diff --git a/smirnovad/lab1/docs/LAB_REPORT.md b/smirnovad/lab1/docs/LAB_REPORT.md new file mode 100644 index 0000000..53ce205 --- /dev/null +++ b/smirnovad/lab1/docs/LAB_REPORT.md @@ -0,0 +1,17 @@ +# Лабораторная работа №1: Структуры данных + +Выполнен замер производительности на выборке N=10000. +## Сводная таблица (Средние значения) +| Тип | Режим | Вставка | Поиск | Удаление | +| :--- | :--- | :--- | :--- | :--- | +| LinkedList | random | 0.00171 | 0.03253 | 0.01889 | +| HashTable | random | 0.00315 | 0.00008 | 0.00005 | +| BST | random | 0.02405 | 0.00021 | 0.00011 | +| LinkedList | sorted | 0.00139 | 0.03529 | 0.01801 | +| HashTable | sorted | 0.00289 | 0.00008 | 0.00004 | +| BST | sorted | 10.51532 | 0.08273 | 0.05241 | + +## Основные выводы +1. **BST** крайне чувствителен к порядку: на отсортированных данных скорость падает из-за превращения дерева в список. +2. **HashTable** — самая стабильная структура, время операций почти не зависит от входной последовательности. +3. **LinkedList** показывает худшее время на операциях поиска из-за необходимости полного перебора. \ No newline at end of file diff --git a/smirnovad/lab1/docs/data/impact_analysis.png b/smirnovad/lab1/docs/data/impact_analysis.png new file mode 100644 index 0000000..c137def Binary files /dev/null and b/smirnovad/lab1/docs/data/impact_analysis.png differ diff --git a/smirnovad/lab1/docs/data/lab1.py b/smirnovad/lab1/docs/data/lab1.py new file mode 100644 index 0000000..6a9ce22 --- /dev/null +++ b/smirnovad/lab1/docs/data/lab1.py @@ -0,0 +1,221 @@ +import time +import random +import csv +from pathlib import Path +import matplotlib.pyplot as plt +import sys + +# Увеличиваем лимит рекурсии для работы с глубокими деревьями (особенно на сортированных данных) +sys.setrecursionlimit(15000) + +# Настройка путей (используем pathlib для гибкости) +ROOT_DIR = Path(r"C:\Users\andre\2026-rff_mp\smirnovad\lab1") +DOCS_DIR = ROOT_DIR / "docs" +DATA_DIR = DOCS_DIR / "data" + +# Создание необходимых директорий +DATA_DIR.mkdir(parents=True, exist_ok=True) + +# --- 1. СВЯЗНЫЙ СПИСОК (LinkedList) --- +def ll_insert(first_node, name, phone): + """Добавление в начало списка (O(1))""" + return {'name': name, 'phone': phone, 'next': first_node} + +def ll_find(first_node, name): + """Линейный поиск по имени""" + item = first_node + while item: + if item['name'] == name: + return item['phone'] + item = item['next'] + return None + +def ll_delete(first_node, name): + """Удаление узла по имени""" + if not first_node: + return None + if first_node['name'] == name: + return first_node['next'] + + prev = first_node + while prev['next']: + if prev['next']['name'] == name: + prev['next'] = prev['next']['next'] + return first_node + prev = prev['next'] + return first_node + +def ll_list_all(first_node): + """Вывод всех записей в алфавитном порядке""" + result_list = [] + item = first_node + while item: + result_list.append((item['name'], item['phone'])) + item = item['next'] + return sorted(result_list) + +# --- 2. ХЕШ-ТАБЛИЦА (Hash Table) --- +def ht_insert(hash_table, name, phone): + slot = hash(name) % len(hash_table) + hash_table[slot] = ll_insert(hash_table[slot], name, phone) + +def ht_find(hash_table, name): + slot = hash(name) % len(hash_table) + return ll_find(hash_table[slot], name) + +def ht_delete(hash_table, name): + slot = hash(name) % len(hash_table) + hash_table[slot] = ll_delete(hash_table[slot], name) + +def ht_list_all(hash_table): + """Сбор данных из всех бакетов""" + total_data = [] + for bucket in hash_table: + node = bucket + while node: + total_data.append((node['name'], node['phone'])) + node = node['next'] + return sorted(total_data) + +# --- 3. ДВОИЧНОЕ ДЕРЕВО ПОИСКА (BST) --- +def bst_insert(root, name, phone): + if not root: + return {'name': name, 'phone': phone, 'left': None, 'right': None} + if name < root['name']: + root['left'] = bst_insert(root['left'], name, phone) + elif name > root['name']: + root['right'] = bst_insert(root['right'], name, phone) + else: + root['phone'] = phone + return root + +def bst_find(root, name): + if not root: + return None + if