bdpy脑磁数据处理

发现博客上没有关于bdpy包的介绍，这里对其进行简单入门的说明。

bdpy是设计用于脑解码分析的 Python 包，以字典形式进行数据存储，存储数据格式为np.array, 支持的保存文件格式为h5.

列出bdpy的主要功能：

bdata: BdPy data format (BData) core package
dataform: Utilities for various data format
distcomp: Distributed computation utilities
dl: Deep learning utilities
feature: Utilities for DNN features
fig: Utilities for figure creation
ml: Machine learning utilities
mri: MRI utilities
opendata: Open data utilities
preproc: Utilities for preprocessing
stats: Utilities for statistics
util: Miscellaneous utilities

创建bdata， 加载数据

from bdpy import BData

# Create an empty BData instance
bdata = BData()

# Load BData from a file
bdata = BData('data_file.h5')

# Load BData from 'data_file.h5'
bdata.load('data_file.h5')

添加新数据：

import numpy 
# Add new data
x = numpy.random.rand(16, 4)
bdata.add(x, 'random_data')

# Set description of metadata
bdata.set_metadatadescription('random_data', 'Random data')

显示数据：

# Show 'key' and 'description' of metadata
bdata.show_meatadata()

# Get 'value' of the metadata specified by 'key'
voxel_x = bdata.get_metadata('voxel_x', where='VoxelData')

提取数据：

# Get an array of voxel data in V1
data_v1 = bdata.select('ROI_V1')  # shape=(M, num voxels in V1)

# `select` accepts some operators
data_v1v2 = bdata.select('ROI_V1 + ROI_V2')
data_hvc = bdata.select('ROI_LOC + ROI_FFA + ROI_PPA - LOC_LVC')

# Wildcard
data_visual = data.select('ROI_V*')

# Get labels ('image_index') in the dataset
label_a  = bdata.select('image_index')

保存数据：

# Save data
bdata.save('output_file.h5')  
# File format is selected automatically by extension. .mat, .h5,and .npy are supported.