Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →mkurman/albumentations
Fast image augmentation library (Albumentations). 70+ transforms for classification, segmentation, object detection, keypoints, and pose estimation. Optimized OpenCV-based pipeline with unified API across all CV tasks. Supports images, masks, bounding boxes, and keypoints simultaneously. Note: classic Albumentations (MIT) is no longer maintained; successor AlbumentationsX uses AGPL-3.0. For torchvision-native augmentations, use torchvision.transforms.v2.
Passed 21 of 22 test cases with the skill, and 19 of 22 without it.
2 fixed19 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
Documentation for the Albumentations library notes that original repository maintenance has transitioned to a newer community project led by the same team. What is the name of this successor project?
| Checked | Without skill | With skill |
|---|---|---|
| Identifies AlbumentationsX as the successor project | Missed | Passed |
The instructions your model reads when the skill loads.
----|---------------|----------| | Pixel-level | BrightnessContrast, Gamma, HueSaturationValue, CLAHE, Blur, GaussNoise, ISONoise, RGBShift, ChannelShuffle, ToGray, Solarize, Posterize, Equalize, ColorJitter | Color/lighting variation | | Spatial-level | RandomCrop, CenterCrop, Resize, Rotate, Flip, ShiftScaleRotate, Affine, ElasticTransform, GridDistortion, OpticalDistortion, Perspective, PiecewiseAffine | Geometric variation | | Weather | RandomRain, RandomSnow, RandomFog, RandomSunFlare | Adverse conditions | | Advanced | CoarseDropout, Cutout, MixUp, Mosaic | Regularization, domain gap | | Special | CLAHE, Emboss, Sharpen, Superpixels, FDA, HistogramMatching, PixelDistributionAdaptation | Medical, domain adaptation |
# OneOf: apply exactly one transform from a list
transform = A.Compose([
A.OneOf([
A.RandomBrightnessContrast(p=1.0),
A.RandomGamma(p=1.0),
A.HueSaturationValue(p=1.0),
], p=0.8),
A.HorizontalFlip(p=0.5),
])
# SomeOf: apply up to N transforms from a list
transform = A.Compose([
A.SomeOf([
A.GaussNoise(p=1.0),
A.ISONoise(p=1.0),
A.MultiplicativeNoise(p=1.0),
], n=2, replace=False, p=0.5),
])
# Per-transform probability
transform = A.Compose([
A.RandomCrop(256, 256, p=1.0), # Always applied
A.HorizontalFlip(p=0.5), # 50% chance
A.RandomBrightnessContrast(p=0.2), # 20% chance
])
test cases passed, vs 19 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.