def test_mask(self):
for is_3d in [True, False]:
mask = '/tmp/mask.tif'
processing.image_mask_from_vector(vector, raster, mask)
mask_src = gdal.Open(mask)
raster_src = gdal.Open(raster)
mask_proj = osr.SpatialReference(wkt=mask_src.GetProjection())
raster_proj = osr.SpatialReference(wkt=raster_src.GetProjection())
assert (raster_proj.GetAttrValue('projcs') ==
mask_proj.GetAttrValue('projcs'))
assert (mask_src.RasterCount == 1)
assert (mask_src.RasterXSize == raster_src.RasterXSize)
assert (mask_src.RasterYSize == raster_src.RasterYSize)
rM_band = processing.RasterMath(raster, return_3d=is_3d)
for idx, band in enumerate(rM_band.read_band_per_band()):
pass
rM_band.add_function(np.mean,
axis=is_3d + 1,
out_image='/tmp/mean.tif')
rM_band.run()
self.assertRaises(MemoryError, rM_band.run, '1K')
assert (idx + 1 == rM_band.n_bands)
x = rM_band.get_random_block()
assert (x.ndim == is_3d + 2)
os.remove('/tmp/mean.tif')
def test_mask(self) :
for is_3d in [True, False]:
mask = '/tmp/mask.tif'
processing.image_mask_from_vector(vector,raster,mask)
mask_src = gdal.Open(mask)
raster_src = gdal.Open(raster)
assert(mask_src.GetProjection() == raster_src.GetProjection())
assert(mask_src.RasterCount == 1)
assert(mask_src.RasterXSize == raster_src.RasterXSize)
assert(mask_src.RasterYSize == raster_src.RasterYSize)
rM_band = processing.RasterMath(raster,return_3d=is_3d)
for idx,band in enumerate(rM_band.read_band_per_band()):
pass
rM_band.add_function(np.mean,axis=1,out_image='/tmp/mean.tif')
rM_band.run()
assert(idx+1 == rM_band.n_bands)
x = rM_band.get_random_block()
assert(x.ndim == is_3d+2)
os.remove('/tmp/mean.tif')
from museotoolbox.processing import image_mask_from_vector
from osgeo import gdal
import os
import tempfile
tempdir = tempfile.mkdtemp()
import shutil
from sklearn.ensemble import RandomForestClassifier
raster, vector = load_historical_data(low_res=True)
X, y, g = load_historical_data(return_X_y_g=True, low_res=True)
param_grid = dict(n_estimators=[1, 10])
classifier = RandomForestClassifier()
image_mask_from_vector(vector, raster, os.path.join(tempdir, 'mask.tif'))
class TestStats(unittest.TestCase):
def test_superLearner(self):
n_cv = 2
for tf in [True, False]:
verbose = tf + 1
model = ai.SuperLearner(classifier,
param_grid=param_grid,
n_jobs=1,
verbose=verbose)
model.fit(X, y, group=g, standardize=tf, cv=n_cv)
assert (model.predict_array(X).shape == y.shape)
len(model.CV) == n_cv
from matplotlib import pyplot as plt
import numpy as np
##############################################################################
# Load HistoricalMap dataset
# -------------------------------------------
raster, vector = datasets.load_historical_data()
##############################################################################
# Initialize rasterMath with raster
# ------------------------------------
# Set return_3d to True to have full block size (not one pixel per row)
# Create raster mask to only keep pixel inside polygons.
image_mask_from_vector(vector, raster, '/tmp/mask.tif', invert=False)
rM = RasterMath(raster, in_image_mask='/tmp/mask.tif', return_3d=True)
#rM.addInputRaster('/tmp/mask.tif')
print(rM.get_random_block().shape)
#######################
# Plot blocks
x = rM.get_random_block()
rM.add_function(np.mean, '/tmp/mean.tif', axis=2, out_np_dt=np.int16)
rM.run()
from osgeo import gdal
dst = gdal.Open('/tmp/mean.tif')
# -*- coding: utf-8 -*-
import unittest
from shutil import copyfile
import numpy as np
from museotoolbox import processing
from museotoolbox.datasets import load_historical_data
from osgeo import gdal
import os
raster,vector = load_historical_data()
rM = processing.RasterMath(raster)
mask = processing.image_mask_from_vector(vector,raster,'/tmp/mask.tif')
class TestRaster(unittest.TestCase):
def test_convert_datatype(self):
self._assert_np_gdt(np.dtype('uint8').name,gdal.GDT_Byte)
self._assert_np_gdt(np.dtype('int16').name,gdal.GDT_Int16)
self._assert_np_gdt(np.dtype('uint16').name,gdal.GDT_UInt16)
self._assert_np_gdt(np.dtype('int32').name,gdal.GDT_Int32)
self._assert_np_gdt(np.dtype('uint32').name,gdal.GDT_UInt32)
self._assert_np_gdt(np.dtype('int64').name,gdal.GDT_Int32)
self._assert_np_gdt(np.dtype('uint64').name,gdal.GDT_Int32)
self._assert_np_gdt(np.dtype('uint16').name,gdal.GDT_UInt16)
self._assert_np_gdt(np.dtype('float32').name,gdal.GDT_Float32)
self._assert_np_gdt(np.dtype('float64').name,gdal.GDT_Float64)
self._assert_np_gdt(gdal.GDT_Byte,np.uint8)
import numpy as np
from museotoolbox import ai
from museotoolbox.datasets import load_historical_data
from museotoolbox.processing import image_mask_from_vector
from osgeo import gdal
import os
import shutil
from sklearn.ensemble import RandomForestClassifier
raster,vector = load_historical_data(low_res=True)
X,y,g = load_historical_data(return_X_y_g=True,low_res=True)
param_grid = dict(n_estimators=[1,10])
classifier = RandomForestClassifier()
image_mask_from_vector(vector,raster,'/tmp/mask.tif')
class TestStats(unittest.TestCase):
def test_superLearner(self):
n_cv = 2
for tf in [True,False]:
verbose = tf+1
model = ai.SuperLearner(classifier,param_grid=param_grid,n_jobs=1,verbose=verbose)
model.fit(X,y,group=g,standardize=tf,cv=n_cv)
assert(model.predict_array(X).shape == y.shape)
len(model.CV) == n_cv
assert(np.all(model.group == g))
model.predict_image(raster,'/tmp/SuperLearner/class.tif',confidence_per_class='/tmp/SuperLearner/confclass.tif',higher_confidence='/tmp/SuperLearner/higherconf.tif')
assert(model._is_standardized == tf)