Python DataGenerator TypeError:不支持*:';发电机&x27;和';浮动';
我创建了一个Python DataGenerator TypeError:不支持*:';发电机&x27;和';浮动';,python,error-handling,typeerror,Python,Error Handling,Typeerror,我创建了一个数据生成器(序列)类: class DataGenerator(Sequence): def __init__(self, x_set, y_set, batch_size): self.x, self.y = x_set, y_set self.batch_size = batch_size def __len__(self): return math.ceil(len(self.x) / self.batch_s
数据生成器(序列)
类:
class DataGenerator(Sequence):
def __init__(self, x_set, y_set, batch_size):
self.x, self.y = x_set, y_set
self.batch_size = batch_size
def __len__(self):
return math.ceil(len(self.x) / self.batch_size)
def __getitem__(self, idx):
batch_x = self.x[idx*self.batch_size : (idx + 1)*self.batch_size]
batch_x = np.array(imread(file_name) for file_name in batch_x)
batch_x = batch_x * 1./255
batch_y = self.y[idx*self.batch_size : (idx + 1)*self.batch_size]
batch_y = np.array(batch_y)
return batch_x, batch_y
这个DataGenerator
应该从两个列表中获取批量数据x_set
是指向图像的文件路径列表y_set
是此图像数据的相应标签列表<代码>批次x是读取并除以255的x\u集的批次<代码>批次y
是y\u集合
的相应批次
之后,我使用此生成器拟合模型:
model.fit_generator(generator=training_generator,
validation_data=validation_generator,
steps_per_epoch = num_train_samples // 128,
validation_steps = num_val_samples // 128,
epochs = 5)
得到了这个错误:
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-77-e56bae3d1c4b> in <module>()
3 steps_per_epoch = num_train_samples // 128,
4 validation_steps = num_val_samples // 128,
----> 5 epochs = 5)
8 frames
<ipython-input-75-6e4037882cc3> in __getitem__(self, idx)
11 batch_x = self.x[idx*self.batch_size : (idx + 1)*self.batch_size]
12 batch_x = np.array(imread(file_name) for file_name in batch_x)
---> 13 batch_x = batch_x * 1./255
14 batch_y = self.y[idx*self.batch_size : (idx + 1)*self.batch_size]
15 batch_y = np.array(batch_y)
TypeError: unsupported operand type(s) for *: 'generator' and 'float'
---------------------------------------------------------------------------
TypeError回溯(最近一次调用上次)
在()
每个历元3个步骤=num列样本//128,
4个验证步骤=num\u val\u样本//128,
---->5个时代=5个)
8帧
in _uGetItem_uuu(self,idx)
11批次x=self.x[idx*自身批次大小:(idx+1)*自身批次大小]
12 batch_x=np.array(batch_x中文件名的imread(文件名))
--->13批次x=批次x*1./255
14批次y=self.y[idx*自身批次大小:(idx+1)*自身批次大小]
15批次y=np.数组(批次y)
TypeError:不支持*:“生成器”和“浮点”的操作数类型
如何修改行batch\u x=batch\u x*1./255
谢谢 试试这个
def __getitem__(self, idx):
batch_x = self.x[idx*self.batch_size : (idx + 1)*self.batch_size]
batch_x = [imread(file_name) for file_name in batch_x]
batch_x = np.array(batch_x)
batch_x = batch_x * 1./255
batch_y = self.y[idx*self.batch_size : (idx + 1)*self.batch_size]
batch_y = np.array(batch_y)
return batch_x, batch_y
是的,很有效!谢谢我不明白的是它非常非常慢…可能是因为批量太大,但是我读到,数据量越大(我有100k个训练文件),批量就应该越大。。。所以,我拿了128。您建议的批处理大小是多少?128批处理大小很好。这种缓慢可能是由于硬件、可用RAM等其他因素造成的。你可以重新启动你的电脑,或者速度很慢。谢谢!我还注意到,我的模型在CPU上的训练速度比在GPU上快。对我来说,这也真的很奇怪。。。