o
    پi                     @   s>   d dl mZ d dlmZ d dlZd dlmZ G dd dZdS )    )SimpleNamespace)OptionalN)run_evalc                   @   sF   e Zd ZU eed< dZee ed< dZeed< dZ	eed< dd	 Z
dS )

GSM8KMixingsm8k_accuracy_thresNgsm8k_accept_length_thres   gsm8k_num_questions   gsm8k_parallelc                 C   s   t | jd  tdd | jd| jdt| jdd d}t|}t	d| | 
|d	 | j | jd urXt | jd
 }| d d d }t	d| | || j d S d S )Nz/flush_cache   i   zhttp://127.0.0.1:)	num_shots	data_pathnum_questionsmax_new_tokensparallelhostportzmetrics=accuracyz/server_infointernal_statesr   avg_spec_accept_lengthzavg_spec_accept_length=)requestsgetbase_urlr   r	   r   intsplitrun_eval_gsm8kprintassertGreaterEqualr   r   jsonassertGreater)selfargsmetricsserver_infor    r'   W/home/ubuntu/.local/lib/python3.10/site-packages/sglang/test/kits/gsm8k_accuracy_kit.py
test_gsm8k   s*   	
zGSM8KMixin.test_gsm8k)__name__
__module____qualname__float__annotations__r   r   r	   r   r   r)   r'   r'   r'   r(   r   	   s   
 r   )	typesr   typingr   r   sglang.test.few_shot_gsm8kr   r   r   r'   r'   r'   r(   <module>   s
    