fix(dataset): 修正数据集输出路径和名称

chinese-c4.py

@@ -18,7 +18,7 @@ if __name__ == "__main__":
         end_idx = min((i + 1) * chunk_size, total_samples)
         chunk = train_dataset.select(range(start_idx, end_idx))
 
-        output_path = f"{output_dir}/{output_dir}_text_chunk_{i}.jsonl"
+        output_path = f"{output_dir}/chinese-c4_text_chunk_{i}.jsonl"
         with open(output_path, "w", encoding="utf-8") as f:
             for example in chunk:
                 # 每行写入一个 {"text": "xxx"} 对象

chinese-wiki.py

@@ -10,7 +10,7 @@ if __name__ == "__main__":
     total_samples = len(train_dataset)
     num_chunks = (total_samples // chunk_size) + 1
     script_dir = os.path.dirname(os.path.abspath(__file__))
-    output_dir = os.path(script_dir, "dataset", "chinese-c4")
+    output_dir = os.path.join(script_dir, "dataset", "chinese-wiki")
     os.makedirs(output_dir, exist_ok=True)
     
     
@@ -19,7 +19,7 @@ if __name__ == "__main__":
         end_idx = min((i + 1) * chunk_size, total_samples)
         chunk = train_dataset.select(range(start_idx, end_idx))
 
-        output_path = f"{output_dir}/{output_dir}_text_chunk_{i}.jsonl"
+        output_path = f"{output_dir}/chinese-wiki_text_chunk_{i}.jsonl"
         with open(output_path, "w", encoding="utf-8") as f:
             for example in chunk:
                 json_line = {"text": example["text"]}

english-fineweb.py

@@ -33,7 +33,7 @@ if __name__ == "__main__":
         end_idx = min((i + 1) * chunk_size, total_samples)
         chunk = train_dataset.select(range(start_idx, end_idx))
         
-        output_path = f"{output_dir}/{output_dir}text_chunk_{i}.jsonl"
+        output_path = f"{output_dir}/english-fineweb_text_chunk_{i}.jsonl"
         with open(output_path, "w", encoding="utf-8") as f:
             for example in chunk:
                 json_line = {"text": comprehensive_normalization(example["text"])}