Dataset

Luo_Suppe2026-08-15

gma.gio.Dataset

[class] Dataset(data = None, shape = None, crs = None, geo_transform = None, dtype = None, nodata = None, metadata = {}, **kwargs):

Create a Dataset object!

**Optional:

data = Dataset||ndarray||list||tuple||None. Default None.

Input data source. Can be a Dataset object, or a NumPy ndarray, list, tuple.

If data is None, a new Dataset will be created according to the shape parameter.

shape = tuple||None. Default None.

Shape of the new Dataset. If shape is a 2-tuple, it represents (rows, columns), and the number of bands is 1. If shape is a 3-tuple, it represents (bands, rows, columns).

If shape is None, data must be set.

crs = str||int||None. Default None.

The output coordinate system. Can be EPSG, WKT, Proj4, and other types of coordinate characters. Default(None) no coordinate system!

geo_transform = tuple||None. Default None.

Geo transformation of the output raster. The affine transformation is not specified by default (None).

dtype = str||None. Default None.

A code or string tag for the output raster data type.

Supported data types include:

'Unknown', 'Byte', 'UInt16', 'Int16', 'UInt32', 'Int32','Float32', 'Float64', 'CInt16', 'CInt32', 'CFloat32', 'CFloat64', ...

For more, see const.RaDT_*.

If dtype is None, the data type is determined by the data parameter.

nodata = float||None. Default None.

Invalid value for output raster. The default is not specified (None).

metadata = dict. Default {}.

Metadata to be added.

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate.

Method

algebraic

[method] .algebraic(agg_fun, block = (None, 2048, 2048), sel_bands = None, scr_nodata = None, dst_nodata = None, **kwargs):

Perform algebraic operations on Dataset.

Parameters:

agg_fun: def.

Calculate functions. For Example:

def func(in_ar):
    return in_ar[0] * 2.1
agg_fun = func

in_ar: Input NumPy arrays with block`s shape.

**Optional:

block = tuple. Default (None, 2048, 2048).

Block sizes for bands, rows, and columns. Use None to represent all.

sel_bands = list||None. Default None.

Extract band numbers. Selecting a subset of bands can reduce memory usage.

The default (None) is all bands.

scr_nodata = float. Default None.

Input nodata value. The default is not specified (None).

dst_nodata = float. Default None.

Output nodata value. The default is scr_nodata (None).

**kwargs:

Other parameters passed to driver.RasterDriver.create.

Returns:

Type: Dataset.

clip

[method] .clip(layer, **kwargs):

Clip off areas that are not covered by the method layer.

Parameters:

layer: Layer||Feature||Geometry||str.

A GMA item use for cliping!

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate

Returns:

Type: Dataset.

extent

[method] .extent(**kwargs):

The extent of a geometry is the bounding rectangle.

**Optional:

**kwargs:

Other parameters passed to _aio.Layer.Layer.

Returns:

Type: Layer.

gen_lonlat

[method] .gen_lonlat(**kwargs):

Generate latitude and longitude raster.

Returns:

Type: Dataset.

Band 1 is longitude, Band 2 is latitude.

get_palette

[method] .get_palette(bid = 1):

Get palette of the band.

**Optional:

bid = int. Default 1.

Number of the element calculation band. Start with 1.

Returns:

Type: matplotlib. colors.ListedColormap or None.

isopleth

[method] .isopleth(interval = 50, base = 0, band = 1, **kwargs):

Generate a smooth curve formed by connecting points with equal quantity index values.

**Optional:

interval = float. Default 50.0.

Isopleth interval.

base = float. Default 0.0.

The offset from the zero point of the start of the interval.

band = int. Default 1.

The number of the band to be calculated starts from 1.

**kwargs:

Other parameters passed to _aio.layer.Layer.

Returns:

Type: Layer.

plot

[method] .plot(ax = None, bands = None, **kwargs):

Make plots of Dataset.

**Optional:

ax = None||matplotlib. ~.AxesSubplot. Default None.

A matplotlib subplot. If None, a default axes will be created.

bands = int||list||None. Default None.

