From 00a7d30ceccc9d0abb5e435512596d0919443ddf Mon Sep 17 00:00:00 2001
From: bskrlj <bskrlj@outbrain.com>
Date: Tue, 26 Mar 2024 12:49:49 +0100
Subject: [PATCH] counter

---
 .../sketches/counting_ultiloglog.py           | 52 +++++++++----------
 1 file changed, 26 insertions(+), 26 deletions(-)

diff --git a/outrank/algorithms/sketches/counting_ultiloglog.py b/outrank/algorithms/sketches/counting_ultiloglog.py
index 38eed99..6a653d6 100644
--- a/outrank/algorithms/sketches/counting_ultiloglog.py
+++ b/outrank/algorithms/sketches/counting_ultiloglog.py
@@ -61,7 +61,7 @@ def __len__(self):
             return len(self.warmup_set)
 
 
-def cardinality_kernel(algo = 'cache'):
+def cardinality_kernel(algo = 'cache', ground = None):
 
     start_time = time.time()
 
@@ -103,7 +103,7 @@ def cardinality_kernel(algo = 'cache'):
     import seaborn as sns
     import tqdm
     from pympler import asizeof
-#    sns.set_style("whitegrid")
+
     plt.rcParams.update({
         'text.usetex': True,
         'font.family': 'Helvetica',
@@ -114,30 +114,30 @@ def get_random_string(length):
         result_str = ''.join(random.choice(letters) for i in range(length))
         return result_str
 
-    # results_df = []
-    # num_vals = 100000
-    # for _ in range(10):
-    #     for j in tqdm.tqdm(range(1000, 100000, 1000)):
-    #         ground = list(set(np.random.randint(0, j, num_vals).tolist()))
-    #         ground = ground + [
-    #             get_random_string(random.randint(1, 15)) for k in range(j)
-    #         ]
-
-
-    #         for algo in ['Hhll (10)', 'Hhll (10000)', 'hll+ (0.005)', 'hll+ (0.01)', 'set']:
-    #             tp, error = cardinality_kernel(algo)
-    #             results_df.append(
-    #                 {
-    #                     'num_samples': len(ground),
-    #                     'time': tp,
-    #                     'algo': algo,
-    #                     'error': error,
-    #                 }
-    #             )
-
-
-    # out_df = pd.DataFrame(results_df)
-    # out_df.to_csv('backup.csv')
+    results_df = []
+    num_vals = 100000
+    for _ in range(10):
+        for j in tqdm.tqdm(range(1000, 100000, 1000)):
+            ground = list(set(np.random.randint(0, j, num_vals).tolist()))
+            ground = ground + [
+                get_random_string(random.randint(1, 15)) for k in range(j)
+            ]
+
+
+            for algo in ['Hhll (10)', 'Hhll (10000)', 'hll+ (0.005)', 'hll+ (0.01)', 'set']:
+                tp, error = cardinality_kernel(algo, ground)
+                results_df.append(
+                    {
+                        'num_samples': len(ground),
+                        'time': tp,
+                        'algo': algo,
+                        'error': error,
+                    },
+                )
+
+
+    out_df = pd.DataFrame(results_df)
+    out_df.to_csv('backup.csv')
     pals = 'coolwarm'
     out_df = pd.read_csv('backup.csv')
     print(out_df)