U ʹj)F@sdZddlZddlmZmZmZddlZddlmZddl m Z m Z m Z ddl mZmZmZddlmZmZmZmZmZmZmZmZddlmZd d d e e feeeeeffeeeed feed fd ddZ d dddddddd dddde e ddddd d feeeeeffeeeefeeeefeeeeeeeed ffeeeeeeeeeed feed feeeeeedddZ!d dddd e e d feeeeeffeeeeeeeeed feed fedddZ"d d d ddddddddddd e e dddddddd d d feeeeefeeeeffeeeeeeeefeeeefeeeeeed ffeeeeeeeeed feed feeeeeeeeeeeeedddZ#dS)z Transforms Factory Factory methods for building image transforms for use with TIMM (PyTorch Image Models) Hacked together by / Copyright 2019, Ross Wightman N)OptionalTupleUnion) transforms)IMAGENET_DEFAULT_MEANIMAGENET_DEFAULT_STDDEFAULT_CROP_PCT)rand_augment_transformaugment_and_mix_transformauto_augment_transform)str_to_interp_modestr_to_pil_interp!RandomResizedCropAndInterpolationResizeKeepRatioCenterCropOrPadRandomCropOrPad TrimBorderToNumpy) RandomErasingbilinearF.)img_size interpolationuse_prefetchermeanstdcCsj|dkr d}tj|t|dt|g}|r:|tg7}n&|ttjt|t|dg7}t |S)aZ No-augmentation image transforms for training. Args: img_size: Target image size. interpolation: Image interpolation mode. mean: Image normalization mean. std: Image normalization standard deviation. use_prefetcher: Prefetcher enabled. Do not convert image to tensor or normalize. Returns: randomrrrr) rResizer CenterCroprToTensor NormalizetorchtensorCompose)rrrrrtflr'[/var/www/html/bwave/python/venv/lib/python3.8/site-packages/timm/data/transforms_factory.pytransforms_noaug_trainsr)?g?rconst)rscaleratiotrain_crop_modehflipvflip color_jittercolor_jitter_probforce_color_jittergrayscale_probgaussian_blur_prob auto_augmentrrrre_probre_modere_count re_num_splitsrseparatec Cs|pd}|dkst|dkrlt|p$d}t|p0d}t|| d|dd|d|d kr\t|d d n t|d d g}n*t|ptd }t|pd }t|||| dg}|dkr|tj|dg7}|dkr|tj|dg7}g}d}| rt | t st|pd| k }t |tt frt |}n|}t t|dtdd| Dd}| rP| dkrPt| |d<| drn|t| |g7}n6| drd|d<|t| |g7}n|t| |g7}|dk r|st |t tfrt|dkstnt|fd}|dk r|tjtj|g|dg7}n|tj|g7}| r6|tj| dg7}| rZ|tjtjd d!g| dg7}g}|rr|tg7}nH|ttjt| t|d"g7}|dkr|t||||d#d$g7}|rt |t |t |fSt |||SdS)%a ImageNet-oriented image transforms for training. Args: img_size: Target image size. train_crop_mode: Training random crop mode ('rrc', 'rkrc', 'rkrr'). scale: Random resize scale range (crop area, < 1.0 => zoom in). ratio: Random aspect ratio range (crop ratio for RRC, ratio adjustment factor for RKR). hflip: Horizontal flip probability. vflip: Vertical flip probability. color_jitter: Random color jitter component factors (brightness, contrast, saturation, hue). Scalar is applied as (scalar,) * 3 (no hue). color_jitter_prob: Apply color jitter with this probability if not None (for SimlCLR-like aug). force_color_jitter: Force color jitter where it is normally disabled (ie with RandAugment on). grayscale_prob: Probability of converting image to grayscale (for SimCLR-like aug). gaussian_blur_prob: Probability of applying gaussian blur (for SimCLR-like aug). auto_augment: Auto augment configuration string (see auto_augment.py). interpolation: Image interpolation mode. mean: Image normalization mean. std: Image normalization standard deviation. re_prob: Random erasing probability. re_mode: Random erasing fill mode. re_count: Number of random erasing regions. re_num_splits: Control split of random erasing across batch size. use_prefetcher: Prefetcher enabled. Do not convert image to tensor or normalize. separate: Output transforms in 3-stage tuple. Returns: If separate==True, the transforms are returned as a tuple of 3 separate transforms for use in a mixing dataset that passes * all data through the first (primary) transform, called the 'clean' data * a portion of the data through the secondary transform * normalizes and converts the branches above with the third, final transform rrc>rkrrr>rkrc)r@r?)g??)g?grq?r*T)rZrandom_scale_probZrandom_scale_rangeZrandom_scale_areaZrandom_aspect_probZrandom_aspect_ranger@Zreflect)Z padding_mode)g{Gz?rA)g?gUUUUUU?)r.r/rr+)pFZ3ag?cSsg|]}tdtd|qS)minround.0xr'r'r( sz-transforms_imagenet_train..)Ztranslate_constZimg_meanrrZrandZaugmixg333333?Z translate_pctN)rK)Z