点击查看代码
# *_*coding:utf-8 *_*
import os
import json
import warnings
import numpy as np
from torch.utils.data import Dataset
warnings.filterwarnings('ignore')
def pc_normalize(pc):
centroid = np.mean(pc, axis=0)
pc = pc - centroid
m = np.max(np.sqrt(np.sum(pc ** 2, axis=1)))
pc = pc / m
return pc
class PartNormalDataset(Dataset):
def __init__(self,root = '/home/yanhua/PycharmProjects/Pointnet_Pointnet2_pytorch-master/data/shapenetcore_partanno_segmentation_benchmark_v0_normal', npoints=2500, split='train', class_choice=None, normal_channel=False):
self.npoints = npoints
self.root = root
self.catfile = os.path.join(self.root, 'synsetoffset2category.txt')
self.cat = {}
self.normal_channel = normal_channel
with open(self.catfile, 'r') as f:
for line in f:
ls = line.strip().split() #strip()用于移除字符串头尾指定的字符(默认为空格或换行符)或字符序列
self.cat[ls[0]] = ls[1] #split()通过指定分隔符对字符串进行分割并返回一个列表，默认分隔符为所有空字符，包括空格、换行(\n)、制表符(\t)等
self.cat = {k: v for k, v in self.cat.items()} #这句代码不是多余？通过实验可知确实多余,self.cat获取了大类名对应的号码
self.classes_original = dict(zip(self.cat, range(len(self.cat)))) #此时字典的键没变，值却变了，作用是为每个大类添加标签
if not class_choice is None: #选择参与训练的数据类别
self.cat = {k:v for k,v in self.cat.items() if k in class_choice}
# print(self.cat)
self.meta = {}
with open(os.path.join(self.root, 'train_test_split', 'shuffled_train_file_list.json'), 'r') as f:
train_ids = set([str(d.split('/')[2]) for d in json.load(f)]) #set()函数创建一个无序不重复元素集;json.load-将json格式字符串转化为dict，读取文件
with open(os.path.join(self.root, 'train_test_split', 'shuffled_val_file_list.json'), 'r') as f:
val_ids = set([str(d.split('/')[2]) for d in json.load(f)])
with open(os.path.join(self.root, 'train_test_split', 'shuffled_test_file_list.json'), 'r') as f:
test_ids = set([str(d.split('/')[2]) for d in json.load(f)])
for item in self.cat:
# print('category', item)
self.meta[item] = []
dir_point = os.path.join(self.root, self.cat[item])
fns = sorted(os.listdir(dir_point)) #sorted()-对所有可迭代的对象进行排序操作,默认升序，此处作用是获取每个大类里所有文件的id
# print(fns[0][0:-4]) #[0:-4]-此时输出的是去掉了文件id后的'.txt'的id
if split == 'trainval': #开始筛选属于train/val/test的item类别的文件
fns = [fn for fn in fns if ((fn[0:-4] in train_ids) or (fn[0:-4] in val_ids))]
elif split == 'train':
fns = [fn for fn in fns if fn[0:-4] in train_ids]
elif split == 'val':
fns = [fn for fn in fns if fn[0:-4] in val_ids]
elif split == 'test':
fns = [fn for fn in fns if fn[0:-4] in test_ids]
else:
print('Unknown split: %s. Exiting..' % (split))
exit(-1) #exit(-1)作用是退出程序
#os.path.basename(path)-返回path最后的文件名,如果path以／或\结尾，那么就会返回空值
for fn in fns:
token = (os.path.splitext(os.path.basename(fn))[0]) #os.path.splitext- 分离文件名与扩展名，默认返回(fname,fextension)元组，可做分片操作
self.meta[item].append(os.path.join(dir_point, token + '.txt'))#两个for循环结束后，此时self.meta包含了所有属于split的单个文件的绝对地址
#此时meta属于一个字典，里头的建是类别，值是列表，列表中又包含了属于该类的文件的绝对地址
#上两句代码明明一句就行啊，干嘛多次一举 self.meta[item].append(os.path.join(dir_point, fn))
self.datapath = []
for item in self.cat:
for fn in self.meta[item]:
self.datapath.append((item, fn)) #此时self.datapath以列表的形式保存所有的文件信息，列表中每个元素都是一个元组，元组的组成为(类别名，文件绝对地址)
self.classes = {}
for i in self.cat.keys():
self.classes[i] = self.classes_original[i] #此时self.classes是一个包含类名信息的字典，键是类名，值是标签
#要是没前头的class_choic，self.classes_original即可代替self.classes的作用
# Mapping from category ('Chair') to a list of int [10,11,12,13] as segmentation labels
self.seg_classes = {'Earphone': [16, 17, 18], 'Motorbike': [30, 31, 32, 33, 34, 35], 'Rocket': [41, 42, 43],
'Car': [8, 9, 10, 11], 'Laptop': [28, 29], 'Cap': [6, 7], 'Skateboard': [44, 45, 46],
'Mug': [36, 37], 'Guitar': [19, 20, 21], 'Bag': [4, 5], 'Lamp': [24, 25, 26, 27],
'Table': [47, 48, 49], 'Airplane': [0, 1, 2, 3], 'Pistol': [38, 39, 40],
'Chair': [12, 13, 14, 15], 'Knife': [22, 23]} #可知此处的seg_classes包含了类别分多少部分，每个部分又是啥标签
# for cat in sorted(self.seg_classes.keys()):
# print(cat, self.seg_classes[cat])
self.cache = {} # from index to (point_set, cls, seg) tuple
self.cache_size = 20000
def __getitem__(self, index):
if index in self.cache:
point_set, cls, seg = self.cache[index]
else:
fn = self.datapath[index]
cat = self.datapath[index][0] #获取类名
cls = self.classes[cat] #获取类名标签
cls = np.array([cls]).astype(np.int32) #将标签数据转为np.int32形式的数组形式
data = np.loadtxt(fn[1]).astype(np.float32) #np.loadtxt(path)-将文件中的每一行数据用[]保存，有多少行就有多少个[]
if not self.normal_channel:
point_set = data[:, 0:3]
else:
point_set = data[:, 0:6]
seg = data[:, -1].astype(np.int32) #取每一行最后一个数据，即取每一点的分割标签，统一用一个[]保存
if len(self.cache) < self.cache_size: #记录调用过的数据，最大不超过20000
self.cache[index] = (point_set, cls, seg)
point_set[:, 0:3] = pc_normalize(point_set[:, 0:3]) #标准化数据
choice = np.random.choice(len(seg), self.npoints, replace=True) #np.random.choice(a,size,replace=True,p=None)-随机采样，replace表示采样的元素是否可以重复，p表示每个元素采样的概率
# resample #choice为选取的采样点
point_set = point_set[choice, :] #提取采样点数据
seg = seg[choice] #提取采样点数据的标签
return point_set, cls, seg
def __len__(self):
return len(self.datapath)
if __name__ == '__main__':
import torch
data=PartNormalDataset(npoints=1250,split='train')
dataloader=torch.utils.data.DataLoader(data,batch_size=12, shuffle=True)
for point_set,cls,seg in dataloader:
print(point_set.shape)
print('-------------')
print(cls.shape)
print('-------------')
print(seg.shape)
break