Replace inline comments with docstrings (#12764)
* Add docstrings * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Auto-format by https://ultralytics.com/actions * Add docstrings * Add docstrings * Add docstrings * Add docstrings * Auto-format by https://ultralytics.com/actions * Update plots.py Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com> --------- Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com> Co-authored-by: UltralyticsAssistant <web@ultralytics.com>pull/12771/head
Glenn Jocher
GitHub
parent
41603da162
commit
574331f984
|
|
@ -57,8 +57,12 @@ from utils.general import (
|
|||
from utils.torch_utils import copy_attr, smart_inference_mode
|
||||
|
||||
|
||||
def autopad(k, p=None, d=1): # kernel, padding, dilation
|
||||
# Pad to 'same' shape outputs
|
||||
def autopad(k, p=None, d=1):
|
||||
"""
|
||||
Pads kernel to 'same' output shape, adjusting for optional dilation; returns padding size.
|
||||
|
||||
`k`: kernel, `p`: padding, `d`: dilation.
|
||||
"""
|
||||
if d > 1:
|
||||
k = d * (k - 1) + 1 if isinstance(k, int) else [d * (x - 1) + 1 for x in k] # actual kernel-size
|
||||
if p is None:
|
||||
|
|
@ -88,13 +92,19 @@ class Conv(nn.Module):
|
|||
|
||||
class DWConv(Conv):
|
||||
# Depth-wise convolution
|
||||
def __init__(self, c1, c2, k=1, s=1, d=1, act=True): # ch_in, ch_out, kernel, stride, dilation, activation
|
||||
def __init__(self, c1, c2, k=1, s=1, d=1, act=True):
|
||||
"""Initializes a depth-wise convolution layer with optional activation; args: input channels (c1), output
|
||||
channels (c2), kernel size (k), stride (s), dilation (d), and activation flag (act).
|
||||
"""
|
||||
super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), d=d, act=act)
|
||||
|
||||
|
||||
class DWConvTranspose2d(nn.ConvTranspose2d):
|
||||
# Depth-wise transpose convolution
|
||||
def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0): # ch_in, ch_out, kernel, stride, padding, padding_out
|
||||
def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0):
|
||||
"""Initializes a depth-wise transpose convolutional layer for YOLOv5; args: input channels (c1), output channels
|
||||
(c2), kernel size (k), stride (s), input padding (p1), output padding (p2).
|
||||
"""
|
||||
super().__init__(c1, c2, k, s, p1, p2, groups=math.gcd(c1, c2))
|
||||
|
||||
|
||||
|
|
@ -148,7 +158,10 @@ class TransformerBlock(nn.Module):
|
|||
|
||||
class Bottleneck(nn.Module):
|
||||
# Standard bottleneck
|
||||
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion
|
||||
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5):
|
||||
"""Initializes a standard bottleneck layer with optional shortcut and group convolution, supporting channel
|
||||
expansion.
|
||||
"""
|
||||
super().__init__()
|
||||
c_ = int(c2 * e) # hidden channels
|
||||
self.cv1 = Conv(c1, c_, 1, 1)
|
||||
|
|
@ -164,7 +177,10 @@ class Bottleneck(nn.Module):
|
|||
|
||||
class BottleneckCSP(nn.Module):
|
||||
# CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks
|
||||
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
|
||||
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
|
||||
"""Initializes CSP bottleneck with optional shortcuts; args: ch_in, ch_out, number of repeats, shortcut bool,
|
||||
groups, expansion.
|
||||
"""
|
||||
super().__init__()
|
||||
c_ = int(c2 * e) # hidden channels
|
||||
self.cv1 = Conv(c1, c_, 1, 1)
|
||||
|
|
@ -206,7 +222,10 @@ class CrossConv(nn.Module):
|
|||
|
||||
class C3(nn.Module):
|
||||
# CSP Bottleneck with 3 convolutions
|
||||
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
|
||||
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
|
||||
"""Initializes C3 module with options for channel count, bottleneck repetition, shortcut usage, group
|
||||
convolutions, and expansion.
|
||||
"""
|
||||
super().__init__()
|
||||
c_ = int(c2 * e) # hidden channels
|
||||
self.cv1 = Conv(c1, c_, 1, 1)
|
||||
|
|
@ -283,7 +302,13 @@ class SPP(nn.Module):
|
|||
|
||||
class SPPF(nn.Module):
|
||||
# Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher
|
||||
def __init__(self, c1, c2, k=5): # equivalent to SPP(k=(5, 9, 13))
|
||||
def __init__(self, c1, c2, k=5):
|
||||
"""
|
||||
Initializes YOLOv5 SPPF layer with given channels and kernel size for YOLOv5 model, combining convolution and
|
||||
max pooling.
