expt: Testing epsilon percentages

code/lib/lmdk_sel.py

@@ -391,6 +391,60 @@ def find_lmdks(seq, lmdks, epsilon):
     lmdks_new = seq[lmdks_seq_new - 1]
   return lmdks_new, epsilon - eps_sel
 
+
+def find_lmdks_eps(seq, lmdks, epsilon):
+  '''
+    Add dummy landmarks to original landmarks.
+
+    Parameters:
+      seq     - All of the data points.
+      lmdks   - The original landmarks.
+      epsilon - The available privacy budget.
+
+    Returns:
+      lmdks_new - The new landmarks.
+  '''
+  # The new landmarks
+  lmdks_new = lmdks
+  if len(lmdks) > 0 and len(seq) != len(lmdks):
+    # Get landmarks timestamps in sequence
+    lmdks_seq = find_lmdks_seq(seq, lmdks)
+    # Turn landmarks to histogram
+    hist, h = get_hist(get_seq(1, len(seq)), lmdks_seq)
+    # Find all possible options
+    opts = get_opts_from_top_h(get_seq(1, len(seq)), lmdks_seq)
+    # Get landmarks histogram with dummy landmarks
+    hist_new, _ = exp_mech.exponential(hist, opts, exp_mech.score, 1.0, epsilon)
+    # Split sequence in parts of size h 
+    pt_idx = []
+    for idx in range(1, len(seq), h):
+      pt_idx.append([idx, idx + h - 1])
+    pt_idx[-1][1] = len(seq)
+    # Get new landmarks indexes
+    lmdks_seq_new = np.array([], dtype=int)
+    for i, pt in enumerate(pt_idx):
+      # Already landmarks
+      lmdks_seq_pt = lmdks_seq[(lmdks_seq >= pt[0]) & (lmdks_seq <= pt[1])]
+      # Sample randomly from the rest of the sequence
+      size = hist_new[i] - len(lmdks_seq_pt)
+      rglr = np.setdiff1d(np.arange(pt[0], pt[1] + 1), lmdks_seq_pt)
+      # Add already landmarks
+      lmdks_seq_new = np.concatenate([lmdks_seq_new, lmdks_seq_pt])
+      # Add new landmarks
+      if size > 0 and len(rglr) > size:
+        lmdks_seq_new = np.concatenate([lmdks_seq_new,
+          np.random.choice(
+            rglr, 
+            size = size, 
+            replace = False
+          )
+        ])
+    # Get actual landmarks values
+    lmdks_new = seq[lmdks_seq_new - 1]
+  return lmdks_new
+
+
+
 def test():
   # Start and end points of the sequence
   # # Nonrandom