In [ ]:

# NBVAL_IGNORE_OUTPUT
import matplotlib
from IPython.display import HTML

from pyramids.dataset import DatasetCollection

%matplotlib inline

# NBVAL_IGNORE_OUTPUT import matplotlib from IPython.display import HTML from pyramids.dataset import DatasetCollection %matplotlib inline

Read multiple files¶

Reading multiple files is being done on two steps

• First use the read_multiple_files method to parse files names and construct the array that will later have the values
• Second use the open_datacube method to open all the raster files and read a specific band from each file

read_multiple_files¶

The given path points to a directory where all the raster we want to read exists The content of the directory is as following

In [2]:

# NBVAL_IGNORE_OUTPUT
import os

path = r"../../../examples/data/geotiff/rhine"
os.listdir(path)

# NBVAL_IGNORE_OUTPUT import os path = r"../../../examples/data/geotiff/rhine" os.listdir(path)

Out[2]:

['Qtot_1979-01-01.tif',
 'Qtot_1979-01-02.tif',
 'Qtot_1979-01-03.tif',
 'Qtot_1979-01-04.tif',
 'Qtot_1979-01-05.tif',
 'Qtot_1979-01-06.tif',
 'Qtot_1979-01-07.tif',
 'Qtot_1979-01-08.tif',
 'Qtot_1979-01-09.tif',
 'Qtot_1979-01-10.tif']

We need raster names to follow a certain pattern in order to be able to read them with a certain order, in our case there is a date in the file name and using this date we will read the rasters and assign the values of each one in the right location in the array based on their date

Regex pattern¶

the parameter regex_string accepts any regex string and apply it to all file names to extract the string that is needed to order the files, this string can be an integer or a date here are some examples for how the regex_string should look like for different file names

• or

• if there is a number at the beginning of the name

fname = "MSWEP_YYYY_M_D.tif" regex_string = r"\d{4}\d{1}\d{1}"

fname = "MSWEP_YYYY.MM.DD.tif" regex_string = r"\d{4}.\d{2}.\d{2}"

fname = "1_MSWEP_YYYY_M_D.tif" regex_string = r"\d+"

In [ ]:

m_dataset = DatasetCollection.read_multiple_files(
    path,
    with_order=True,
    regex_string=r"\d{4}-\d{2}-\d{2}",
    date=True,
    file_name_data_fmt="%Y-%m-%d",
)

m_dataset = DatasetCollection.read_multiple_files( path, with_order=True, regex_string=r"\d{4}-\d{2}-\d{2}", date=True, file_name_data_fmt="%Y-%m-%d", )

Now the Datacube object is created and we can check it by printing the object

In [4]:

print(m_dataset)

print(m_dataset)

            Files: 10
            Cell size: 5000.0
            EPSG: 4647
            Dimension: 125 * 93
            Mask: 2147483647.0
        

open_datacube¶

To read a specific band from each file and assign it to its location in the array we can pass the band index to the open_datacube method, (the default band value is 0)

In [5]:

m_dataset.open_multi_dataset()

m_dataset.open_multi_dataset()

In [6]:

# NBVAL_IGNORE_OUTPUT
print(m_dataset.values)

# NBVAL_IGNORE_OUTPUT print(m_dataset.values)

[[[0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  ...
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]]

 [[0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  ...
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]]

 [[0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  ...
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]]

 ...

 [[0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  ...
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]]

 [[0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  ...
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]]

 [[0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  ...
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]
  [0. 0. 0. ... 0. 0. 0.]]]

plot¶

To animate the Datacube use the plot function

In [7]:

# NBVAL_IGNORE_OUTPUT
cleo = m_dataset.plot(
    exclude_value=0, text_loc=(1, 3), color_scale="linear", vmin=1, vmax=100
)
print(cleo)

# NBVAL_IGNORE_OUTPUT cleo = m_dataset.plot( exclude_value=0, text_loc=(1, 3), color_scale="linear", vmin=1, vmax=100 ) print(cleo)

Matplotlib backend set to inline for static plots in Jupyter notebook

                    Min: 1.0
                    Max: 3375.0
                    Exclude values: [2147483647.0, 0]
                    RGB: False
                

In [8]:

# NBVAL_IGNORE_OUTPUT
HTML(cleo.anim.to_jshtml())

# NBVAL_IGNORE_OUTPUT HTML(cleo.anim.to_jshtml())

Out[8]:

Once Loop Reflect