|
||||
|
||||
Equivalent to SPP(k=(5, 9, 13)).
|
||||
"""
|
||||
super().__init__()
|
||||
c_ = c1 // 2 # hidden channels
|
||||
self.cv1 = Conv(c1, c_, 1, 1)
|
||||
|
|
@ -302,19 +327,26 @@ class SPPF(nn.Module):
|
|||
|
||||
class Focus(nn.Module):
|
||||
# Focus wh information into c-space
|
||||
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
|
||||
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):
|
||||
"""Initializes Focus module to concentrate width-height info into channel space with configurable convolution
|
||||
parameters.
|
||||
"""
|
||||
super().__init__()
|
||||
self.conv = Conv(c1 * 4, c2, k, s, p, g, act=act)
|
||||
# self.contract = Contract(gain=2)
|
||||
|
||||
def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2)
|
||||
def forward(self, x):
|
||||
"""Processes input through Focus mechanism, reshaping (b,c,w,h) to (b,4c,w/2,h/2) then applies convolution."""
|
||||
return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1))
|
||||
# return self.conv(self.contract(x))
|
||||
|
||||
|
||||
class GhostConv(nn.Module):
|
||||
# Ghost Convolution https://github.com/huawei-noah/ghostnet
|
||||
def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups
|
||||
def __init__(self, c1, c2, k=1, s=1, g=1, act=True):
|
||||
"""Initializes GhostConv with in/out channels, kernel size, stride, groups, and activation; halves out channels
|
||||
for efficiency.
|
||||
"""
|
||||
super().__init__()
|
||||
c_ = c2 // 2 # hidden channels
|
||||
self.cv1 = Conv(c1, c_, k, s, None, g, act=act)
|
||||
|
|
@ -328,7 +360,8 @@ class GhostConv(nn.Module):
|
|||
|
||||
class GhostBottleneck(nn.Module):
|
||||
# Ghost Bottleneck https://github.com/huawei-noah/ghostnet
|
||||
def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride
|
||||
def __init__(self, c1, c2, k=3, s=1):
|
||||
"""Initializes GhostBottleneck with ch_in `c1`, ch_out `c2`, kernel size `k`, stride `s`; see https://github.com/huawei-noah/ghostnet."""
|
||||
super().__init__()
|
||||
c_ = c2 // 2
|
||||
self.conv = nn.Sequential(
|
||||
|
|
@ -982,10 +1015,14 @@ class Detections:
|
|||
"""Logs the string representation of the current object's state via the LOGGER."""
|
||||
LOGGER.info(self.__str__())
|
||||
|
||||
def __len__(self): # override len(results)
|
||||
def __len__(self):
|
||||
"""Returns the number of results stored, overrides the default len(results)."""
|
||||
return self.n
|
||||
|
||||
def __str__(self): # override print(results)
|
||||
def __str__(self):
|
||||
"""Returns a string representation of the model's results, suitable for printing, overrides default
|
||||
print(results).
|
||||
"""
|
||||
return self._run(pprint=True) # print results
|
||||
|
||||
def __repr__(self):
|
||||
|
|
@ -995,7 +1032,8 @@ class Detections:
|
|||
|
||||
class Proto(nn.Module):
|
||||
# YOLOv5 mask Proto module for segmentation models
|
||||
def __init__(self, c1, c_=256, c2=32): # ch_in, number of protos, number of masks
|
||||
def __init__(self, c1, c_=256, c2=32):
|
||||
"""Initializes YOLOv5 Proto module for segmentation with input, proto, and mask channels configuration."""
|
||||
super().__init__()
|
||||
self.cv1 = Conv(c1, c_, k=3)
|
||||
self.upsample = nn.Upsample(scale_factor=2, mode="nearest")
|
||||
|
|
|
|||
|
|
@ -10,8 +10,12 @@ from utils.downloads import attempt_download
|
|||
|
||||
|
||||
class Sum(nn.Module):
|
||||
# Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070
|
||||
def __init__(self, n, weight=False): # n: number of inputs
|
||||
"""Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070."""
|
||||
|
||||
def __init__(self, n, weight=False):
|
||||
"""Initializes a module to sum outputs of layers with number of inputs `n` and optional weighting, supporting 2+
|
||||
inputs.
|
||||
"""
|
||||
super().__init__()
|
||||
self.weight = weight # apply weights boolean
|
||||
self.iter = range(n - 1) # iter object
|
||||
|
|
@ -32,8 +36,12 @@ class Sum(nn.Module):
|
|||
|
||||
|
||||
class MixConv2d(nn.Module):
|
||||
# Mixed Depth-wise Conv https://arxiv.org/abs/1907.09595
|
||||
def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True): # ch_in, ch_out, kernel, stride, ch_strategy
|
||||
"""Mixed Depth-wise Conv https://arxiv.org/abs/1907.09595."""
