experiment.py

@@ -1,132 +0,0 @@
-"""Экспериментальное сравнение структур данных."""
-
-import argparse
-import csv
-import random
-from pathlib import Path
-from statistics import mean
-from time import perf_counter
-
-from phonebook import (
-    bst_delete, bst_find, bst_insert,
-    ht_delete, ht_find, ht_insert,
-    ll_delete, ll_find, ll_insert,
-)
-
-
-def generate_records(size):
-    return [(f"User_{index:05d}", f"+7{index:010d}") for index in range(size)]
-
-
-def measure_once(structure, records, existing_names, missing_names, deleted_names, bucket_count):
-    started = perf_counter()
-    if structure == "LinkedList":
-        data = None
-        for name, phone in records:
-            data = ll_insert(data, name, phone)
-        insert_time = perf_counter() - started
-
-        started = perf_counter()
-        for name in existing_names + missing_names:
-            ll_find(data, name)
-        find_time = perf_counter() - started
-
-        started = perf_counter()
-        for name in deleted_names:
-            data = ll_delete(data, name)
-        delete_time = perf_counter() - started
-    elif structure == "HashTable":
-        data = [None] * bucket_count
-        for name, phone in records:
-            ht_insert(data, name, phone)
-        insert_time = perf_counter() - started
-
-        started = perf_counter()
-        for name in existing_names + missing_names:
-            ht_find(data, name)
-        find_time = perf_counter() - started
-
-        started = perf_counter()
-        for name in deleted_names:
-            ht_delete(data, name)
-        delete_time = perf_counter() - started
-    else:
-        data = None
-        for name, phone in records:
-            data = bst_insert(data, name, phone)
-        insert_time = perf_counter() - started
-
-        started = perf_counter()
-        for name in existing_names + missing_names:
-            bst_find(data, name)
-        find_time = perf_counter() - started
-
-        started = perf_counter()
-        for name in deleted_names:
-            data = bst_delete(data, name)
-        delete_time = perf_counter() - started
-
-    return {"insert": insert_time, "find_110": find_time, "delete_50": delete_time}
-
-
-def run_experiment(size=3000, repeats=5, seed=2026, output_dir="docs/data"):
-    rng = random.Random(seed)
-    sorted_records = generate_records(size)
-    shuffled_records = sorted_records.copy()
-    rng.shuffle(shuffled_records)
-    modes = {"shuffled": shuffled_records, "sorted": sorted_records}
-    rows = []
-
-    for mode, records in modes.items():
-        names = [record[0] for record in records]
-        test_cases = []
-        for run in range(1, repeats + 1):
-            test_cases.append((
-                run,
-                rng.sample(names, min(100, size)),
-                [f"None_{index}" for index in range(10)],
-                rng.sample(names, min(50, size)),
-            ))
-        for structure in ("LinkedList", "HashTable", "BST"):
-            for run, existing, missing, deleted in test_cases:
-                timings = measure_once(structure, records, existing, missing, deleted, max(17, size * 2 + 1))
-                for operation, elapsed in timings.items():
-                    rows.append({
-                        "structure": structure,
-                        "mode": mode,
-                        "operation": operation,
-                        "run": run,
-                        "time_seconds": elapsed,
-                    })
-
-    output = Path(output_dir)
-    output.mkdir(parents=True, exist_ok=True)
-    raw_path = output / "results_raw.csv"
-    with raw_path.open("w", newline="", encoding="utf-8-sig") as file:
-        writer = csv.DictWriter(file, fieldnames=rows[0].keys())
-        writer.writeheader()
-        writer.writerows(rows)
-
-    groups = {}
-    for row in rows:
-        key = (row["structure"], row["mode"], row["operation"])
-        groups.setdefault(key, []).append(row["time_seconds"])
-    summary = [
-        {"structure": key[0], "mode": key[1], "operation": key[2], "mean_seconds": mean(values)}
-        for key, values in groups.items()
-    ]
-    summary_path = output / "results_summary.csv"
-    with summary_path.open("w", newline="", encoding="utf-8-sig") as file:
-        writer = csv.DictWriter(file, fieldnames=summary[0].keys())
-        writer.writeheader()
-        writer.writerows(summary)
-    return raw_path, summary_path
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description="Сравнение структур телефонного справочника")
-    parser.add_argument("--size", type=int, default=3000)
-    parser.add_argument("--repeats", type=int, default=5)
-    args = parser.parse_args()
-    raw, summary = run_experiment(args.size, args.repeats)
-    print(f"Полные замеры: {raw}\nСредние значения: {summary}")