当前位置:网站首页>[PaddleSeg 源码阅读] PaddleSeg计算Dice
[PaddleSeg 源码阅读] PaddleSeg计算Dice
2022-07-04 03:32:00 【氵文大师】
D i c e Dice Dice 系数是一种集合相似度量函数,通常用于计算两个集合的相似度,取值范围是 [ 0 , 1 ] [0, 1] [0,1]:
D i c e ( X , Y ) = 2 ∣ X ∩ Y ∣ ∣ X ∣ + ∣ Y ∣ Dice(X, Y) = \frac {2 | X\cap Y|} { |X| + |Y| } Dice(X,Y)=∣X∣+∣Y∣2∣X∩Y∣
∣ . ∣ |.| ∣.∣ 表示取集合元素的数量函数
D i c e ( X , Y ) = 两 者 交 集 元 素 的 个 数 × 2 两 者 元 素 数 量 之 和 Dice(X, Y) = \frac {两者交集元素的个数 \times 2} {两者元素数量之和} Dice(X,Y)=两者元素数量之和两者交集元素的个数×2
PaddleSeg 中的实现:
def dice(intersect_area, pred_area, label_area):
""" Calculate DICE. 计算 DICE 值. Args: intersect_area (Tensor): The intersection area of prediction and ground truth on all classes. pred_area (Tensor): The prediction area on all classes. label_area (Tensor): The ground truth area on all classes. Returns: np.ndarray: DICE on all classes. float: mean DICE of all classes. """
# 将三者转化为 np.adarray
intersect_area = intersect_area.numpy()
pred_area = pred_area.numpy()
label_area = label_area.numpy()
# 两个集合元素的数量
union = pred_area + label_area
class_dice = []
for i in range(len(intersect_area)): # 在每一类中开始迭代
if union[i] == 0:
# 如果分子为 0,也就是当前图片没有该类元素
dice = 0
else:
dice = (2 * intersect_area[i]) / union[i]
class_dice.append(dice)
mdice = np.mean(class_dice)
return np.array(class_dice), mdice
intersect_area, pred_area, label_area 这三个参数是通过:
calculate_area(pred, label, num_classes, ignore_index=255)
来计算的,接上一篇博客:
[PaddleSeg 源码阅读] PaddleSeg计算 mIoU
边栏推荐
- 機器學習基礎:用 Lasso 做特征選擇
- [Wu Enda deep learning] beginner learning record 3 (regularization / error reduction)
- 150 ppt! The most complete "fair perception machine learning and data mining" tutorial, Dr. AIST Toshihiro kamishima, Japan
- Nbear introduction and use diagram
- Solve the problems encountered by the laravel framework using mongodb
- XSS prevention
- 查詢效率提昇10倍!3種優化方案,幫你解决MySQL深分頁問題
- What are the conditions for the opening of Tiktok live broadcast preview?
- Want to do something in production? Then try these redis commands
- POSTECH | option compatible reward reverse reinforcement learning
猜你喜欢

PID of sunflower classic

1day vulnerability pushback skills practice (3)

What is the difference between enterprise wechat applet and wechat applet

Setting methods, usage methods and common usage scenarios of environment variables in postman

Setting methods, usage methods and common usage scenarios of environment variables in postman

WordPress collection WordPress hang up collection plug-in
![Stm32bug [the project references devices, files or libraries that are not installed appear in keilmdk]](/img/0d/7a8370d153a8479b706377c3487220.jpg)
Stm32bug [the project references devices, files or libraries that are not installed appear in keilmdk]

How about the ratings of 2022 Spring Festival Gala in all provinces? Map analysis helps you show clearly!

96% of the collected traffic is prevented by bubble mart of cloud hosting

Practical multifunctional toolbox wechat applet source code / support traffic master
随机推荐
Setting methods, usage methods and common usage scenarios of environment variables in postman
No clue about the data analysis report? After reading this introduction of smartbi, you will understand!
I stepped on a foundation pit today
Contest3145 - the 37th game of 2021 freshman individual training match_ 1: Origami
[source code analysis] model parallel distributed training Megatron (5) -- pipestream flush
New year's first race, submit bug reward more!
The first spring of the new year | a full set of property management application templates are presented, and Bi construction is "out of the box"
Zigzag scan
JVM family -- heap analysis
Slurm view node configuration information
Apple submitted the new MAC model to the regulatory database before the spring conference
Solve the problem that the tabbar navigation at the bottom of vantui does not correspond to the page (window.loading.hash)
Jenkins continuous integration environment construction V (Jenkins common construction triggers)
POSTECH | option compatible reward reverse reinforcement learning
長文綜述:大腦中的熵、自由能、對稱性和動力學
JVM family -- monitoring tools
2022 attached lifting scaffold worker (special type of construction work) free test questions and attached lifting scaffold worker (special type of construction work) examination papers 2022 attached
Audio and video technology development weekly | 232
Backpropagation formula derivation [Li Hongyi deep learning version]
Remote work guide