weighted derived raster

scripts/ceno_stuff.py

@@ -0,0 +1,67 @@
+from pathlib import Path
+import numpy as np
+import geopandas as gpd
+
+
+# from sklearn.neighbors import BallTree
+#
+data_dir = Path("/home/sudipta/Documents/nci/Ceno_test")
+# shapefile = gpd.read_file(data_dir.joinpath("ceno_polygon_stats.shp"))
+# polygons = shapefile["geometry"]
+# print(polygons)
+#
+# index = BallTree(polygons)
+
+
+# import numpy as np
+# from sklearn.neighbors import KDTree, BallTree
+#
+# polygons = np.array([
+#     [0, 0],
+#     [1, 0],
+#     [1, 1],
+#     [0, 1]
+# ])
+#
+# query_point = np.array([0.5, 0.5])
+#
+# kdtree = KDTree(polygons)
+# distances, indices = kdtree.query(query_point, k=3)
+#
+# print(distances)
+# print(indices)
+
+import rasterio as rio
+
+smoothed = "spline_p1_sm.tif"
+ceno = "euc_geo.tif"
+
+# import IPython; IPython.embed(); import sys; sys.exit()
+src = rio.open(data_dir.joinpath(smoothed))
+X = rio.open(data_dir.joinpath(smoothed)).read(masked=True)
+Y = rio.open(data_dir.joinpath(ceno)).read(masked=True)
+# W = rio.open(data_dir.joinpath(ceno)).read(masked=True)
+
+# _, rows, cols = X.shape
+
+mask = X.mask
+
+greater = Y > 40000
+less = Y < 20000
+W = (40000 - Y)/20000
+W[greater] = 0
+W[less] = 1
+W = np.ma.MaskedArray(data=W, mask=X.mask)
+
+output = X * (1-W**2) + Y*W**2
+
+profile = src.profile
+profile.update({
+    'driver': 'COG'
+})
+
+with rio.open(f'weighted_average_quadratic.tif', 'w', ** profile, compress='lzw') as dst:
+    dst.write(output)
+
+
+# output = X * W**2 + Y*(1-W**2)