benchmark script used for some performance optimizations

test/benchmark.py

@@ -0,0 +1,41 @@
+import pandas as pd
+from collections import Counter
+from timeit import timeit
+
+print(
+    "pandas read_csv:",
+    timeit(
+        'pd.read_csv("dataset.csv")',
+        setup='import pandas as pd',
+        number=10
+    )
+)
+
+print(
+    "pandas read_csv dtype=str:",
+    timeit(
+        'pd.read_csv("dataset.csv", dtype=str)',
+        setup='import pandas as pd',
+        number=10
+    )
+)
+
+df = pd.read_csv("dataset.csv")
+
+for col in df.columns:
+    print(col)
+    print(
+        "  pandas value_counts:",
+        timeit(
+            'df[col].value_counts()',
+            globals={"df": df, "col": col},
+            number=10
+        )
+    )
+    print("  Counter:",
+        timeit(
+            'Counter(df[col])',
+            globals={"df": df, "col": col, "Counter": Counter},
+            number=10
+        )
+    )