Merge pull request #166 from githubnext/autoloop/perf-comparison-a1c7965c

[Autoloop] [Autoloop: perf-comparison]

Author: mrjf

benchmarks/pandas/bench_nancumops_extended.py

@@ -0,0 +1,38 @@
+"""Benchmark: nanprod / nanmedian / nancount — nan-ignoring aggregates on a 100k-element array.
+Outputs JSON: {"function": "nancumops_extended", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import math
+import numpy as np
+import pandas as pd
+
+SIZE = 100_000
+WARMUP = 5
+ITERATIONS = 50
+
+# Array with ~10% NaN values; small floats to keep product finite
+data = np.array([
+    np.nan if i % 10 == 0 else 1.0 + math.sin(i * 0.001) * 0.001
+    for i in range(SIZE)
+])
+s = pd.Series(data)
+
+for _ in range(WARMUP):
+    np.nanprod(data)
+    np.nanmedian(data)
+    np.count_nonzero(~np.isnan(data))
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    np.nanprod(data)
+    np.nanmedian(data)
+    np.count_nonzero(~np.isnan(data))
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "nancumops_extended",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_string_ops_extended.py

@@ -0,0 +1,31 @@
+"""Benchmark: string_ops_extended — strip, replace, startswith/endswith on 100k strings"""
+import json, time
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+data = [f"  hello_world_{i % 200}  " for i in range(ROWS)]
+s = pd.Series(data)
+
+for _ in range(WARMUP):
+    s.str.strip()
+    s.str.replace("hello", "hi", regex=False)
+    s.str.startswith("hello")
+    s.str.endswith("world")
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.str.strip()
+    s.str.replace("hello", "hi", regex=False)
+    s.str.startswith("hello")
+    s.str.endswith("world")
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "string_ops_extended",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))