|
||||
|
||||
def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True):
|
||||
"""Initializes MixConv2d with mixed depth-wise convolutional layers, taking input and output channels (c1, c2),
|
||||
kernel sizes (k), stride (s), and channel distribution strategy (equal_ch).
|
||||
"""
|
||||
super().__init__()
|
||||
n = len(k) # number of convolutions
|
||||
if equal_ch: # equal c_ per group
|
||||
|
|
|
|||
42
models/tf.py
42
models/tf.py
|
|
@ -190,15 +190,25 @@ class TFFocus(keras.layers.Layer):
|
|||
super().__init__()
|
||||
self.conv = TFConv(c1 * 4, c2, k, s, p, g, act, w.conv)
|
||||
|
||||
def call(self, inputs): # x(b,w,h,c) -> y(b,w/2,h/2,4c)
|
||||
# inputs = inputs / 255 # normalize 0-255 to 0-1
|
||||
def call(self, inputs):
|
||||
"""
|
||||
Performs pixel shuffling and convolution on input tensor, downsampling by 2 and expanding channels by 4.
|
||||
|
||||
Example x(b,w,h,c) -> y(b,w/2,h/2,4c).
|
||||
"""
|
||||
inputs = [inputs[:, ::2, ::2, :], inputs[:, 1::2, ::2, :], inputs[:, ::2, 1::2, :], inputs[:, 1::2, 1::2, :]]
|
||||
return self.conv(tf.concat(inputs, 3))
|
||||
|
||||
|
||||
class TFBottleneck(keras.layers.Layer):
|
||||
# Standard bottleneck
|
||||
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5, w=None): # ch_in, ch_out, shortcut, groups, expansion
|
||||
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5, w=None):
|
||||
"""
|
||||
Initializes a standard bottleneck layer for TensorFlow models, expanding and contracting channels with optional
|
||||
shortcut.
|
||||
|
||||
Arguments are ch_in, ch_out, shortcut, groups, expansion.
|
||||
"""
|
||||
super().__init__()
|
||||
c_ = int(c2 * e) # hidden channels
|
||||
self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
|
||||
|
|
@ -364,7 +374,10 @@ class TFSPPF(keras.layers.Layer):
|
|||
|
||||
class TFDetect(keras.layers.Layer):
|
||||
# TF YOLOv5 Detect layer
|
||||
def __init__(self, nc=80, anchors=(), ch=(), imgsz=(640, 640), w=None): # detection layer
|
||||
def __init__(self, nc=80, anchors=(), ch=(), imgsz=(640, 640), w=None):
|
||||
"""Initializes YOLOv5 detection layer for TensorFlow with configurable classes, anchors, channels, and image
|
||||
size.
|
||||
"""
|
||||
super().__init__()
|
||||
self.stride = tf.convert_to_tensor(w.stride.numpy(), dtype=tf.float32)
|
||||
self.nc = nc # number of classes
|
||||
|
|
@ -454,7 +467,13 @@ class TFProto(keras.layers.Layer):
|
|||
|
||||
class TFUpsample(keras.layers.Layer):
|
||||
# TF version of torch.nn.Upsample()
|
||||
def __init__(self, size, scale_factor, mode, w=None): # warning: all arguments needed including 'w'
|
||||
def __init__(self, size, scale_factor, mode, w=None):
|
||||
"""
|
||||
Initializes a TensorFlow upsampling layer with specified size, scale_factor, and mode, ensuring scale_factor is
|
||||
even.
|
||||
|
||||
Warning: all arguments needed including 'w'
|
||||
"""
|
||||
super().__init__()
|
||||
assert scale_factor % 2 == 0, "scale_factor must be multiple of 2"
|
||||
self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * scale_factor, x.shape[2] * scale_factor), mode)
|
||||
|
|
@ -481,7 +500,8 @@ class TFConcat(keras.layers.Layer):
|
|||
return tf.concat(inputs, self.d)
|
||||
|
||||
|
||||
def parse_model(d, ch, model, imgsz): # model_dict, input_channels(3)
|
||||
def parse_model(d, ch, model, imgsz):
|
||||
"""Parses a model definition dict `d` to create YOLOv5 model layers, including dynamic channel adjustments."""
