问题描述
代码如下:
class Dataset(Dataset):
# Constructor
def __init__(self,csv_file,data_dir,transform=None):
# Image directory
self.data_dir=data_dir
# The transform is going to be used on an image.
self.transform = transform
# Load the CSV file that contians image info
self.data_name= pd.read_csv(csv_file)
# Number of images in dataset
self.len=self.data_name.shape[0]
# Get the length
def __len__(self):
return self.len
# Getter
def __getitem__(self,idx):
# Image file path
img_name=self.data_dir + self.data_name.iloc[idx,2]
# Open image file
image = Image.open(img_name)
# The class label for the image
y = self.data_name.iloc[idx,3]
# If there is any transform method,apply it onto the image
if self.transform:
image = self.transform(image)
return image,y
#=================
# Create the dataset objects
train_dataset = Dataset(csv_file=train_csv_file,data_dir='/resources/data/training_data_pytorch/')
我想回答以下问题:
问题:打印出三个样本图像,它们的类别构成训练数据。
样本= [53、23、10]
谢谢!
解决方法
暂无找到可以解决该程序问题的有效方法,小编努力寻找整理中!
如果你已经找到好的解决方法,欢迎将解决方案带上本链接一起发送给小编。
小编邮箱:dio#foxmail.com (将#修改为@)