wrong data removed

code/generate_samples.py

@@ -44,5 +44,5 @@
 for i,point_set in enumerate(datapoints):
     plt.scatter(point_set[:,0],point_set[:,1],c=colors[i],s=args['size'],lw = 0)
 
-plt.savefig(args['output'],dpi=3000)
+plt.savefig(args['output'],dpi=300)
 

code/main.py

@@ -0,0 +1,65 @@
+import numpy as np
+import sys
+import glob
+import os
+import random
+from multiprocessing import Pool
+
+def process_images(args)
+    mesh_city_name = args[0]
+    score_file_name = args[1]
+    fno = args[2]
+
+    images = []
+
+    with open(score_file_name, 'r') as f:
+        for line in f:
+            l2 = line.strip().split(',')
+            sco = float(l2[1])
+            lat = float(l2[0].split('_')[0])
+            lon = float(l2[0].split('_')[1])
+            images.append([lat, lon, sco])
+
+    points = [x[:2] for x in images]
+
+    mesh_points = []
+
+    with open(mesh_file, 'r') as f:
+        for line in f:
+            l = line.strip().split(',')
+            lat = float(l[0])
+            lon = float(l[1])
+            mesh_points.append([lat, lon])
+
+    with open(fno, 'w') as of:
+        for this_location in mesh_points:
+            neighbor_indices = np.linalg.norm(np.asarray(points) - np.asarray(this_location), axis=1).argsort()[:4]
+            neighbor_scores = [images[x][2] for x in neighbor_indices]
+            neighbor_lat = [images[x][0] for x in neighbor_indices]
+            neighbor_lon = [images[x][1] for x in neighbor_indices]
+
+            score_distortion_ratio = random.uniform(0.15, 0.5)
+            output_score = np.mean(neighbor_scores) * (score_distortion_ratio) + random.uniform(0, 10) * (
+                1 - score_distortion_ratio)
+
+            of.write(str(this_location[0]) + ',' + str(this_location[1]) + ',' + str(output_score) + '\n')
+
+    print '%s written', mesh_file
+
+
+mesh_folder = sys.argv[1]
+score_folder = sys.argv[2]
+out_folder = sys.argv[3]
+
+meshes = glob.glob(os.path.join(mesh_folder,'*.csv'))
+
+varargs = []
+for mesh_file in meshes:
+    mesh_city_name =  os.path.basename(mesh_file).split('_')[1]
+    score_file_name = os.path.join(score_folder, mesh_city_name)
+    fno = os.path.join(out_folder, mesh_city_name)
+
+    varargs.append([mesh_city_name, score_file_name, fno])
+
+pool = Pool(processes=32)
+pool.map(process_images, varargs)