|
||||
LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}")
|
||||
anchors, nc, gd, gw, ch_mul = (
|
||||
d["anchors"],
|
||||
|
|
@ -562,7 +582,10 @@ def parse_model(d, ch, model, imgsz): # model_dict, input_channels(3)
|
|||
|
||||
class TFModel:
|
||||
# TF YOLOv5 model
|
||||
def __init__(self, cfg="yolov5s.yaml", ch=3, nc=None, model=None, imgsz=(640, 640)): # model, channels, classes
|
||||
def __init__(self, cfg="yolov5s.yaml", ch=3, nc=None, model=None, imgsz=(640, 640)):
|
||||
"""Initializes TF YOLOv5 model with specified configuration, channels, classes, model instance, and input
|
||||
size.
|
||||
"""
|
||||
super().__init__()
|
||||
if isinstance(cfg, dict):
|
||||
self.yaml = cfg # model dict
|
||||
|
|
@ -640,7 +663,10 @@ class AgnosticNMS(keras.layers.Layer):
|
|||
)
|
||||
|
||||
@staticmethod
|
||||
def _nms(x, topk_all=100, iou_thres=0.45, conf_thres=0.25): # agnostic NMS
|
||||
def _nms(x, topk_all=100, iou_thres=0.45, conf_thres=0.25):
|
||||
"""Performs agnostic non-maximum suppression (NMS) on detected objects, filtering based on IoU and confidence
|
||||
thresholds.
|
||||
"""
|
||||
boxes, classes, scores = x
|
||||
class_inds = tf.cast(tf.argmax(classes, axis=-1), tf.float32)
|
||||
scores_inp = tf.reduce_max(scores, -1)
|
||||
|
|
|
|||
|
|
@ -75,7 +75,8 @@ class Detect(nn.Module):
|
|||
dynamic = False # force grid reconstruction
|
||||
export = False # export mode
|
||||
|
||||
def __init__(self, nc=80, anchors=(), ch=(), inplace=True): # detection layer
|
||||
def __init__(self, nc=80, anchors=(), ch=(), inplace=True):
|
||||
"""Initializes YOLOv5 detection layer with specified classes, anchors, channels, and inplace operations."""
|
||||
super().__init__()
|
||||
self.nc = nc # number of classes
|
||||
self.no = nc + 5 # number of outputs per anchor
|
||||
|
|
@ -183,7 +184,8 @@ class BaseModel(nn.Module):
|
|||
if c:
|
||||
LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s} Total")
|
||||
|
||||
def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers
|
||||
def fuse(self):
|
||||
"""Fuses Conv2d() and BatchNorm2d() layers in the model to improve inference speed."""
|
||||
LOGGER.info("Fusing layers... ")
|
||||
for m in self.model.modules():
|
||||
if isinstance(m, (Conv, DWConv)) and hasattr(m, "bn"):
|
||||
|
|
@ -193,7 +195,8 @@ class BaseModel(nn.Module):
|
|||
self.info()
|
||||
return self
|
||||
|
||||
def info(self, verbose=False, img_size=640): # print model information
|
||||
def info(self, verbose=False, img_size=640):
|
||||
"""Prints model information given verbosity and image size, e.g., `info(verbose=True, img_size=640)`."""
|
||||
model_info(self, verbose, img_size)
|
||||
|
||||
def _apply(self, fn):
|
||||
|
|
@ -212,7 +215,8 @@ class BaseModel(nn.Module):
|
|||
|
||||
class DetectionModel(BaseModel):
|
||||
# YOLOv5 detection model
|
||||
def __init__(self, cfg="yolov5s.yaml", ch=3, nc=None, anchors=None): # model, input channels, number of classes
|
||||
def __init__(self, cfg="yolov5s.yaml", ch=3, nc=None, anchors=None):
|
||||
"""Initializes YOLOv5 model with configuration file, input channels, number of classes, and custom anchors."""
|
||||
super().__init__()
|
||||
if isinstance(cfg, dict):
|
||||
self.yaml = cfg # model dict
|
||||
|
|
@ -303,8 +307,12 @@ class DetectionModel(BaseModel):
|
|||
y[-1] = y[-1][:, i:] # small
|
||||
return y
|
||||
|
||||
def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency
|
||||
# https://arxiv.org/abs/1708.02002 section 3.3
|
||||
def _initialize_biases(self, cf=None):
|
||||
"""
|
||||
Initializes biases for YOLOv5's Detect() module, optionally using class frequencies (cf).
|
||||
|
||||
For details see https://arxiv.org/abs/1708.02002 section 3.3.