kernel_sizercpu)mode max_countZ num_splitsZdevice)!AssertionErrortuplerrrrrZRandomHorizontalFlipZRandomVerticalFlip isinstancestrlistrEdictintr startswithr r r lenfloatZ RandomApplyZ ColorJitterZRandomGrayscaleZ GaussianBlurrr!r"r#r$rr%)rr.r/r0r1r2r3r4r5r6r7r8rrrr9r:r;r<rr=Z primary_tflZ secondary_tflZdisable_color_jitterZ img_size_minZ aa_paramsZ final_tflr'r'r(transforms_imagenet_train;s8                r[)rcrop_pct crop_modecrop_border_pixelsrrrrc sZptt|ttfr>t|dks&ttfdd|D}nt|}||f}g} |rj| t|g7} |dkr| t j |t |dt |g7} n|dkrdd|D} | t ||dd t|| d g7} nL|d |d kr| t j |d t |dg7} n| t |g7} | t |g7} |r*| tg7} n&| t t jt|t|d g7} t | S)a ImageNet-oriented image transform for evaluation and inference. Args: img_size: Target image size. crop_pct: Crop percentage. Defaults to 0.875 when None. crop_mode: Crop mode. One of ['squash', 'border', 'center']. Defaults to 'center' when None. crop_border_pixels: Trim a border of specified # pixels around edge of original image. interpolation: Image interpolation mode. mean: Image normalization mean. std: Image normalization standard deviation. use_prefetcher: Prefetcher enabled. Do not convert image to tensor or normalize. Returns: Composed transform pipeline csg|]}t|qSr')mathfloorrGr\r'r(rJ sz,transforms_imagenet_eval..ZsquashrbordercSsg|]}td|qSrC)rF)rHvr'r'r(rJsrA)rZlongest)fillrr-r)rrSrRrUrYrQr`rarrrr r rrrr!r"r#r$r%) rr\r]r^rrrrZ scale_sizer&rer'rbr(transforms_imagenet_evalsH  rf) input_size is_trainingno_augr0r.r/r1r2r3r4r6r7r8rrrr9r:r;r<r\r]r^tf_preprocessingrr=cCst|ttfr|dd}n|}|rP|rP|r4tdddlm}|||| d}n|rx|rx|rdtdt|| |||d}nZ|rt|||||||| | | | | ||||||||d }n$|rtd t|| ||||||d }|S) a Args: input_size: Target input size (channels, height, width) tuple or size scalar. is_training: Return training (random) transforms. no_aug: Disable augmentation for training (useful for debug). train_crop_mode: Training random crop mode ('rrc', 'rkrc', 'rkrr'). scale: Random resize scale range (crop area, < 1.0 => zoom in). ratio: Random aspect ratio range (crop ratio for RRC, ratio adjustment factor for RKR). hflip: Horizontal flip probability. vflip: Vertical flip probability. color_jitter: Random color jitter component factors (brightness, contrast, saturation, hue). Scalar is applied as (scalar,) * 3 (no hue). color_jitter_prob: Apply color jitter with this probability if not None (for SimlCLR-like aug). grayscale_prob: Probability of converting image to grayscale (for SimCLR-like aug). gaussian_blur_prob: Probability of applying gaussian blur (for SimCLR-like aug). auto_augment: Auto augment configuration string (see auto_augment.py). interpolation: Image interpolation mode. mean: Image normalization mean. std: Image normalization standard deviation. re_prob: Random erasing probability. re_mode: Random erasing fill mode. re_count: Number of random erasing regions. re_num_splits: Control split of random erasing across batch size. crop_pct: Inference crop percentage (output size / resize size). crop_mode: Inference crop mode. One of ['squash', 'border', 'center']. Defaults to 'center' when None. crop_border_pixels: Inference crop border of specified # pixels around edge of original image. tf_preprocessing: Use TF 1.0 inference preprocessing for testing model ports use_prefetcher: Pre-fetcher enabled. Do not convert image to tensor or normalize. separate: Output transforms in 3-stage tuple. Returns: Composed transforms or tuple thereof Nz6Separate transforms not supported for TF preprocessingr)TfPreprocessTransform)rhsizerz-Cannot perform split augmentation with no_aug)rrrr)r0r.r/r1r2r3r4r6r7r8rrrrr9r:r;r<r=z>Separate transforms not supported for validation preprocessing)rrrrr\r]r^) rSrRrUrQZtimm.data.tf_preprocessingrlr)r[rf)rgrhrir0r.r/r1r2r3r4r6r7r8rrrr9r:r;r<r\r]r^rjrr=rrlZ transformr'r'r(create_transform?sn>     rn)$__doc__r`typingrrrr#Z torchvisionrZtimm.data.constantsrrrZtimm.data.auto_augmentr r r Ztimm.data.transformsr r rrrrrrZtimm.data.random_erasingrrWrTboolrZr)r[rfrnr'r'r'r(s (   )   4   S