diff --git a/clean_python/extractors.py b/clean_python/extractors.py
index eaa7a1f..234b7af 100644
--- a/clean_python/extractors.py
+++ b/clean_python/extractors.py
@@ -43,9 +43,9 @@ class Evaluator(ABC):
     @timeit
     def generate_mask(self, img: FFT, merge=False):
         hash = str(img.intens.shape)
-        self.mask = self.gen_mask_helper(img, hash=hash)
+        self.mask = self.gen_mask_helper(img)
         if merge:
-            self.mask = self.merge_mask_helper(hash=hash)
+            self.mask = self.merge_mask_helper()
             self.eval_points = [[a] for a in np.arange(len(self.eval_points))]
 
     @abstractmethod
diff --git a/clean_python/main.py b/clean_python/main.py
index f1cba20..c78e84d 100644
--- a/clean_python/main.py
+++ b/clean_python/main.py
@@ -1,4 +1,5 @@
 from lattices import SCC_Lattice, VO2_Lattice
+import sys
 from spin_image import SpinImage
 import numpy as np
 import matplotlib.pyplot as plt
@@ -121,7 +122,7 @@ def random(seed):
     for i in tqdm.tqdm(ind):
         maske[np.unravel_index(i, (LEN, LEN))] = True
         counter += 1
-        if np.mod(counter, 100) != 0:
+        if np.mod(counter, 100) != 0 and not i == ind[-1]:
             continue
 
         si.apply_mask(lat.parse_mask(maske))
@@ -204,21 +205,34 @@ def ising(seed, temp=0.5):
              out_2=out_rect[1], out_3=out_rect[2], out_4=out_rect[3])
 
 
+def runner():
+    np.random.seed(1234)
+    seeds = np.random.randint(0, 10000, 200)
+    idx = int(sys.argv[1])
+    if idx < 0:
+        return
+    if idx >= seeds.size:
+        return
+    seed = seeds[idx]
+    random(seed)
+
+
 if __name__ == "__main__":
     np.random.seed(1234)
+    runner()
     # test_me()
     # test_square()
     # test_mixed()
     # plt.show()
     # random(1234)
     # ising(1234)
-    test_pdf()
-    plt.show()
-    exit()
-    for i in np.random.randint(0, 10000, 5):
-        random(i)
-        ising(i, 0.5)
-        ising(i, 1.0)
-        ising(i, 1.5)
+    # test_pdf()
+    # plt.show()
+    # exit()
+    # for i in np.random.randint(0, 10000, 5):
+    #    random(i)
+    #    ising(i, 0.5)
+    #    ising(i, 1.0)
+    #    ising(i, 1.5)
 
     # plt.show()