The band ID to be plotted. For multi-band data, only 1, 3 (RGB), and 4 (RGBA) bands are supported for selection. Duplicate bands will only keep one. Default is the first four bands.

**kwargs:

Other parameters passed to matplotlib.image.AxesImage.

Returns:

matplotlib.~.AxesSubplot

reclassify

[method] .reclassify(remap, method = "Range", **kwargs):

Reclassifies (or changes) the values in a raster.

Parameters:

remap: dict.

Remap object. Used to remap an input value or range definition to a new value.

The format is {mapping1: value1, mapping2: value2, ...}.

Note: Items containing NAN will be rejected directly.

**Optional:

method = str. Default 'Range'.

Specifies how to reclassify the values of the input raster. May be: 'Range' or 'Unique'.

'Range': The data in the range will be mapped.

'Unique': Mapped strictly on a one-to-one basis.

For example:

Remap = {1000: 1, 2000: 2, 3000: 3} 1. ReClaMethod = 'Range'.

Data <= 1000: --> 1

1000 < Data <= 1000: --> 2

2000 < Data <= 3000: --> 3

Data > 3000: --> NoData

2. ReClaMethod = 'Unique'.

Data == 1000: --> 1

Data == 2000: --> 2

Data == 3000: --> 3

Other values: --> NoData

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate

Returns:

Type: Dataset.

resample

[method] .resample(resolution, method = "Nearest", **kwargs):

Change the spatial resolution of the Dataset.

Parameters:

resolution: float||list.

Resampling resolution(X, Y).

**Optional:

method = str. Default 'Nearest'.

Resampling method. Supported resampling methods include:

'Nearest', 'Bilinear', 'Cubic', 'CubicSpline', 'Lanczos', 'Average', 'RMS', 'Mode'.

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate

Returns:

Type: Dataset.

set_crs

[method] .set_crs(crs):

Set dataset's crs!

Parameters:

crs: str||int||None.

The output coordinate system. Can be EPSG, WKT, Proj4, and other types of coordinate characters. Default(None) no coordinate system!

set_gcps

[method] .set_gcps(gcps = [(0, 0, 0, 0, 0, '', '')], crs = None):

Set GCPs.

**Optional:

gcps = list[tuple]. Default [(0, 0, 0, 0, 0, '', '')].

list(4 foloat + 2str) of gcps values.

format:

[(x(float), y(float), z(float), pixel(float), line(float), info(str, opt), id(str, opt)), ...]

crs = str||int||None. Default None.

The gcps' coordinate system. Can be EPSG, WKT, Proj4, and other types of coordinate characters. Default(None) no coordinate system!

set_geo_transform

[method] .set_geo_transform(geo_transform):

Set dataset's geo transform!

Parameters:

geo_transform: tuple.

Geo transformation of the output raster(6 float of tuple). format:

[x min, x resolution, x rotation, y max, y rotation, y resolution]

set_metadata

[method] .set_metadata(metadata):

Set dataset's metadata!

Parameters:

metadata: dict.

Metadata to be added.

set_nodata

[method] .set_nodata(nodata, bids = None):

Set dataset's nodata value of band!

Parameters:

nodata: float.

Invalid value for output raster.

**Optional:

bids = int||lint of int||None. Default None.

Number of the element calculation band. Start with 1. None for all bands.

set_palette

[method] .set_palette(palette):

Set dataset's palette of band!

Parameters:

palette: dict||matplotlib. ~.cmap or a list of them.

Palette(color table) of raster bands.

For example:

1.dict({value: (R,G,B,A)}):

palette = {10:(200,50,100,255), 20:(200,50,100,255)}.

2.camp:

import matplotlib.colors as cor palette = cor.ListedColormap(['r', 'g', 'b'], N = 256)

set_statistics

[method] .set_statistics(statistics, bids = None):

Set dataset's statistics of band!

Parameters:

statistics: None||list of tuple(vmin||vmax||vmean||vstddev).

Statistics values. None for automatically compute statistics.

**Optional:

bids = int||lint of int||None. Default None.