root['name'] == name: + return root['phone'] + if name < root['name']: + return bst_find(root['left'], name) + return bst_find(root['right'], name) + +def bst_delete(root, name): + """Удаление узла в BST""" + if not root: + return None + if name < root['name']: + root['left'] = bst_delete(root['left'], name) + elif name > root['name']: + root['right'] = bst_delete(root['right'], name) + else: + if not root['left']: return root['right'] + if not root['right']: return root['left'] + # Поиск минимального в правом поддереве + min_node = root['right'] + while min_node['left']: + min_node = min_node['left'] + root['name'], root['phone'] = min_node['name'], min_node['phone'] + root['right'] = bst_delete(root['right'], min_node['name']) + return root + +# --- ЭКСПЕРИМЕНТАЛЬНАЯ ЧАСТЬ --- +log_entries = [] +stats_summary = [] + +def run_test(structure_name, data_mode, dataset): + print(f"Тестирование: {structure_name} | Режим: {data_mode}") + t_ins, t_find, t_del = [], [], [] + + for run_idx in range(5): # 5 итераций + # Инициализация хранилища + storage = [None] * 1024 if structure_name == "HashTable" else None + + # 1. Замер вставки + start = time.perf_counter() + for n, p in dataset: + if structure_name == "LinkedList": storage = ll_insert(storage, n, p) + elif structure_name == "HashTable": ht_insert(storage, n, p) + elif structure_name == "BST": storage = bst_insert(storage, n, p) + t_ins.append(time.perf_counter() - start) + + # 2. Замер поиска (100 существующих + 10 отсутствующих) + test_names = [x[0] for x in random.sample(dataset, 100)] + [f"Missing_{j}" for j in range(10)] + start = time.perf_counter() + for name_to_find in test_names: + if structure_name == "LinkedList": ll_find(storage, name_to_find) + elif structure_name == "HashTable": ht_find(storage, name_to_find) + elif structure_name == "BST": bst_find(storage, name_to_find) + t_find.append(time.perf_counter() - start) + + # 3. Замер удаления (50 записей) + test_dels = [x[0] for x in random.sample(dataset, 50)] + start = time.perf_counter() + for name_to_del in test_dels: + if structure_name == "LinkedList": storage = ll_delete(storage, name_to_del) + elif structure_name == "HashTable": ht_delete(storage, name_to_del) + elif structure_name == "BST": bst_delete(storage, name_to_del) + t_del.append(time.perf_counter() - start) + + log_entries.append([structure_name, data_mode, f"Run_{run_idx+1}", t_ins[-1], t_find[-1], t_del[-1]]) + + # Считаем среднее + avg_i, avg_f, avg_d = sum(t_ins)/5, sum(t_find)/5, sum(t_del)/5 + stats_summary.append({"type": structure_name, "mode": data_mode, "ins": avg_i, "find": avg_f, "del": avg_d}) + +# Генерация данных +N_COUNT = 10000 +raw_data = [(f"User_{i:05d}", f"{random.randint(100, 999)}-{random.randint(10, 99)}") for i in range(N_COUNT)] +data_shuffled = random.sample(raw_data, len(raw_data)) +data_sorted = sorted(raw_data) + +# Запуск тестов +for mode_label, data_src in [("random", data_shuffled), ("sorted", data_sorted)]: + for s_kind in ["LinkedList", "HashTable", "BST"]: + run_test(s_kind, mode_label, data_src) + +# Сохранение CSV +with open(DATA_DIR / "performance_stats.csv", "w", newline="", encoding="utf-8") as f: + writer = csv.writer(f) + writer.writerow(["Structure", "Input_Mode", "Iteration", "Insert_Sec", "Find_Sec", "Delete_Sec"]) + writer.writerows(log_entries) + +# Построение графиков +def generate_visuals(): + ops = ["Вставка", "Поиск", "Удаление"] + structs = ["LinkedList", "HashTable", "BST"] + palette = ["#3498db", "#9b59b6", "#2ecc71"] # Другие цвета + + # График влияния порядка + fig, axes = plt.subplots(1, 3, figsize=(16, 5)) + fig.suptitle("Анализ влияния упорядоченности данных", fontsize=14) + + for idx, s_name in enumerate(structs): + r_vals = next(s for s in stats_summary if s['type'] == s_name and s['mode'] == "random") + s_vals = next(s for s in stats_summary if s['type'] == s_name and s['mode'] == "sorted") + + pos = [0, 1, 2] + axes[idx].bar([p - 0.2 for p in pos], [r_vals['ins'], r_vals['find'], r_vals['del']], 0.4, label='Random', color=palette[0]) + axes[idx].bar([p + 0.2 for p in pos], [s_vals['ins'], s_vals['find'], s_vals['del']], 0.4, label='Sorted', color="#e74c3c") + axes[idx].set_title(s_name) + axes[idx].set_xticks(pos) + axes[idx].set_xticklabels(ops) + axes[idx].legend() + + plt.tight_layout() + plt.savefig(DATA_DIR / "impact_analysis.png") + +generate_visuals() + +# Генерация отчета +with open(DOCS_DIR / "LAB_REPORT.md", "w", encoding="utf-8") as f: + f.write("# Лабораторная работа №1: Структуры данных\n\n") + f.write(f"Выполнен замер производительности на выборке N={N_COUNT}.