Add processing tools.

.gitignore

@@ -0,0 +1 @@
+__pycache__

loader.py → dataset_loader.py

@@ -4,9 +4,9 @@ from datasets import load_dataset
 version = 'original' # latest version of this dataset
 
 # Load the dataset with version control
-dataset = load_dataset('json', data_files={'train': 'original/train.jsonl',
-                                           'validation': 'original/valid.jsonl',
-                                           'test': 'original/test.jsonl'},
+dataset = load_dataset('json', data_files={'train': f'{version}/train.jsonl',
+                                           'validation': f'{version}/valid.jsonl',
+                                           'test': f'{version}/test.jsonl'},
                        version=version)
 
 # Access the data
@@ -16,7 +16,7 @@ test_dataset = dataset['test']
 
 # Example usage
 for example in train_dataset:
-    print(example['text'])
+    print(example['user'], example['assistant'])
     break
 
 # Get the dataset version

functions/__init__.py

@@ -0,0 +1 @@
+from . import regular

functions/regular.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env python
+
+import pandas as pd
+import re
+
+def process(df: pd.DataFrame) -> pd.DataFrame:
+
+    pattern = r'<\|im_start\|>(\w+)\n(.*?)<\|im_end\|>'
+    
+    def process_text(text):
+        matches = re.findall(pattern, text, re.DOTALL)
+        return dict(matches)
+    
+    result = df['text'].apply(process_text)
+    return pd.DataFrame(result.tolist())
+
+
+def main(file: str = None) -> None:
+    if file != None:
+        df = pd.read_json(file, lines=True)
+        result = process(df)
+        print(result)
+    else:
+        data = {
+            "text": [
+                '<|im_start|>system You are a helpful assistant<|im_end|> <|im_start|>user "阿猛，你觉得做视频最重要的是什么，除了内容好，还有什么能让视频更吸引人呢？"<|im_end|> <|im_start|>assistant 视频做的很好，但美中不足的是。。。时长太短，不！够！看！<|im_end|>',
+                '<|im_start|>system You are a helpful assistant<|im_end|> <|im_start|>user "要如何學會加法"<|im_end|> <|im_start|>assistant 這個問題，我也不太會<|im_end|>',
+            ]
+        }
+        df = pd.DataFrame(data)
+        result = process(df)
+        print(result)
+
+if __name__ == "__main__":
+    import fire
+    fire.Fire(main)

separation.py

@@ -0,0 +1,24 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+import functions
+import pandas as pd
+
+def main(
+    in_file: str,
+    out_file: str,
+    func: str = '',
+):
+    try:
+        # Default using jsonl
+        processor = getattr(functions, func)
+        df = pd.read_json(in_file, lines=True)
+        out = processor.process(df)
+        with open(out_file, 'w', encoding='utf-8') as file:
+            out.to_json(file, force_ascii=False, lines=True, orient='records')
+    except ValueError as e:
+        print(e)
+
+if __name__ == '__main__':
+    import fire
+    fire.Fire(main)