官术网_书友最值得收藏!

Using HDF5 with h5py

The h5py module is the most popular way to handle HDF5 files in Python. A new or existing HDF5 file can be opened with the h5py.File() function. After the file is open, its groups can simply be accessed by subscripting the file object as if it was a dictionary object. For example, the following code opens an HDF5 file with h5py and then prints the array stored in the /global_power group:

import h5py
hdf5file = h5py.File('pytable_demo.hdf5')
ds=hdf5file['/global_power']
print(ds)
for i in range(len(ds)):
print(arr[i])
hdf5file.close()

The arr variable prints an HDF5 dataset type:

<HDF5 dataset "global_power": shape (9, 2), type "<f8">
[2.58  0.136]
[2.552 0.1  ]
[2.55 0.1 ]
[2.55 0.1 ]
[2.554 0.1  ]
[2.55 0.1 ]
[2.534 0.096]
[2.484 0.   ]
[2.468 0.   ]

For a new hdf5file, datasets and groups can be created by using the hdf5file.create_dataset() function, returning the dataset object, and the hdf5file.create_group() function, returning the folder object. The hdf5file file object is also a folder object representing /, the root folder. Dataset objects support array style slicing and dicing to set or read values from them. For example, the following code creates an HDF5 file and stores one dataset:

import numpy as np
arr = np.loadtxt('temp.csv', skiprows=1, usecols=(2,3), delimiter=',')

import h5py
hdf5file = h5py.File('h5py_demo.hdf5')
dataset1 = hdf5file.create_dataset('global_power',data=arr)
hdf5file.close()

h5py provides an attrs proxy object with a dictionary-like interface to store and retrieve metadata about the file, folders, and datasets. For example, the following code sets and then prints the dataset and file attribute:

dataset1.attrs['owner']='City Corp.'
print(dataset1.attrs['owner'])

hdf5file.attrs['security_level']='public'
print(hdf5file.attrs['security_level'])

For more information about the h5py library, refer to the documentation at the following link: http://docs.h5py.org/en/latest/index.html.

So far, we have learned about different data formats. Often, large data is stored commercially in databases, therefore we will explore how to access both SQL and NoSQL databases next.

主站蜘蛛池模板: 天长市| 平远县| 平罗县| 修文县| 铜鼓县| 右玉县| 澄城县| 吴川市| 怀来县| 揭东县| 新绛县| 保康县| 肃宁县| 竹山县| 中卫市| 左权县| 文水县| 梁平县| 临西县| 凤阳县| 句容市| 桐乡市| 宣城市| 渑池县| 神农架林区| 阜南县| 宁城县| 古浪县| 德州市| 公安县| 昆山市| 长岭县| 游戏| 涿鹿县| 湄潭县| 商河县| 义乌市| 乐都县| 罗源县| 若羌县| 蓬溪县|