Number of the element calculation band. Start with 1. None for all bands.

shift

[method] .shift(distance = 1):

Shift a distance of the dataset. **Warining: This method will modify the original data.

Parameters:

distance = float||list. Default 1.

X, Y shift distance. If a single value is provided, it is assumed that X = Y.

Returns:

Type: Dataset.

to_bytearray

[method] .to_bytearray(row_off = 0, column_off = 0, row_size = None, column_size = None, sel_bands = None):

Reads the raster dataset as a bytearray.

**Optional:

row_off = int. Default 0.

The starting row of the data to be read.

column_off = int. Default 0.

The starting column of the data to be read.

row_size = int. Default None.

The number of rows to read from the start row (the row width of the data window). The default (None) is all rows from the start row.

column_size = int. Default None.

The number of columns to read from the start column (the column width of the data window). Default (None) All columns from the start column.

sel_bands = int||list. Default None.

The band number or list of band numbers to be read. Default (None) All bands (numbering starts from 1).

Returns:

Type: bytearray.

to_crs

[method] .to_crs(crs, **kwargs):

Reproject to another coordinate system.

Parameters:

crs: str||int.

The output coordinate system. Can be EPSG, WKT, Proj4, and other types of coordinate characters.

**Optional:

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate

Returns:

Type: Dataset.

to_dtype

[method] .to_dtype(dtype, **kwargs):

Convert the data type of raster.

Parameters:

dtype: str.

A code or string tag for the output raster data type.

Supported data types include:

'Unknown', 'Byte', 'UInt16', 'Int16', 'UInt32', 'Int32','Float32', 'Float64', 'CInt16', 'CInt32', 'CFloat32', 'CFloat64', ...

for more, see const.RaDT_*.

**Optional:

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate

Returns:

Type: Dataset.

to_file

[method] .to_file(out_dst, driver = "GTiff", **kwargs):

Save Dataset as a raster file.

Parameters:

out_dst: str.

The output raster file path.

**Optional:

driver = str. Default 'GTiff'.

Output file driver. For other drivers, see gft.RasterSummary().CreatableFormats.

**kwargs:

Other parameters passed to _algos.geo.dataset.raster_translate.

to_layer

[method] .to_layer(field = None, band = 1, gtype = "Polygon", mask = False, **kwargs):

Translate Dataset to Layer!

**Optional:

field = str. Default None.

The name of the output vector field. None for band.

band = int. Default 1.

Number of the element calculation band. Start with 1.

gtype = str. Default 'Polygon'.

Geometry type! Can be 'Point', 'MultiPoint', 'Line', 'MultiLine', 'MultiPolygon' or 'Polygon'. The default (None) is the automatic geometry type.

mask = bool. Default False.

Whether to build mask layer(nodata:0, other:1).

**kwargs:

Other parameters passed to _algos.geo.layer.Layer.

Returns:

Type: Layer.

to_numpy

[method] .to_numpy(row_off = 0, column_off = 0, row_size = None, column_size = None, sel_bands = None, mask_nodata = False):

Reads the raster dataset as a 3-ndim Numpy array.

**Optional:

row_off = int. Default 0.

The starting row of the data to be read.

column_off = int. Default 0.

The starting column of the data to be read.

row_size = int. Default None.

The number of rows to read from the start row (the row width of the data window). The default (None) is all rows from the start row.

column_size = int. Default None.

The number of columns to read from the start column (the column width of the data window). Default (None) All columns from the start column.

sel_bands = int||list. Default None.

The band number or list of band numbers to be read. Default (None) All bands (numbering starts from 1).

mask_nodata = bool. Default False.

Whether to mask array with nodata value.

Returns:

Type: array.

write_array

[method] .write_array(ar, row_off = 0, column_off = 0, sel_bands = None):

Write array to dataset.

Parameters:

ar: np. ndarray.

The NumPy array that needs to be set.

**Optional:

row_off = int. Default 0.

The starting row of the data to be read.

column_off = int. Default 0.

The starting column of the data to be read.

sel_bands = int||list. Default None.

The band number or list of band numbers to be read. Default (None) All bands (numbering starts from 1).

Dataset

ON THIS PAGE