code: Comparing Wasserstein and Euclidean distance

code/expt/lmdk_sel_cmp.py

@@ -32,7 +32,8 @@ def main(args):
   plt.xlabel('Landmarks (%)')  # Set x axis label.
   plt.xlim(x_i.min() - x_margin, x_i.max() + x_margin)
   # The y axis
-  plt.ylabel('Mean absolute error')  # Set y axis label.
+  # plt.yscale('log')
+  plt.ylabel('Euclidean distance')  # Set y axis label.
   # Bar offset
   x_offset = -(bar_width/2)*(len(dist_type) - 1)
   for d_i, d in enumerate(dist_type):
@@ -50,8 +51,9 @@ def main(args):
         hist, h = lmdk_lib.get_hist(seq, lmdks)
         opts = lmdk_sel.get_opts_from_top_h(seq, lmdks)
         delta = 1.0
-        res, _ = exp_mech.exponential_pareto(hist, opts, exp_mech.score, delta, epsilon)
+        res, _ = exp_mech.exponential(hist, opts, exp_mech.score, delta, epsilon)
-        mae[n_i] += lmdk_lib.get_norm(hist, res)/args.reps
+        mae[n_i] += lmdk_lib.get_norm(hist, res)/args.reps  # Euclidean
+        # mae[n_i] += lmdk_lib.get_emd(hist, res)/args.reps  # Wasserstein
     print('[OK]', flush=True)
     # Plot bar for current distribution
     plt.bar(
@@ -63,7 +65,7 @@ def main(args):
     )
     # Change offset for next bar
     x_offset += bar_width
-  path = str('../../rslt/lmdk_sel_cmp-pareto/' + 'lmdk_sel_cmp')
+  path = str('../../rslt/lmdk_sel_cmp/' + 'lmdk_sel_cmp-norm')
   # Plot legend
   lmdk_lib.plot_legend()
   # Show plot

code/lib/lmdk_sel.py

@@ -6,6 +6,7 @@ import exp_mech
 import numpy as np
 import random
 import time
+from scipy.spatial.distance import cdist
 
 
 '''
@@ -174,7 +175,8 @@ def get_opts_from_top_h(seq, lmdks):
         hist_tmp = np.copy(hist_cur)
         hist_tmp[i] += 1
         # Find difference from original
-        diff_cur = get_norm(hist, hist_tmp)
+        diff_cur = get_norm(hist, hist_tmp)  # Euclidean
+        # diff_cur = get_emd(hist, hist_tmp)  # Wasserstein
         # Remember if it is the best that you've seen
         if diff_cur < diff_min:
           diff_min = diff_cur
@@ -331,6 +333,12 @@ def find_lmdks(seq, lmdks, epsilon):
   return lmdks_new, epsilon - eps_sel
 
 def test():
+
+  A = np.array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1])
+  B = np.array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])
+  print(get_norm(A, B))
+  exit()
+
   # Start and end points of the sequence
   # # Nonrandom
   # start = 1