From e260e481cffbdda166e0574b5d710caf32919994 Mon Sep 17 00:00:00 2001
From: Yomguithereal <guillaumeplique@gmail.com>
Date: Wed, 20 Jun 2018 18:35:12 +0200
Subject: [PATCH] Improvements

---
 fog/clustering/minhash.py | 2 +-
 fog/lsh/minhash.py        | 4 +---
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/fog/clustering/minhash.py b/fog/clustering/minhash.py
index c6a4299..4712b70 100644
--- a/fog/clustering/minhash.py
+++ b/fog/clustering/minhash.py
@@ -11,7 +11,7 @@ from collections import defaultdict
 import math
 
 from fog.clustering.utils import merge_buckets_into_clusters
-from fog.lsh.minhash import LSBMinHash, MinHash
+from fog.lsh.minhash import MinHash
 from fog.metrics.jaccard import jaccard_similarity
 
 
diff --git a/fog/lsh/minhash.py b/fog/lsh/minhash.py
index e425004..db49150 100644
--- a/fog/lsh/minhash.py
+++ b/fog/lsh/minhash.py
@@ -33,10 +33,8 @@ def crc32(x):
 class MinHash(object):
 
     def __init__(self, h=256, seed=None):
-        # TODO: weighted
         # TODO: cheap_hashes
         # TODO: lsb
-        # TODO: superminhash
 
         rng = Random(seed)
 
@@ -124,7 +122,7 @@ class SuperMinHash(object):
 
             while j <= a:
                 r = rng.random()
-                k = rng.randint(j, m - 1)
+                k = j + math.floor(r * (m - j))
 
                 if q[j] != i:
                     q[j] = i