Merge 517ac89 into 082e167

analytics_utils/__init__.py

@@ -9,6 +9,7 @@
 from scipy import ndimage
 from scipy.io import netcdf
 import csv
+from osgeo import gdal, osr
 
 '''
 Utils class for:
@@ -116,14 +117,14 @@ def writeTXY_to_GeoTiff(array_result, filename):
 		filename: name of output GeoTiff file
 	'''
 
-	no_data_value = array_result['plan']['array_output'].values()[0]['no_data_value']
+	no_data_value = array_result['array_output']['no_data_value']
 
-	dims = array_result['array_result'].shape
-	dim_order = array_result['plan']['array_output'].values()[0]['dimensions_order']
+	dims = array_result['array_output']['shape']
+	dim_order = array_result['array_output']['dimensions_order']
 
-	num_t = array_result['array_result'].shape[0]
-	rows = array_result['array_result'].shape[1]
-	cols = array_result['array_result'].shape[2]
+	num_t = dims[0]
+	rows = int(dims[1])
+	cols = int(dims[2])
 
 	driver = gdal.GetDriverByName('GTiff')
 	dataset = driver.Create(filename, rows, cols, num_t, gdal.GDT_Int16)
@@ -137,15 +138,15 @@ def writeTXY_to_GeoTiff(array_result, filename):
 	dataset.SetProjection(proj.ExportToWkt())
 
 	# set geo transform
-	xmin = array_result['plan']['array_output'].values()[0]['dimensions']['X']['range'][0]
-	ymax = array_result['plan']['array_output'].values()[0]['dimensions']['Y']['range'][1]
+	xmin = array_result['array_output']['dimensions']['X']['range'][0]
+	ymax = array_result['array_output']['dimensions']['Y']['range'][1]
 	pixel_size = 0.00025
 	geotransform = (xmin, pixel_size, 0, ymax,0, -pixel_size)  
 	dataset.SetGeoTransform(geotransform)
 
 	for i in range(num_t):
 		band = dataset.GetRasterBand(i+1)
-		band.WriteArray(array_result['array_result'][i])
+		band.WriteArray(array_result['array_result'].values()[0][i])
 		band.SetNoDataValue(no_data_value)
 		band.FlushCache()
 
@@ -158,11 +159,11 @@ def writeNDVI2NetCDF(array_result, filename):
 		filename: name of output NetCDF file
 	'''
 
-	no_data_value = array_result['plan']['array_output'].values()[0]['no_data_value']
-
-	num_t = array_result['array_result'].shape[0]
-	rows = array_result['array_result'].shape[1]
-	cols = array_result['array_result'].shape[2]
+	no_data_value = array_result['array_output']['no_data_value']
+	dims = array_result['array_output']['shape']
+	num_t = dims[0]
+	rows = int(dims[1])
+	cols = int(dims[2])
 
 	pixel_size = 0.00025
 	grid_size = rows * pixel_size
@@ -231,9 +232,9 @@ def writeToCSV(array_result, filename):
 
 	with open(filename, 'w') as fp:
 		writer = csv.writer(fp, delimiter=',')
-		for i in range(array_result['array_result'].shape[0]):
-			data = array_result['array_result'][i].tolist()
-			if len(array_result['array_result'].shape) == 1:
+		for i in range(int(array_result['array_output']['shape'][0])):
+			data = array_result['array_result'].values()[0][i].tolist()
+			if len(array_result['array_result'].values()[0].shape) == 1:
 				writer.writerow([data])
 			else:
 				writer.writerow(data)
@@ -282,4 +283,4 @@ def get_pqa_mask(pqa_ndarray, good_pixel_masks=[32767,16383,2457], dilation=3):
 
 		for good_pixel_mask in good_pixel_masks:
 			pqa_mask[i][pqa_array == good_pixel_mask] = True
-	return pqa_mask
+	return pqa_mask