|
||||
"""
|
||||
# cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1.
|
||||
m = self.model[-1] # Detect() module
|
||||
for mi, s in zip(m.m, m.stride): # from
|
||||
|
|
@ -328,7 +336,10 @@ class SegmentationModel(DetectionModel):
|
|||
|
||||
class ClassificationModel(BaseModel):
|
||||
# YOLOv5 classification model
|
||||
def __init__(self, cfg=None, model=None, nc=1000, cutoff=10): # yaml, model, number of classes, cutoff index
|
||||
def __init__(self, cfg=None, model=None, nc=1000, cutoff=10):
|
||||
"""Initializes YOLOv5 model with config file `cfg`, input channels `ch`, number of classes `nc`, and `cuttoff`
|
||||
index.
|
||||
"""
|
||||
super().__init__()
|
||||
self._from_detection_model(model, nc, cutoff) if model is not None else self._from_yaml(cfg)
|
||||
|
||||
|
|
@ -354,8 +365,8 @@ class ClassificationModel(BaseModel):
|
|||
self.model = None
|
||||
|
||||
|
||||
def parse_model(d, ch): # model_dict, input_channels(3)
|
||||
# Parse a YOLOv5 model.yaml dictionary
|
||||
def parse_model(d, ch):
|
||||
"""Parses a YOLOv5 model from a dict `d`, configuring layers based on input channels `ch` and model architecture."""
|
||||
LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}")
|
||||
anchors, nc, gd, gw, act, ch_mul = (
|
||||
d["anchors"],
|
||||
|
|
|
|||
|
|
@ -95,7 +95,12 @@ WORLD_SIZE = int(os.getenv("WORLD_SIZE", 1))
|
|||
GIT_INFO = check_git_info()
|
||||
|
||||
|
||||
def train(hyp, opt, device, callbacks): # hyp is path/to/hyp.yaml or hyp dictionary
|
||||
def train(hyp, opt, device, callbacks):
|
||||
"""
|
||||
Trains the YOLOv5 model on a dataset, managing hyperparameters, model optimization, logging, and validation.
|
||||
|
||||
`hyp` is path/to/hyp.yaml or hyp dictionary.
|
||||
"""
|
||||
(
|
||||
save_dir,
|
||||
epochs,
|
||||
|
|
|
|||
8
train.py
8
train.py
|
|
@ -100,7 +100,13 @@ WORLD_SIZE = int(os.getenv("WORLD_SIZE", 1))
|
|||
GIT_INFO = check_git_info()
|
||||
|
||||
|
||||
def train(hyp, opt, device, callbacks): # hyp is path/to/hyp.yaml or hyp dictionary
|
||||
def train(hyp, opt, device, callbacks):
|
||||
"""
|
||||
Trains YOLOv5 model with given hyperparameters, options, and device, managing datasets, model architecture, loss
|
||||
computation, and optimizer steps.
|
||||
|
||||
`hyp` argument is path/to/hyp.yaml or hyp dictionary.
|
||||
"""
|
||||
save_dir, epochs, batch_size, weights, single_cls, evolve, data, cfg, resume, noval, nosave, workers, freeze = (
|
||||
Path(opt.save_dir),
|
||||
opt.epochs,
|
||||
|
|
|
|||
|
|
@ -59,8 +59,10 @@ class MemoryEfficientMish(nn.Module):
|
|||
|
||||
|
||||
class FReLU(nn.Module):
|
||||
# FReLU activation https://arxiv.org/abs/2007.11824
|
||||
"""FReLU activation https://arxiv.org/abs/2007.11824."""
|
||||
|
||||
def __init__(self, c1, k=3): # ch_in, kernel
|
||||
"""Initializes FReLU activation with channel `c1` and kernel size `k`."""
|
||||
super().__init__()
|
||||
self.conv = nn.Conv2d(c1, c1, k, 1, 1, groups=c1, bias=False)
|
||||
self.bn = nn.BatchNorm2d(c1)
|
||||
|
|
@ -103,7 +105,8 @@ class MetaAconC(nn.Module):
|
|||
See "Activate or Not: Learning Customized Activation" https://arxiv.org/pdf/2009.04759.pdf.
|
||||
"""
|
||||
|
||||
def __init__(self, c1, k=1, s=1, r=16): # ch_in, kernel, stride, r
|
||||
def __init__(self, c1, k=1, s=1, r=16):
|
||||
"""Initializes MetaAconC with params: channel_in (c1), kernel size (k=1), stride (s=1), reduction (r=16)."""
