is an int and `default_to_square` is `False`, then smaller edge of the image will be matched to this number. i.e, if height > width, then image will be rescaled to (size * height / width, size). resample (`int`, *optional*, defaults to `PILImageResampling.BILINEAR`): The filter to user for resampling. default_to_square (`bool`, *optional*, defaults to `True`): How to convert `size` when it is a single int. If set to `True`, the `size` will be converted to a square (`size`,`size`). If set to `False`, will replicate [`torchvision.transforms.Resize`](https://pytorch.org/vision/stable/transforms.html#torchvision.transforms.Resize) with support for resizing only the smallest edge and providing an optional `max_size`. max_size (`int`, *optional*, defaults to `None`): The maximum allowed for the longer edge of the resized image: if the longer edge of the image is greater than `max_size` after being resized according to `size`, then the image is resized again so that the longer edge is equal to `max_size`. As a result, `size` might be overruled, i.e the smaller edge may be shorter than `size`. Only used if `default_to_square` is `False`. Returns: image: A resized `PIL.Image.Image`. r rz max_size = zN must be strictly greater than the requested size for the smaller edge size = )resample)PILImageResamplingBILINEARrr)r*r+rr4r5rSrhsizerFresize) rr:rrdefault_to_squaremax_sizewidthheightshortlongrequested_new_short new_shortnew_longs r&rz"ImageFeatureExtractionMixin.resizesy< (389K9T9T %%e,%1%%e,E dD !;D dC CIN '1$'<d|47DQRGBT % v16&ufovuo t.8s.Cda#// L&93?RUY?Y\a?a;b8 '#66()(4@@DvG (*.1(Y2F2Q.RT\8 05 8,hPYEZ||D8|44r%c|j|t|ts||f}t|st|tj rP|j dk(r|j|}|jddvr|jddn|jdd}n|jd|jdf}|d|dz dz}||dz}|d|dz dz}||dz}t|tjjr|j||||fS|jddvrdnd}|sKt|tj r|jddd}t|r|jddd}|dk\r!||dkr|dk\r||dkr |d||||fS|jdd t|d|dt|d|dfz} t|tj rt j || } nt|r|j#| } | d |dz dz} | |dz} | d |dz dz} | |dz}| d| | | |f<|| z }|| z }|| z }|| z }| dtd|t%| jd |td|t%| jd |f} | S) a Crops `image` to the given size using a center crop. Note that if the image is too small to be cropped to the size given, it will be padded (so the returned result has the size asked). Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor` of shape (n_channels, height, width) or (height, width, n_channels)): The image to resize. size (`int` or `Tuple[int, int]`): The size to which crop the image. Returns: new_image: A center cropped `PIL.Image.Image` or `np.ndarray` or `torch.Tensor` of shape: (n_channels, height, width). rQrrOr NTF.r^)rTr_)rr)r5rr0r1rErrTrr*r+croprpermuter@ zeros_like new_zerosr?)rr:r image_shapetopbottomleftrightr new_shape new_imagetop_pad bottom_padleft_pad right_pads r& center_cropz'ImageFeatureExtractionMixin.center_crop*s %%e,$&$1a7{1~- bMKN2q8{1~- AF #wz)8I+==> w'   Qs9??2#6??QPST]TcTcdfTginPoAo o r%c|j|t|tjjr|j |}|dddddddfS)a Flips the channel order of `image` from RGB to BGR, or vice versa. Note that this will trigger a conversion of `image` to a NumPy array if it's a PIL Image. Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor`): The image whose color channels to flip. If `np.ndarray` or `torch.Tensor`, the channel dimension should be first. Nr_)rr)r*r+rLrs r&flip_channel_orderz.ImageFeatureExtractionMixin.flip_channel_orderusI %%e, eSYY__ -''.ETrT1aZ  r%c||ntjj}|j|t |tjjs|j |}|j ||||||S)a Returns a rotated copy of `image`. This method returns a copy of `image`, rotated the given number of degrees counter clockwise around its centre. Args: image (`PIL.Image.Image` or `np.ndarray` or `torch.Tensor`): The image to rotate. If `np.ndarray` or `torch.Tensor`, will be converted to `PIL.Image.Image` before rotating. Returns: image: A rotated `PIL.Image.Image`. )rexpandcenter translate fillcolor)r*r+NEARESTrr)rrotate)rr:anglerrrrrs r&rz"ImageFeatureExtractionMixin.rotatesn (389J9J %%e,%1%%e,E|| HVFicl  r%r()NT)F)NTN)NrNNN)rr r!__doc__rrrr0r1r rrSrrLrrrrrrr$r%r&rrFsj< $RZZeSj0Abjj@(,(\A5FIV!" r%r)rPr()Brriortypingrrrrrr r rr0r packagingr utilsr rrrrrrrutils.constantsrrrrrr PIL.Imager* PIL.ImageOpsparse __version__ base_versionr+ Resamplingrrr1 ImageInputrr.r2r6r9boolrArSrIrLrWrr[rarjrnryr|rrrr$r%r&rs  NNN   w}}]W]]3??3@@A]W]]SZE[[ YY11 YYrzz>48I3JDQSQ[Q[L\^bcq^rr | F  52::5$5$$D