Update multi QA generation process

Each thread stores the generated content independently, and finally integrates the generated content into a file

scripts/qa_generation/main.py

@@ -4,14 +4,46 @@ import time
 from tqdm import tqdm
 import concurrent.futures
 from datetime import datetime
+import numpy as np
 
 from config.config import result_dir, storage_interval, window_size, overlap_size, multi_process_num
 from model.qwen import call_qwen_single_turn
 from util.logger import get_logger
-from util.data_loader import get_file_list, get_txt_content, capture_qa
+from util.data_loader import get_file_list, get_txt_content, capture_qa, merge_sub_qa_generation, save_to_file
 
 logger = get_logger()
 
+"""
+每个线程产生 QA 对以及存储到子文件中
+"""
+def single_thread_generate(thread_num, interval, model_caller, storage_jsonl_path, contents):
+
+    storage_counter = 0
+    storage_list = []
+
+    for content in tqdm(contents):
+        try:
+            response = model_caller(content)
+            captured_qa = capture_qa(response)
+
+            if captured_qa is None:
+                continue
+
+            storage_list.extend(captured_qa)
+            storage_counter += 1
+
+            if storage_counter % interval == 0:
+                save_to_file(storage_jsonl_path, storage_list)
+                storage_counter = 0
+                storage_list = []
+        except Exception as exc:
+            logger.error("QA generation error : %s" % (exc))
+
+    if storage_list:
+        save_to_file(storage_jsonl_path, storage_list)
+        storage_list = []
+   
+ 
 """
 生成 QA 对
 model_name: 可调用的模型名称，暂时只实现了 qwen
@@ -45,43 +77,29 @@ def generate_qa(
             result_dir, f'{current_time}-{file_name}-{model_name}.jsonl')
         logger.info(f'The generated QA will be stored in {storage_jsonl_path}.')
 
-        response_list = []
+        # 基于并发个数切分 contents 内容
-        logger.info("Begin generate QA data")
+        contents_array = np.array(contents)
-        with tqdm(total=len(contents)) as pbar:
+        chunks = np.array_split(contents_array, multi_process_num)
-            with concurrent.futures.ThreadPoolExecutor(max_workers=multi_process_num) as executor:
-                # 创建一个Future列表，它们将对应每个worker_function的结果
-                futures = {executor.submit(model_caller, content): content for content in contents}
 
-                for future in concurrent.futures.as_completed(futures):
+        # 构建并发参数 list
-                    try:
+        parameters_list = list()
-                        response_list.append(future.result())
+        for thread_num, chunk in enumerate(chunks):
-                        pbar.update(1)
+            parameters_list.append(
-                    except Exception as exc:
+                    [thread_num, interval, model_caller, storage_jsonl_path + f'-{thread_num}', list(chunk)]
-                        logger.error("Item generated an exception: %s" % (exc))
+            )
 
-        logger.info("Begin capture LLM response")
+        # 并发生成 QA 对
-        for response in tqdm(response_list):
+        with concurrent.futures.ThreadPoolExecutor(max_workers=multi_process_num) as executor:
-            captured_qa = capture_qa(response)
+            # 创建一个Future列表，它们将对应每个worker_function的结果
-            # print(captured_qa) # 打印QA对
+            futures = [executor.submit(single_thread_generate, *parameters) for parameters in parameters_list]
-            if captured_qa is None:
-                continue
         
-            storage_list.extend(captured_qa)
+            for future in concurrent.futures.as_completed(futures):
-            storage_counter += 1
+                try:
-            if storage_counter % interval == 0:
+                    future.result()
-                storage_counter = 0
+                except Exception as exc:
-                with open(storage_jsonl_path, 'a', encoding='utf-8') as f:
+                    logger.error("Thread generated an exception: %s" % (exc))
-                    for item in storage_list:
-                        f.write(json.dumps(item, ensure_ascii=False) + '\n')
-                    storage_list = []
-    
-        # 如果有剩余，存入
-        if storage_list:
-            with open(storage_jsonl_path, 'a', encoding='utf-8') as f:
-                for item in storage_list:
-                    f.write(json.dumps(item, ensure_ascii=False) + '\n')
-                storage_list = []
 
+        merge_sub_qa_generation(result_dir, storage_jsonl_path)
 
 if __name__ == '__main__':
 

scripts/qa_generation/util/data_loader.py

@@ -1,6 +1,7 @@
 import os
 import re
 import json
+import glob
 from typing import List, Dict
 
 from config.config import data_dir
@@ -78,3 +79,28 @@ def capture_qa(content: str) -> List[Dict]:
     else:
         logger.warning("No JSON block found.")
         return None
+
+"""
+将 storage_list 存入到 storage_jsonl_path
+"""
+def save_to_file(storage_jsonl_path, storage_list):
+    with open(storage_jsonl_path, 'a', encoding='utf-8') as f:
+        for item in storage_list:
+            f.write(json.dumps(item, ensure_ascii=False) + '\n')
+
+"""
+将并发产生的文件合并成为一个文件
+"""
+def merge_sub_qa_generation(directory, storage_jsonl_path):
+
+    # 查找以指定前缀开始的所有文件
+    matching_files = glob.glob(os.path.join(directory, storage_jsonl_path + "*"))
+    
+    file_contents = []
+    for file_path in matching_files:
+        with open(file_path, 'r', encoding='utf-8') as f:
+            for line in f:
+                file_contents.append(json.loads(line))
+            os.remove(file_path)
+    save_to_file(storage_jsonl_path, file_contents)
+