Add docstrings to YOLOv5 functions (#12760)
* Add docstrings to top level files * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Auto-format by https://ultralytics.com/actions * Update activations.py Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Auto-format by https://ultralytics.com/actions --------- Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com> Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
This commit is contained in:
Glenn JocherandUltralyticsAssistant authored and GitHub committed
1 parent
ca00c34c2a
commit
41603da162
42 files changed
+983
-322
No files matched your search
+37
-10
@@ -20,6 +20,7 @@ IMAGENET_STD = 0.229, 0.224, 0.225 # RGB standard deviation
|
||||
class Albumentations:
|
||||
# YOLOv5 Albumentations class (optional, only used if package is installed)
|
||||
def __init__(self, size=640):
|
||||
"""Initializes Albumentations class for optional data augmentation in YOLOv5 with specified input size."""
|
||||
self.transform = None
|
||||
prefix = colorstr("albumentations: ")
|
||||
try:
|
||||
@@ -46,6 +47,7 @@ class Albumentations:
|
||||
LOGGER.info(f"{prefix}{e}")
|
||||
|
||||
def __call__(self, im, labels, p=1.0):
|
||||
"""Applies transformations to an image and labels with probability `p`, returning updated image and labels."""
|
||||
if self.transform and random.random() < p:
|
||||
new = self.transform(image=im, bboxes=labels[:, 1:], class_labels=labels[:, 0]) # transformed
|
||||
im, labels = new["image"], np.array([[c, *b] for c, b in zip(new["class_labels"], new["bboxes"])])
|
||||
@@ -53,19 +55,23 @@ class Albumentations:
|
||||
|
||||
|
||||
def normalize(x, mean=IMAGENET_MEAN, std=IMAGENET_STD, inplace=False):
|
||||
# Denormalize RGB images x per ImageNet stats in BCHW format, i.e. = (x - mean) / std
|
||||
"""
|
||||
Applies ImageNet normalization to RGB images in BCHW format, modifying them in-place if specified.
|
||||
|
||||
Example: y = (x - mean) / std
|
||||
"""
|
||||
return TF.normalize(x, mean, std, inplace=inplace)
|
||||
|
||||
|
||||
def denormalize(x, mean=IMAGENET_MEAN, std=IMAGENET_STD):
|
||||
# Denormalize RGB images x per ImageNet stats in BCHW format, i.e. = x * std + mean
|
||||
"""Reverses ImageNet normalization for BCHW format RGB images by applying `x = x * std + mean`."""
|
||||
for i in range(3):
|
||||
x[:, i] = x[:, i] * std[i] + mean[i]
|
||||
return x
|
||||
|
||||
|
||||
def augment_hsv(im, hgain=0.5, sgain=0.5, vgain=0.5):
|
||||
# HSV color-space augmentation
|
||||
"""Applies HSV color-space augmentation to an image with random gains for hue, saturation, and value."""
|
||||
if hgain or sgain or vgain:
|
||||
r = np.random.uniform(-1, 1, 3) * [hgain, sgain, vgain] + 1 # random gains
|
||||
hue, sat, val = cv2.split(cv2.cvtColor(im, cv2.COLOR_BGR2HSV))
|
||||
@@ -81,7 +87,7 @@ def augment_hsv(im, hgain=0.5, sgain=0.5, vgain=0.5):
|
||||
|
||||
|
||||
def hist_equalize(im, clahe=True, bgr=False):
|
||||
# Equalize histogram on BGR image 'im' with im.shape(n,m,3) and range 0-255
|
||||
"""Equalizes image histogram, with optional CLAHE, for BGR or RGB image with shape (n,m,3) and range 0-255."""
|
||||
yuv = cv2.cvtColor(im, cv2.COLOR_BGR2YUV if bgr else cv2.COLOR_RGB2YUV)
|
||||
if clahe:
|
||||
c = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))
|
||||
@@ -92,7 +98,11 @@ def hist_equalize(im, clahe=True, bgr=False):
|
||||
|
||||
|
||||
def replicate(im, labels):
|
||||
# Replicate labels
|
||||
"""
|
||||
Replicates half of the smallest object labels in an image for data augmentation.
|
||||
|
||||
Returns augmented image and labels.
|
||||
"""
|
||||
h, w = im.shape[:2]
|
||||
boxes = labels[:, 1:].astype(int)
|
||||
x1, y1, x2, y2 = boxes.T
|
||||
@@ -109,7 +119,7 @@ def replicate(im, labels):
|
||||
|
||||
|
||||
def letterbox(im, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True, stride=32):
|
||||
# Resize and pad image while meeting stride-multiple constraints
|
||||
"""Resizes and pads image to new_shape with stride-multiple constraints, returns resized image, ratio, padding."""
