fix param definition (#424)

* fix param def * add param check
modelscope · Sep 11, 2024 · 8ab74a3 · 8ab74a3
1 parent da1acc9
commit 8ab74a3
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 4 deletions.
diff --git a/data_juicer/ops/mapper/extract_qa_mapper.py b/data_juicer/ops/mapper/extract_qa_mapper.py
@@ -41,7 +41,7 @@ class ExtractQAMapper(Mapper):
 
     def __init__(self,
                  hf_model: str = 'alibaba-pai/pai-qwen1_5-7b-doc2qa',
-                 trust_remote_code=False,
+                 trust_remote_code: bool = False,
                  pattern: str = None,
                  qa_format: str = 'chatml',
                  enable_vllm: bool = True,

diff --git a/data_juicer/ops/mapper/generate_instruction_mapper.py b/data_juicer/ops/mapper/generate_instruction_mapper.py
@@ -51,9 +51,9 @@ class GenerateInstructionMapper(Mapper):
     _accelerator = 'cuda'
 
     def __init__(self,
-                 hf_model,
-                 seed_file,
-                 instruct_num,
+                 hf_model: str = 'Qwen/Qwen-7B-Chat',
+                 seed_file: str = None,
+                 instruct_num: int = 3,
                  trust_remote_code: bool = False,
                  similarity_threshold: float = 0.7,
                  prompt_template: str = None,
@@ -111,6 +111,10 @@ def __init__(self,
         super().__init__(*args, **kwargs)
         self.num_proc = 1
 
+        if not seed_file:
+            raise ValueError('Please provide `seed_file` parameter, a file in chatml format. '\
+                'Reference data: data-juicer/demos/data/demo-dataset-chatml.jsonl ')
+
         self.instruct_num = instruct_num
         self.similarity_threshold = similarity_threshold
         self.similarity_type = 'rouge_l'