|
||||
super().__init__()
|
||||
c2 = max(r, c1 // r)
|
||||
self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1))
|
||||
|
|
|
|||
|
|
@ -310,8 +310,13 @@ def mixup(im, labels, im2, labels2):
|
|||
return im, labels
|
||||
|
||||
|
||||
def box_candidates(box1, box2, wh_thr=2, ar_thr=100, area_thr=0.1, eps=1e-16): # box1(4,n), box2(4,n)
|
||||
# Compute candidate boxes: box1 before augment, box2 after augment, wh_thr (pixels), aspect_ratio_thr, area_ratio
|
||||
def box_candidates(box1, box2, wh_thr=2, ar_thr=100, area_thr=0.1, eps=1e-16):
|
||||
"""
|
||||
Filters bounding box candidates by minimum width-height threshold `wh_thr` (pixels), aspect ratio threshold
|
||||
`ar_thr`, and area ratio threshold `area_thr`.
|
||||
|
||||
box1(4,n) is before augmentation, box2(4,n) is after augmentation.
|
||||
"""
|
||||
w1, h1 = box1[2] - box1[0], box1[3] - box1[1]
|
||||
w2, h2 = box2[2] - box2[0], box2[3] - box2[1]
|
||||
ar = np.maximum(w2 / (h2 + eps), h2 / (w2 + eps)) # aspect ratio
|
||||
|
|
@ -380,7 +385,12 @@ class LetterBox:
|
|||
self.auto = auto # pass max size integer, automatically solve for short side using stride
|
||||
self.stride = stride # used with auto
|
||||
|
||||
def __call__(self, im): # im = np.array HWC
|
||||
def __call__(self, im):
|
||||
"""
|
||||
Resizes and pads input image `im` (HWC format) to specified dimensions, maintaining aspect ratio.
|
||||
|
||||
im = np.array HWC
|
||||
"""
|
||||
imh, imw = im.shape[:2]
|
||||
r = min(self.h / imh, self.w / imw) # ratio of new/old
|
||||
h, w = round(imh * r), round(imw * r) # resized image
|
||||
|
|
@ -398,7 +408,12 @@ class CenterCrop:
|
|||
super().__init__()
|
||||
self.h, self.w = (size, size) if isinstance(size, int) else size
|
||||
|
||||
def __call__(self, im): # im = np.array HWC
|
||||
def __call__(self, im):
|
||||
"""
|
||||
Applies center crop to the input image and resizes it to a specified size, maintaining aspect ratio.
|
||||
|
||||
im = np.array HWC
|
||||
"""
|
||||
imh, imw = im.shape[:2]
|
||||
m = min(imh, imw) # min dimension
|
||||
top, left = (imh - m) // 2, (imw - m) // 2
|
||||
|
|
@ -412,7 +427,13 @@ class ToTensor:
|
|||
super().__init__()
|
||||
self.half = half
|
||||
|
||||
def __call__(self, im): # im = np.array HWC in BGR order
|
||||
def __call__(self, im):
|
||||
"""
|
||||
Converts BGR np.array image from HWC to RGB CHW format, and normalizes to [0, 1], with support for FP16 if
|
||||
`half=True`.
|
||||
|
||||
im = np.array HWC in BGR order
|
||||
"""
|
||||
im = np.ascontiguousarray(im.transpose((2, 0, 1))[::-1]) # HWC to CHW -> BGR to RGB -> contiguous
|
||||
im = torch.from_numpy(im) # to torch
|
||||
im = im.half() if self.half else im.float() # uint8 to fp16/32
|
||||
|
|
|
|||
|
|
@ -1060,8 +1060,13 @@ def flatten_recursive(path=DATASETS_DIR / "coco128"):
|
|||
shutil.copyfile(file, new_path / Path(file).name)
|
||||
|
||||
|
||||
def extract_boxes(path=DATASETS_DIR / "coco128"): # from utils.dataloaders import *; extract_boxes()
|
||||
# Convert detection dataset into classification dataset, with one directory per class
|
||||
def extract_boxes(path=DATASETS_DIR / "coco128"):
|
||||
"""
|
||||
Converts a detection dataset to a classification dataset, creating a directory for each class and extracting
|
||||
bounding boxes.
|
||||
|
||||
Example: from utils.dataloaders import *; extract_boxes()
|
||||
"""
|
||||
path = Path(path) # images dir
|
||||
shutil.rmtree(path / "classification") if (path / "classification").is_dir() else None # remove existing
|
||||
files = list(path.rglob("*.*"))
|
||||
|
|
@ -1253,7 +1258,7 @@ class HUBDatasetStats:
|
|||
"""Generates dataset JSON for Ultralytics HUB, optionally saves or prints it; save=bool, verbose=bool."""
|
||||
|
||||
def _round(labels):
|
||||
# Update labels to integer class and 6 decimal place floats
|
||||
"""Rounds class labels to integers and coordinates to 4 decimal places for improved label accuracy."""