|
||||
shape = im.shape[:2] # current shape [height, width]
|
||||
if isinstance(new_shape, int):
|
||||
new_shape = (new_shape, new_shape)
|
||||
@@ -232,7 +242,11 @@ def random_perspective(
|
||||
|
||||
|
||||
def copy_paste(im, labels, segments, p=0.5):
|
||||
# Implement Copy-Paste augmentation https://arxiv.org/abs/2012.07177, labels as nx5 np.array(cls, xyxy)
|
||||
"""
|
||||
Applies Copy-Paste augmentation by flipping and merging segments and labels on an image.
|
||||
|
||||
Details at https://arxiv.org/abs/2012.07177.
|
||||
"""
|
||||
n = len(segments)
|
||||
if p and n:
|
||||
h, w, c = im.shape # height, width, channels
|
||||
@@ -254,7 +268,11 @@ def copy_paste(im, labels, segments, p=0.5):
|
||||
|
||||
|
||||
def cutout(im, labels, p=0.5):
|
||||
# Applies image cutout augmentation https://arxiv.org/abs/1708.04552
|
||||
"""
|
||||
Applies cutout augmentation to an image with optional label adjustment, using random masks of varying sizes.
|
||||
|
||||
Details at https://arxiv.org/abs/1708.04552.
|
||||
"""
|
||||
if random.random() < p:
|
||||
h, w = im.shape[:2]
|
||||
scales = [0.5] * 1 + [0.25] * 2 + [0.125] * 4 + [0.0625] * 8 + [0.03125] * 16 # image size fraction
|
||||
@@ -281,7 +299,11 @@ def cutout(im, labels, p=0.5):
|
||||
|
||||
|
||||
def mixup(im, labels, im2, labels2):
|
||||
# Applies MixUp augmentation https://arxiv.org/pdf/1710.09412.pdf
|
||||
"""
|
||||
Applies MixUp augmentation by blending images and labels.
|
||||
|
||||
See https://arxiv.org/pdf/1710.09412.pdf for details.
|
||||
"""
|
||||
r = np.random.beta(32.0, 32.0) # mixup ratio, alpha=beta=32.0
|
||||
im = (im * r + im2 * (1 - r)).astype(np.uint8)
|
||||
labels = np.concatenate((labels, labels2), 0)
|
||||
@@ -341,7 +363,7 @@ def classify_albumentations(
|
||||
|
||||
|
||||
def classify_transforms(size=224):
|
||||
# Transforms to apply if albumentations not installed
|
||||
"""Applies a series of transformations including center crop, ToTensor, and normalization for classification."""
|
||||
assert isinstance(size, int), f"ERROR: classify_transforms size {size} must be integer, not (list, tuple)"
|
||||
# T.Compose([T.ToTensor(), T.Resize(size), T.CenterCrop(size), T.Normalize(IMAGENET_MEAN, IMAGENET_STD)])
|
||||
return T.Compose([CenterCrop(size), ToTensor(), T.Normalize(IMAGENET_MEAN, IMAGENET_STD)])
|
||||
@@ -350,6 +372,9 @@ def classify_transforms(size=224):
|
||||
class LetterBox:
|
||||
# YOLOv5 LetterBox class for image preprocessing, i.e. T.Compose([LetterBox(size), ToTensor()])
|
||||
def __init__(self, size=(640, 640), auto=False, stride=32):
|
||||
"""Initializes a LetterBox object for YOLOv5 image preprocessing with optional auto sizing and stride
|
||||
adjustment.
|
||||
"""
|
||||
super().__init__()
|
||||
self.h, self.w = (size, size) if isinstance(size, int) else size
|
||||
self.auto = auto # pass max size integer, automatically solve for short side using stride
|
||||
@@ -369,6 +394,7 @@ class LetterBox:
|
||||
class CenterCrop:
|
||||
# YOLOv5 CenterCrop class for image preprocessing, i.e. T.Compose([CenterCrop(size), ToTensor()])
|
||||
def __init__(self, size=640):
|
||||
"""Initializes CenterCrop for image preprocessing, accepting single int or tuple for size, defaults to 640."""
|
||||
super().__init__()
|
||||
self.h, self.w = (size, size) if isinstance(size, int) else size
|
||||
|
||||
@@ -382,6 +408,7 @@ class CenterCrop:
|
||||
class ToTensor:
|
||||
# YOLOv5 ToTensor class for image preprocessing, i.e. T.Compose([LetterBox(size), ToTensor()])
|
||||
def __init__(self, half=False):
|
||||
"""Initializes ToTensor for YOLOv5 image preprocessing, with optional half precision (half=True for FP16)."""
|
||||
super().__init__()
|
||||
self.half = half
|
||||
|
||||
|
||||
Reference in new issue
Block a user