\n") + f.write("## Сводная таблица (Средние значения)\n") + f.write("| Тип | Режим | Вставка | Поиск | Удаление |\n| :--- | :--- | :--- | :--- | :--- |\n") + for s in stats_summary: + f.write(f"| {s['type']} | {s['mode']} | {s['ins']:.5f} | {s['find']:.5f} | {s['del']:.5f} |\n") + f.write("\n## Основные выводы\n") + f.write("1. **BST** крайне чувствителен к порядку: на отсортированных данных скорость падает из-за превращения дерева в список.\n") + f.write("2. **HashTable** — самая стабильная структура, время операций почти не зависит от входной последовательности.\n") + f.write("3. **LinkedList** показывает худшее время на операциях поиска из-за необходимости полного перебора.") + +print(f"Все файлы успешно сохранены в {DOCS_DIR}") \ No newline at end of file diff --git a/smirnovad/lab1/docs/data/performance_stats.csv b/smirnovad/lab1/docs/data/performance_stats.csv new file mode 100644 index 0000000..2dff65b --- /dev/null +++ b/smirnovad/lab1/docs/data/performance_stats.csv @@ -0,0 +1,31 @@ +Structure,Input_Mode,Iteration,Insert_Sec,Find_Sec,Delete_Sec +LinkedList,random,Run_1,0.0023578000254929066,0.031007900135591626,0.01914949994534254 +LinkedList,random,Run_2,0.0014043999835848808,0.035298199858516455,0.02021360001526773 +LinkedList,random,Run_3,0.0016908999532461166,0.031882400158792734,0.016559200128540397 +LinkedList,random,Run_4,0.0016268000472337008,0.033718999940901995,0.01765769999474287 +LinkedList,random,Run_5,0.00146949989721179,0.030750300036743283,0.020888000028207898 +HashTable,random,Run_1,0.003547400003299117,7.690000347793102e-05,6.220000796020031e-05 +HashTable,random,Run_2,0.0028089999686926603,6.849993951618671e-05,3.900006413459778e-05 +HashTable,random,Run_3,0.002920700004324317,7.139984518289566e-05,4.020007327198982e-05 +HashTable,random,Run_4,0.003132300218567252,6.630015559494495e-05,3.860006108880043e-05 +HashTable,random,Run_5,0.003326199948787689,0.00011060014367103577,6.16998877376318e-05 +BST,random,Run_1,0.021002399967983365,0.00020360015332698822,0.00011419993825256824 +BST,random,Run_2,0.020290900021791458,0.00019980012439191341,0.00010569998994469643 +BST,random,Run_3,0.019706800114363432,0.00019660010002553463,0.00010689999908208847 +BST,random,Run_4,0.019484999822452664,0.00018949992954730988,0.0001066999975591898 +BST,random,Run_5,0.03975450014695525,0.00024339999072253704,0.00013699987903237343 +LinkedList,sorted,Run_1,0.0015730001032352448,0.03809090005233884,0.01893949997611344 +LinkedList,sorted,Run_2,0.001297699986025691,0.033360299887135625,0.01909619988873601 +LinkedList,sorted,Run_3,0.0015416000969707966,0.03634240012615919,0.016841999953612685 +LinkedList,sorted,Run_4,0.0012899001594632864,0.03580150008201599,0.0170306998770684 +LinkedList,sorted,Run_5,0.0012546998914331198,0.03284729993902147,0.018117799889296293 +HashTable,sorted,Run_1,0.0034030000679194927,8.430005982518196e-05,3.66999302059412e-05 +HashTable,sorted,Run_2,0.002653100062161684,6.769993342459202e-05,3.760005347430706e-05 +HashTable,sorted,Run_3,0.0026434999890625477,6.690016016364098e-05,3.8400059565901756e-05 +HashTable,sorted,Run_4,0.002997299889102578,7.299985736608505e-05,4.179985262453556e-05 +HashTable,sorted,Run_5,0.002777900081127882,8.819997310638428e-05,3.800005652010441e-05 +BST,sorted,Run_1,9.951400500023738,0.07922379998490214,0.0600940000731498 +BST,sorted,Run_2,10.377625699853525,0.08713930007070303,0.05045670014806092 +BST,sorted,Run_3,12.112230099970475,0.08630810002796352,0.050702399807050824 +BST,sorted,Run_4,10.117846999783069,0.0832209000363946,0.05910569988191128 +BST,sorted,Run_5,10.017497000051662,0.07774659991264343,0.041689899982884526