0.0.195

Author: joocer

orso/dataframe.py

@@ -425,7 +425,7 @@ def __hash__(self):
 
         _hash = 0
         for i, row in enumerate(self._rows):
-            row_hash = xxh64(str(row).encode())
+            row_hash = xxh64(str(row).encode()).intdigest()
             _hash = i ^ _hash ^ row_hash
         return _hash
 

orso/profiler/profiler.py

@@ -79,6 +79,7 @@ def find_mfvs(data, top_n=MOST_FREQUENT_VALUE_SIZE):
 
 def get_kvm_hashes(data, size: int):  # slowest function
     from xxhash import xxh32
+
     min_hashes = []
 
     data = list(set(data))

tests/test_profiler.py

@@ -131,8 +131,8 @@ def test_profile_estimators():
     values = source.fetchone().as_dict
     estimated_cardinality = profile.column("Lauched_at").estimate_cardinality()
     assert (
-        estimated_cardinality * 0.75 < values["unique_timestamps"] < estimated_cardinality * 1.25
-    ), f"{profile.column('Launched_at').estimate_cardinality()} != {values['unique_timestamps']}"
+        estimated_cardinality * 0.70 < values["unique_timestamps"] < estimated_cardinality * 1.30
+    ), f"{profile.column('Lauched_at').estimate_cardinality()} != {values['unique_timestamps']}"
 
     source = opteryx.query("SELECT COUNT(*) AS price_over_100 FROM $missions WHERE Price > 100")
     values = source.fetchone().as_dict