Iteration 266: 10 new benchmark pairs (609 total, +10 vs canonical 599)

Added 10 new benchmark pairs for recently merged functions:
- ffill_bfill_series_na: ffillSeries / bfillSeries
- ffill_bfill_df_na: dataFrameFfill / dataFrameBfill
- diff_shift_df_na: diffDataFrame / shiftDataFrame
- interval_range_na: intervalRange
- date_range_stats_na: dateRange (stats module)
- timedelta_ops_na: toTimedelta / formatTimedelta / parseFrac
- date_utils_na: advanceDate / parseFreq / toDateInput
- nunique_df_standalone_na: nunique (DataFrame standalone)
- any_all_reduce_na: anySeries / allSeries / anyDataFrame / allDataFrame
- pct_change_na: pctChangeSeries / pctChangeDataFrame

Run: https://github.com/githubnext/tsessebe/actions/runs/24690878721

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Author: github-actions[bot]

benchmarks/pandas/bench_any_all_reduce_na.py

@@ -0,0 +1,42 @@
+"""
+Benchmark: Series.any() / all() / DataFrame.any() / all() — boolean reductions.
+Outputs JSON: {"function": "any_all_reduce_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import numpy as np
+import pandas as pd
+
+SIZE = 100_000
+ROWS = 10_000
+WARMUP = 5
+ITERATIONS = 100
+
+bool_data = np.arange(SIZE) % 3 != 0
+s = pd.Series(bool_data)
+df = pd.DataFrame({
+    "a": np.arange(ROWS) % 2 == 0,
+    "b": np.arange(ROWS) > ROWS // 2,
+    "c": np.ones(ROWS, dtype=bool),
+})
+
+for _ in range(WARMUP):
+    s.any()
+    s.all()
+    df.any()
+    df.all()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.any()
+    s.all()
+    df.any()
+    df.all()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "any_all_reduce_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_date_range_stats_na.py

@@ -0,0 +1,30 @@
+"""
+Benchmark: pd.date_range — generate date arrays with various frequencies.
+Outputs JSON: {"function": "date_range_stats_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import pandas as pd
+
+WARMUP = 5
+ITERATIONS = 100
+
+start_ = "2020-01-01"
+end_ = "2022-12-31"
+
+for _ in range(WARMUP):
+    pd.date_range(start=start_, end=end_, freq="D")
+    pd.date_range(start=start_, periods=365, freq="D")
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    pd.date_range(start=start_, end=end_, freq="D")
+    pd.date_range(start=start_, periods=365, freq="D")
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "date_range_stats_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_date_utils_na.py

@@ -0,0 +1,40 @@
+"""
+Benchmark: date parsing utilities — equivalent to advanceDate / parseFreq / toDateInput.
+Outputs JSON: {"function": "date_utils_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import pandas as pd
+from datetime import datetime
+
+WARMUP = 5
+ITERATIONS = 200
+
+d = pd.Timestamp("2023-06-15")
+
+for _ in range(WARMUP):
+    pd.tseries.frequencies.to_offset("D")
+    pd.tseries.frequencies.to_offset("MS")
+    d + pd.tseries.frequencies.to_offset("D")
+    d + pd.tseries.frequencies.to_offset("MS")
+    d + pd.tseries.frequencies.to_offset("QS")
+    pd.Timestamp("2023-06-15")
+    pd.Timestamp(1686787200000, unit="ms")
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    pd.tseries.frequencies.to_offset("D")
+    pd.tseries.frequencies.to_offset("MS")
+    d + pd.tseries.frequencies.to_offset("D")
+    d + pd.tseries.frequencies.to_offset("MS")
+    d + pd.tseries.frequencies.to_offset("QS")
+    pd.Timestamp("2023-06-15")
+    pd.Timestamp(1686787200000, unit="ms")
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "date_utils_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_diff_shift_df_na.py

@@ -0,0 +1,35 @@
+"""
+Benchmark: DataFrame.diff() / shift() — diff and shift on 10k-row DataFrame.
+Outputs JSON: {"function": "diff_shift_df_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import numpy as np
+import pandas as pd
+
+ROWS = 10_000
+WARMUP = 5
+ITERATIONS = 50
+
+df = pd.DataFrame({
+    "a": np.arange(ROWS) * 2.0,
+    "b": np.arange(ROWS) * 3.0,
+    "c": np.arange(ROWS) * 0.5,
+})
+
+for _ in range(WARMUP):
+    df.diff(periods=1)
+    df.shift(periods=2)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.diff(periods=1)
+    df.shift(periods=2)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "diff_shift_df_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_ffill_bfill_df_na.py

