methods/Hardtest/count.py

from config import cfg
import sys
sys.path.append("/home/i-luoxianzhen/data/TestCase-Gen/methods/utils")
from dataset_all import get_datasets_by_name

def see_all():
    all_count = 0
    for batch in range(9):
        al_dataset = get_datasets_by_name(cfg.dataset_name)
        start_pos = (batch) * 100
        end_pos = (batch + 1) * 100 if (batch + 1) * 100 <= len(al_dataset) else len(al_dataset)
        al_dataset = al_dataset[start_pos: end_pos]
        import os
        total_count = 0
        for item in al_dataset:
            if os.path.exists(cfg.tests_path.format(item['tcb_id'])):
                total_count += 1
                all_count += 1

        print(total_count)

    print(f"共计：{all_count}")

def find_index(batch, tcb_id):
    al_dataset = get_datasets_by_name(cfg.dataset_name)
    print(f"数据集大小{len(al_dataset)}")
    # start_pos = (batch) * 100
    # end_pos = (batch + 1) * 100 if (batch + 1) * 100 <= len(al_dataset) else len(al_dataset)
    # al_dataset = al_dataset[start_pos: end_pos]
    import os
    total_count = 0
    for item in al_dataset:
        if tcb_id == item['tcb_id']:
            break
        total_count += 1
    print(total_count)

print("执行进度")
see_all()


def find_unexist():
    all_count = 0
    for batch in range(8):
        al_dataset = get_datasets_by_name(cfg.dataset_name)
        start_pos = (batch) * 100 
        end_pos = (batch + 1) * 100 if (batch + 1) * 100 <= len(al_dataset) else len(al_dataset)
        al_dataset = al_dataset[start_pos: end_pos]
        import os
        total_count = 0
        for item in al_dataset:
            if not os.path.exists(cfg.tests_path.format(item['tcb_id'])):
                print(item['tcb_id'])
                all_count += 1

    print(f"共计：{all_count}")     

tcb = "数组游戏"
print(f"寻找idx {tcb}")
find_index(2, tcb)
# find_unexist()

# import os
# import json

# def calculate_average_samples(directory):
#     """
#     读取指定目录下所有的jsonl文件，计算平均的样本个数。
    
#     参数：
#     directory: 文件夹路径，包含 .jsonl 文件。
    
#     返回：
#     float: 所有文件中的平均样本个数。
#     """
#     sample_counts = []
#     # 遍历目录中的所有文件
#     for filename in os.listdir(directory):
#         if filename.endswith('.jsonl') and "test_pass_rate" not in filename:
#             file_path = os.path.join(directory, filename)
            
#             # 读取每个jsonl文件并统计样本个数
#             with open(file_path, 'r', encoding='utf-8') as file:
#                 lines = file.readlines()
#                 sample_counts.append(len(lines))  # 每个文件的行数即为样本数
    
#     # 计算并返回平均值
#     if sample_counts:
#         print(f"平均testcase个数: {sum(sample_counts) / 731}")
#         print(f"平均testcase个数(不计入生成0个的题目): {sum(sample_counts) / len(sample_counts)}")
#         return sum(sample_counts) / 731
#     else:
#         return 0.0

# # 示例用法
# directory = '/home/i-luoxianzhen/data/TestCase-Gen/saved_tests/predo/'  # 替换为你的文件夹路径
# average_samples = calculate_average_samples(directory)
# # print(f"平均样本个数: {average_samples}")

# def calculate_average_pass(directory):
#     all_gen = 0
#     right = 0
#     # 遍历目录中的所有文件
#     for filename in os.listdir(directory):
#         if filename.endswith('.json') and "test_pass_rate" in filename:
#             file_path = os.path.join(directory, filename)
            
#             # 读取每个jsonl文件并统计样本个数
#             with open(file_path, 'r', encoding='utf-8') as file:
#                 for line in file.readlines():
#                     item = json.loads(line)
#                     all_gen += item['gen_nums']
#                     right += item['right_nums']
#     # 计算并返回平均值

#     return right / all_gen

# print(f"pass rate {calculate_average_pass(directory)}")