|
||||
return [[int(c), *(round(x, 4) for x in points)] for c, *points in labels]
|
||||
|
||||
for split in "train", "val", "test":
|
||||
|
|
|
|||
|
|
@ -351,8 +351,12 @@ def check_online():
|
|||
return run_once() or run_once() # check twice to increase robustness to intermittent connectivity issues
|
||||
|
||||
|
||||
def git_describe(path=ROOT): # path must be a directory
|
||||
# Return human-readable git description, i.e. v5.0-5-g3e25f1e https://git-scm.com/docs/git-describe
|
||||
def git_describe(path=ROOT):
|
||||
"""
|
||||
Returns a human-readable git description of the repository at `path`, or an empty string on failure.
|
||||
|
||||
Example output is 'fv5.0-5-g3e25f1e'. See https://git-scm.com/docs/git-describe.
|
||||
"""
|
||||
try:
|
||||
assert (Path(path) / ".git").is_dir()
|
||||
return check_output(f"git -C {path} describe --tags --long --always", shell=True).decode()[:-1]
|
||||
|
|
@ -767,8 +771,12 @@ def labels_to_image_weights(labels, nc=80, class_weights=np.ones(80)):
|
|||
return (class_weights.reshape(1, nc) * class_counts).sum(1)
|
||||
|
||||
|
||||
def coco80_to_coco91_class(): # converts 80-index (val2014) to 91-index (paper)
|
||||
# https://tech.amikelive.com/node-718/what-object-categories-labels-are-in-coco-dataset/
|
||||
def coco80_to_coco91_class():
|
||||
"""
|
||||
Converts COCO 80-class index to COCO 91-class index used in the paper.
|
||||
|
||||
Reference: https://tech.amikelive.com/node-718/what-object-categories-labels-are-in-coco-dataset/
|
||||
"""
|
||||
# a = np.loadtxt('data/coco.names', dtype='str', delimiter='\n')
|
||||
# b = np.loadtxt('data/coco_paper.names', dtype='str', delimiter='\n')
|
||||
# x1 = [list(a[i] == b).index(True) + 1 for i in range(80)] # darknet to coco
|
||||
|
|
@ -1108,8 +1116,13 @@ def non_max_suppression(
|
|||
return output
|
||||
|
||||
|
||||
def strip_optimizer(f="best.pt", s=""): # from utils.general import *; strip_optimizer()
|
||||
# Strip optimizer from 'f' to finalize training, optionally save as 's'
|
||||
def strip_optimizer(f="best.pt", s=""):
|
||||
"""
|
||||
Strips optimizer and optionally saves checkpoint to finalize training; arguments are file path 'f' and save path
|
||||
's'.
|
||||
|
||||
Example: from utils.general import *; strip_optimizer()
|
||||
"""
|
||||
x = torch.load(f, map_location=torch.device("cpu"))
|
||||
if x.get("ema"):
|
||||
x["model"] = x["ema"] # replace model with ema
|
||||
|
|
|
|||
|
|
@ -8,8 +8,8 @@ from utils.metrics import bbox_iou
|
|||
from utils.torch_utils import de_parallel
|
||||
|
||||
|
||||
def smooth_BCE(eps=0.1): # https://github.com/ultralytics/yolov3/issues/238#issuecomment-598028441
|
||||
# return positive, negative label smoothing BCE targets
|
||||
def smooth_BCE(eps=0.1):
|
||||
"""Returns label smoothing BCE targets for reducing overfitting; pos: `1.0 - 0.5*eps`, neg: `0.5*eps`. For details see ttps://github.com/ultralytics/yolov3/issues/238#issuecomment-598028441"""
|
||||
return 1.0 - 0.5 * eps, 0.5 * eps
|
||||
|
||||
|
||||
|
|
@ -132,6 +132,7 @@ class ComputeLoss:
|
|||
self.device = device
|
||||
|
||||
def __call__(self, p, targets): # predictions, targets
|
||||
"""Performs forward pass, calculating class, box, and object loss for given predictions and targets."""
|
||||
lcls = torch.zeros(1, device=self.device) # class loss
|
||||
lbox = torch.zeros(1, device=self.device) # box loss
|
||||
lobj = torch.zeros(1, device=self.device) # object loss
|
||||
|
|
|
|||
|
|
@ -67,7 +67,8 @@ class Colors:
|
|||
return (c[2], c[1], c[0]) if bgr else c
|
||||
|
||||
@staticmethod
|
||||
def hex2rgb(h): # rgb order (PIL)
|
||||
def hex2rgb(h):
|
||||
"""Converts hexadecimal color `h` to an RGB tuple (PIL-compatible) with order (R, G, B)."""
