Use unique values

pandas-dev · Aug 2, 2023 · e360fea · e360fea
1 parent 467941f
commit e360fea
Show file tree

Hide file tree

Showing 3 changed files with 8 additions and 6 deletions.
diff --git a/asv_bench/benchmarks/array.py b/asv_bench/benchmarks/array.py
@@ -54,7 +54,7 @@ def time_from_tuples(self):
 class StringArray:
     def setup(self):
         N = 100_000
-        values = np.array(["a"] * N, dtype=object)
+        values = np.array([str(i) for i in range(N)], dtype=object)
         self.values_obj = np.array(values, dtype="object")
         self.values_str = np.array(values, dtype="U")
         self.values_list = values.tolist()
@@ -78,7 +78,7 @@ def setup(self, multiple_chunks):
             import pyarrow as pa
         except ImportError:
             raise NotImplementedError
-        strings = np.array(["a"] * 10_000, dtype=object)
+        strings = np.array([str(i) for i in range(10_000)], dtype=object)
         if multiple_chunks:
             chunks = [strings[i : i + 100] for i in range(0, len(strings), 100)]
             self.array = pd.arrays.ArrowStringArray(pa.chunked_array(chunks))
@@ -125,7 +125,7 @@ def setup(self, dtype, hasna):
         elif dtype == "int64[pyarrow]":
             data = np.arange(N)
         elif dtype == "string[pyarrow]":
-            data = np.array(["a"] * N, dtype=object)
+            data = np.array([str(i) for i in range(N)], dtype=object)
         elif dtype == "timestamp[ns][pyarrow]":
             data = pd.date_range("2000-01-01", freq="s", periods=N)
         else:

diff --git a/asv_bench/benchmarks/series_methods.py b/asv_bench/benchmarks/series_methods.py
@@ -104,7 +104,7 @@ def setup(self, dtype):
             data = np.arange(N)
             na_value = NA
         elif dtype in ("string", "string[pyarrow]"):
-            data = np.array(["a"] * N, dtype=object)
+            data = np.array([str(i) * 5 for i in range(N)], dtype=object)
             na_value = NA
         else:
             raise NotImplementedError

diff --git a/asv_bench/benchmarks/strings.py b/asv_bench/benchmarks/strings.py
@@ -34,7 +34,9 @@ class Construction:
     dtype_mapping = {"str": "str", "string[python]": object, "string[pyarrow]": object}
 
     def setup(self, pd_type, dtype):
-        series_arr = np.array(["a" * 10] * 100_000, dtype=self.dtype_mapping[dtype])
+        series_arr = np.array(
+            [str(i) * 10 for i in range(100_000)], dtype=self.dtype_mapping[dtype]
+        )
         if pd_type == "series":
             self.arr = series_arr
         elif pd_type == "frame":
@@ -274,7 +276,7 @@ def time_iter(self, dtype):
 
 class StringArrayConstruction:
     def setup(self):
-        self.series_arr = np.array(["a" * 10] * 10**5, dtype=object)
+        self.series_arr = np.array([str(i) * 10 for i in range(10**5)], dtype=object)
         self.series_arr_nan = np.concatenate([self.series_arr, np.array([NA] * 1000)])
 
     def time_string_array_construction(self):