@@ -0,0 +1,36 @@
+"""
+Benchmark: DataFrame.ffill() / bfill() — forward/backward fill on 10k-row DataFrame.
+Outputs JSON: {"function": "ffill_bfill_df_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import numpy as np
+import pandas as pd
+
+ROWS = 10_000
+WARMUP = 5
+ITERATIONS = 50
+
+data = {}
+for col, offset in zip("abcde", range(5)):
+    arr = np.arange(ROWS, dtype=float) + offset
+    arr[::10] = np.nan
+    data[col] = arr
+df = pd.DataFrame(data)
+
+for _ in range(WARMUP):
+    df.ffill()
+    df.bfill()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.ffill()
+    df.bfill()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "ffill_bfill_df_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_ffill_bfill_series_na.py

@@ -0,0 +1,33 @@
+"""
+Benchmark: Series.ffill() / bfill() — forward/backward fill on 100k-element Series.
+Outputs JSON: {"function": "ffill_bfill_series_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import numpy as np
+import pandas as pd
+
+SIZE = 100_000
+WARMUP = 5
+ITERATIONS = 50
+
+data = np.arange(SIZE, dtype=float) * 1.5
+data[::10] = np.nan
+s = pd.Series(data)
+
+for _ in range(WARMUP):
+    s.ffill()
+    s.bfill()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.ffill()
+    s.bfill()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "ffill_bfill_series_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_interval_range_na.py

@@ -0,0 +1,27 @@
+"""
+Benchmark: pd.interval_range — generate numeric IntervalIndex ranges.
+Outputs JSON: {"function": "interval_range_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import pandas as pd
+
+WARMUP = 5
+ITERATIONS = 100
+
+for _ in range(WARMUP):
+    pd.interval_range(start=0, end=1000, periods=100)
+    pd.interval_range(start=0, periods=500, freq=2)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    pd.interval_range(start=0, end=1000, periods=100)
+    pd.interval_range(start=0, periods=500, freq=2)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "interval_range_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_nunique_df_standalone_na.py

@@ -0,0 +1,37 @@
+"""
+Benchmark: DataFrame.nunique() — count unique values per column.
+Outputs JSON: {"function": "nunique_df_standalone_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import numpy as np
+import pandas as pd
+
+ROWS = 10_000
+WARMUP = 5
+ITERATIONS = 100
+
+df = pd.DataFrame({
+    "a": np.arange(ROWS) % 100,
+    "b": np.arange(ROWS) % 50,
+    "c": np.arange(ROWS) % 200,
+    "d": [None if i % 10 == 0 else i % 75 for i in range(ROWS)],
+    "e": np.arange(ROWS) % 500,
+})
+
+for _ in range(WARMUP):
+    df.nunique()
+    df.nunique(axis=0)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.nunique()
+    df.nunique(axis=0)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "nunique_df_standalone_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_pct_change_na.py

@@ -0,0 +1,39 @@
+"""
+Benchmark: Series.pct_change() / DataFrame.pct_change() — percent change computations.
+Outputs JSON: {"function": "pct_change_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import numpy as np
+import pandas as pd
+
+SIZE = 100_000
+ROWS = 10_000
+WARMUP = 5
+ITERATIONS = 50
+
+s = pd.Series(100 + np.sin(np.arange(SIZE) / 100))
+df = pd.DataFrame({
+    "price": 100 + np.arange(ROWS) * 0.01,
+    "volume": 1000 + (np.arange(ROWS) % 100) * 10,
+    "ratio": 0.5 + np.cos(np.arange(ROWS) / 1000),
+})
+
+for _ in range(WARMUP):
+    s.pct_change()
+    s.pct_change(periods=5)
+    df.pct_change()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.pct_change()
+    s.pct_change(periods=5)
+    df.pct_change()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "pct_change_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_timedelta_ops_na.py

@@ -0,0 +1,32 @@
+"""
+Benchmark: pd.Timedelta parsing / formatting — timedelta ops.
+Outputs JSON: {"function": "timedelta_ops_na", "mean_ms": ..., "iterations": ..., "total_ms": ...}
+"""
+import json
+import time
+import pandas as pd
+
+WARMUP = 5
+ITERATIONS = 100
+
+td = pd.Timedelta(hours=1, minutes=1, seconds=1)
+vals = ["1h", "30min", "2.5s", "100ms", "1D 2h"]
+
+for _ in range(WARMUP):
+    for v in vals:
+        pd.Timedelta(v)
+    str(td)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    for v in vals:
+        pd.Timedelta(v)
+    str(td)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "timedelta_ops_na",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))