我有8个大型h5文件(每个文件~100 G),每个文件都有许多不同的数据集(例如“x”、“y”、“z”、“h”)。我想将所有8个'x‘和'y’数据集合并到一个test.h5和train.h5文件中。有什么快速的方法吗?总共有800080行,所以我首先创建了列车文件save_file = h5py.File(os.path.join(base_path,'data/train.h5'),'w',libver='latest'),在计算了随机分割后,我创建了数据集:
train_file.create_dataset('x', (num_train, 256, 256, 1))
train_file.create_dataset('y',(num_train,1))
[similarly for test_file]
train_indeces = np.asarray([1]*num_train + [0]*num_test)
np.random.shuffle(train_indeces)然后,我尝试遍历我的8个文件中的每一个,并保存训练/测试。
indeces_index = 0
last_train_index = 0
last_test_index = 0
for e in files:
print(f'FILE: {e}')
rnd_file = h5py.File(f'{base_path}data/{e}', 'r', libver='latest')
for j in tqdm(range(rnd_file['x'].shape[0] )):
if train_indeces[indeces_index]==1:
train_file['x'][last_train_index] = rnd_file['x'][j]
train_file['y'][last_train_index] = rnd_file['y'][j]
last_train_index+=1
else:
test_file['x'][last_test_index] = rnd_file['x'][j]
test_file['y'][last_test_index] = rnd_file['y'][j]
last_test_index +=1
indeces_index +=1
rnd_file.close()但根据我的计算,这将需要12天的时间。有更快的方法吗?提前谢谢。
发布于 2021-04-08 20:17:49
如果我理解你的方法,它有800,080读/写操作。是大量的“写作”让你丧命。为了提高性能,您必须重新排序I/O操作,以便每次读取和写入大量数据。
通常,我会将整个数据集读入数组中,然后将其写入新文件。我通过您的代码阅读,并看到您使用train_indeces随机选择要写入train_file或test_file的数据行。这会使事情变得“有点复杂”。:-)
为了复制随机性,我使用np.where()查找训练和测试行。然后,我使用NumPy“想像力索引”将数据作为数组访问(在转换成列表之后)。然后,我将该数组写入适当数据集中的下一个打开的槽。(我重用了您的三个计数器:indeces_index、last_train_index和last_test_index来跟踪事物。)
我认为这能做你想做的事:
警告:我99%肯定这会起作用,但没有用真实的数据进行测试。
for e in files:
print(f'FILE: {e}')
rnd_file = h5py.File(f'{base_path}data/{e}', 'r', libver='latest')
rnd_size = rnd_file['x'].shape[0]
# get an array with the next "rnd_size" indices
ind_arr = train_indeces[indeces_index:indeces_index+rnd_size]
# Get training data indices where index==1
train_idx = np.where(ind_arr==1)[0] # np.where() returns a tuple
train_size = len(train_idx)
x_train_arr = rnd_file['x'][train_idx.tolist()]
train_file['x'][last_train_index:last_train_index+train_size] = x_train_arr
y_train_arr = rnd_file['y'][train_idx.tolist()]
train_file['y'][last_train_index:last_train_index+train_size] = y_train_arr
# Get test data indices where index==0
test_idx = np.where(ind_arr==0)[0] # np.where() returns a tuple
test_size = len(test_idx)
x_test_arr = rnd_file['x'][test_idx.tolist()]
test_file['x'][last_test_index:last_test_index+test_size] = x_test_arr
y_test_arr = rnd_file['y'][test_idx.tolist()]
test_file['y'][last_test_index:last_test_index+test_size] = y_test_arr
indeces_index += rnd_size
last_train_index+= train_size
last_test_index += test_size
rnd_file.close()您应该考虑使用Python的with/as:上下文管理器打开文件。用这个:
with h5py.File(f'{base_path}data/{e}', 'r', libver='latest') as rnd_file:您不需要使用上下文管理器的rnd_file.close。
而不是这样:
rnd_file = h5py.File(f'{base_path}data/{e}', 'r', libver='latest')https://stackoverflow.com/questions/67009165
复制相似问题