CoolFace
Apppublic

AMD21/codefixerenv

sourceHugging Faceupdated 6mo agoView on Hugging Face
0likes
rootenv__pycache__environment.cpython-310.pyc
environment.cpython-310.pyc38 linesDownload Raw Back to __pycache__
1o

2e��ip�@s`ddlZddlZej�dej�ej�e���ddlmZmZm	Z	ddl3mZGdd�d�ZdS)�N)�Observation�Action�Reward)�TASKSc@sdeZdZUdZeed<ddd�Zddedefd	d4�Z	de5deeee
effdd
�Zdefdd�ZdS)�CodeFixerEnv��	MAX_STEPS�returnNcCs4d|_d|_d|_g|_d|_d|_d|_d|_dS)NrF�)�_task�_grader�_difficulty�_history�_step_number�_done�_best_score�_cumulative_reward��self�r�5C:\Users\swaro\Downloads\CodeFixer\env\environment.py�__init__s6zCodeFixerEnv.__init__�easy�7difficultycCs||tvrtd|�dtt�������t|\}}||_||_||_g|_d|_d|_	d|_8d|_t|d|dgd|j
d�S)	NzUnknown difficulty 'z'. Valid choices: rFr9�10buggy_code�context��inputr�history�step_number�	max_steps)r�11ValueError�sorted�keysrrr
rrrrrrr)rrZtask_defZgraderrrr�resets(��zCodeFixerEnv.reset�actioncCsZ|jrtd��|jdurtd��|jd7_|jd|j|jd�}d}d}d}d}d}d}|jdkrAd	}|}d12|_d}d|d
<n||jdkrWd}|}d|jd��}d|d
<nf|jdkr�|�|j�\}}	|	|d<||d<||j}|dkr{|}||_nd}|}|dkr�|jdkr�d}||7}d13|d<|dkr�d14|_d}d|d
<n|	�	dd�}15|	�	dd�}d |d!�d"|16�d#|�d$�}d%|d
<t17|d&�}t18|j|d&�|_|j|jkr�|js�d19|_|�	d
d'�|d
<t
|||dur�t20|d&�nd|||d(�}|j�|j|jt|j�d)k�r21|jdd)�d*n|j||d+��t|jd,|jdt|j�|j|jd-�}
|
||j|fS).Nz5Episode is over. Call reset() to start a new episode.z0Environment not initialised. Call reset() first.��id)�task_idr�stepr22�Zgive_upg��������TzAgent chose to give up.Zgave_upZoutcomeZexplaing��������z4Hint: re-read the loop bounds carefully. Task goal: rZ	explainedZfix�grade_detail�grader_scorerg�?�g�������?�efficiency_bonusu&All test cases passed — perfect fix!�solvedZtests_passed�?Ztests_totalzScore z.2fu — �/z tests passed. Keep refining.�partial��timeout)�valuer,�improvementr.�penalty�reason�xz...)r)�action_type�action_content�reward�feedbackrr)r�RuntimeErrorrrr
�typer�contentr�get�roundrrrr�append�lenr�list)rr%�infoZ	raw_valuer,r6r.r7r=r+Zpassed�totalr<Znext_obsrrrr)-s�23�242526��2728293031���32�	���zCodeFixerEnv.stepc	Cs8|jr|jdnd|j|j|j|j|j|jt|j�d�S)Nr')r(rrr �doneZ33best_scoreZcumulative_rewardZhistory_length)	rr
rrrrrrDrrrrr�state�s�zCodeFixerEnv.state)r	N)r)�__name__�34__module__�__qualname__r�int�__annotations__r�strrr$r�tupler�bool�dictr)rIrrrrrs353637hr)
�os�sys�path�insert�dirname�__file__�38env.modelsrrr�tasks.tasksrrrrrr�<module>s