|
||||
return tuple(int(h[1 + i : 1 + i + 2], 16) for i in (0, 2, 4))
|
||||
|
||||
|
||||
|
|
@ -225,8 +226,13 @@ def plot_lr_scheduler(optimizer, scheduler, epochs=300, save_dir=""):
|
|||
plt.close()
|
||||
|
||||
|
||||
def plot_val_txt(): # from utils.plots import *; plot_val()
|
||||
# Plot val.txt histograms
|
||||
def plot_val_txt():
|
||||
"""
|
||||
Plots 2D and 1D histograms of bounding box centers from 'val.txt' using matplotlib, saving as 'hist2d.png' and
|
||||
'hist1d.png'.
|
||||
|
||||
Example: from utils.plots import *; plot_val()
|
||||
"""
|
||||
x = np.loadtxt("val.txt", dtype=np.float32)
|
||||
box = xyxy2xywh(x[:, :4])
|
||||
cx, cy = box[:, 0], box[:, 1]
|
||||
|
|
@ -242,8 +248,12 @@ def plot_val_txt(): # from utils.plots import *; plot_val()
|
|||
plt.savefig("hist1d.png", dpi=200)
|
||||
|
||||
|
||||
def plot_targets_txt(): # from utils.plots import *; plot_targets_txt()
|
||||
# Plot targets.txt histograms
|
||||
def plot_targets_txt():
|
||||
"""
|
||||
Plots histograms of object detection targets from 'targets.txt', saving the figure as 'targets.jpg'.
|
||||
|
||||
Example: from utils.plots import *; plot_targets_txt()
|
||||
"""
|
||||
x = np.loadtxt("targets.txt", dtype=np.float32).T
|
||||
s = ["x targets", "y targets", "width targets", "height targets"]
|
||||
fig, ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True)
|
||||
|
|
@ -255,8 +265,13 @@ def plot_targets_txt(): # from utils.plots import *; plot_targets_txt()
|
|||
plt.savefig("targets.jpg", dpi=200)
|
||||
|
||||
|
||||
def plot_val_study(file="", dir="", x=None): # from utils.plots import *; plot_val_study()
|
||||
# Plot file=study.txt generated by val.py (or plot all study*.txt in dir)
|
||||
def plot_val_study(file="", dir="", x=None):
|
||||
"""
|
||||
Plots validation study results from 'study*.txt' files in a directory or a specific file, comparing model
|
||||
performance and speed.
|
||||
|
||||
Example: from utils.plots import *; plot_val_study()
|
||||
"""
|
||||
save_dir = Path(file).parent if file else Path(dir)
|
||||
plot2 = False # plot additional results
|
||||
if plot2:
|
||||
|
|
@ -381,8 +396,12 @@ def imshow_cls(im, labels=None, pred=None, names=None, nmax=25, verbose=False, f
|
|||
return f
|
||||
|
||||
|
||||
def plot_evolve(evolve_csv="path/to/evolve.csv"): # from utils.plots import *; plot_evolve()
|
||||
# Plot evolve.csv hyp evolution results
|
||||
def plot_evolve(evolve_csv="path/to/evolve.csv"):
|
||||
"""
|
||||
Plots hyperparameter evolution results from a given CSV, saving the plot and displaying best results.
|
||||
|
||||
Example: from utils.plots import *; plot_evolve()
|
||||
"""
|
||||
evolve_csv = Path(evolve_csv)
|
||||
data = pd.read_csv(evolve_csv)
|
||||
keys = [x.strip() for x in data.columns]
|
||||
|
|
|
|||
|
|
@ -44,6 +44,7 @@ class ComputeLoss:
|
|||
self.device = device
|
||||
|
||||
def __call__(self, preds, targets, masks): # predictions, targets, model
|
||||
"""Evaluates YOLOv5 model's loss for given predictions, targets, and masks; returns total loss components."""
|
||||
p, proto = preds
|
||||
bs, nm, mask_h, mask_w = proto.shape # batch size, number of masks, mask height, mask width
|
||||
lcls = torch.zeros(1, device=self.device)
|
||||
|
|
|
|||
|
|
@ -325,7 +325,9 @@ def model_info(model, verbose=False, imgsz=640):
|
|||
|
||||
|
||||
def scale_img(img, ratio=1.0, same_shape=False, gs=32): # img(16,3,256,416)
|
||||
# Scales img(bs,3,y,x) by ratio constrained to gs-multiple
|
||||
"""Scales an image tensor `img` of shape (bs,3,y,x) by `ratio`, optionally maintaining the original shape, padded to
|
||||
multiples of `gs`.
|
||||
"""
|
||||
if ratio == 1.0:
|
||||
return img
|
||||
h, w = img.shape[2:]
|
||||
|
|
|
|||
Loading…
Reference in New Issue