转-matlab读取hdf5

Question

如题所述

Answer 1

MATLAB读取hdf文件

https://blog.csdn.net/good_learner_1/article/details/100047950 

版权

1. 显示hdf文件中的数据集以及属性等等信息

2. 读取hdf文件中的数据

3. 读取数据集的属性

4. 创建hdf文件

5. 往新建的hdf中写入数据

6. 给数据集写属性

官方中文文档

1. 显示hdf文件中的数据集以及属性等等信息

h5disp('文件名.hdf');

1

例如：

h5disp('dataset/sample_A_padded_20160501.hdf');

# 输出：

HDF5 sample_A_padded_20160501.hdf

Group'/'

    Attributes:

'file_format':  '0.2'

Group'/annotations'

        Attributes:

'offset':  1520.000000 3644.000000 3644.000000

Dataset'ids'

Size:432

MaxSize:432

            Datatype:  H5T_STD_U64LE (uint64)

            ChunkSize:  []

            Filters:  none

FillValue:0

Dataset'locations'

Size:3x432

MaxSize:3x432

Datatype:  H5T_IEEE_F32LE (single)

            ChunkSize:  []

            Filters:  none

FillValue:0.000000

Dataset'types'

Size:432

MaxSize:432

            Datatype:  H5T_STRING

StringLength: variable

                Padding: H5T_STR_NULLTERM

CharacterSet: H5T_CSET_UTF8

                Character Type: H5T_C_S1

ChunkSize:432

Filters:  deflate(4)

Group'/annotations/comments'

Dataset'comments'

Size:17

MaxSize:17

                Datatype:  H5T_STRING

StringLength: variable

                    Padding: H5T_STR_NULLTERM

CharacterSet: H5T_CSET_UTF8

                    Character Type: H5T_C_S1

                ChunkSize:  []

                Filters:  none

Dataset'target_ids'

Size:17

MaxSize:17

                Datatype:  H5T_STD_U64LE (uint64)

                ChunkSize:  []

                Filters:  none

FillValue:0

Group'/annotations/presynaptic_site'

Dataset'partners'

Size:2x216

MaxSize:2x216

                Datatype:  H5T_STD_U64LE (uint64)

                ChunkSize:  []

                Filters:  none

FillValue:0

Group '/volumes'

Dataset 'raw'

Size:3072x3072x200

MaxSize:3072x3072x200

            Datatype:  H5T_STD_U8LE (uint8)

ChunkSize:192x96x7

Filters:  deflate(4)

            Attributes:

'resolution':  40.000000 4.000000 4.000000

Group'/volumes/labels'

Dataset'clefts'

Size:1250x1250x125

MaxSize:1250x1250x125

                Datatype:  H5T_STD_U64LE (uint64)

ChunkSize:79x79x4

Filters:  deflate(4)

                Attributes:

'resolution':  40.000000 4.000000 4.000000

'offset':  1520.000000 3644.000000 3644.000000

Dataset'neuron_ids'

Size:1250x1250x125

MaxSize:1250x1250x125

                Datatype:  H5T_STD_U64LE (uint64)

ChunkSize:79x79x4

Filters:  deflate(4)

                Attributes:

'resolution':  40.000000 4.000000 4.000000

'offset':  1520.000000 3644.000000 3644.000000

1

2. 读取hdf文件中的数据

data= h5read('文件名.hdf','数据集名');

1

例如：

raw = h5read('dataset/sample_A_padded_20160501.hdf',' /volumes/raw ');

1

3. 读取数据集的属性

attribute = h5readatt('文件名.hdf','数据集名','属性名');

1

例如：

resolution = h5readatt('dataset/sample_A_padded_20160501.hdf','/volumes/raw','resolution');

1

4. 创建hdf文件

h5create('文件名.hdf','数据集名', 数据大小,'Datatype','数据类型');

1

例如：

h5create('downs_dataset/downsample_A_padded_20160501.hdf','/volumes/raw', size_downs_raw,'Datatype','uint8');

1

5. 往新建的hdf中写入数据

h5write('文件名.hdf','数据集名', 数据变量);

1

例如：

h5write('downs_dataset/downsample_A_padded_20160501.hdf','/volumes/raw', downs_raw);

1

6. 给数据集写属性

h5writeatt('文件名.hdf','数据集名','属性名', 属性变量);

1

例如：

h5writeatt('downs_dataset/downsample_A_padded_20160501.hdf','/volumes/raw','resolution', resolution);

1