o
    –hˆ  ã                   @   s@   d Z ddlZddlZddlmZ dZdd„ Zdd„ Zd	d
„ ZdS )z8
Utilities for the processing of constituency treebanks
é    N)Ú
parse_tree)ÚtrainÚdevÚtestc                 C   sT   t  tj | d| ¡tj | d| ¡¡ t  tj | d| ¡tj | d| ¡¡ d S )Nz
%s_dev.mrgz%s_test.mrg)ÚshutilÚcopy2ÚosÚpathÚjoin)Ú	base_pathÚinput_datasetÚoutput_dataset© r   úc/var/www/html/env_mimamsha/lib/python3.10/site-packages/stanza/utils/datasets/constituency/utils.pyÚcopy_dev_test   s   ÿÿr   c                 C   sP   t | tƒD ] \}}tj |d||f ¡}td t|ƒ|¡ƒ tj	 
||¡ qd S )Nz	%s_%s.mrgzWriting {} trees to {})ÚzipÚSHARDSr   r	   r
   ÚprintÚformatÚlenr   ÚTreeÚwrite_treebank)ÚdatasetsÚ
output_dirÚdataset_nameÚdatasetÚshardÚoutput_filenamer   r   r   Úwrite_dataset   s
   ýr   c                 C   sF   t t| ƒ| ƒ}t t| ƒ||  ƒ}| d|… | ||… | |d… fS )z,
    Split a treebank deterministically
    N)Úintr   )ÚtreebankÚ
train_sizeÚdev_sizeÚ	train_endÚdev_endr   r   r   Úsplit_treebank   s   "r%   )	Ú__doc__r   r   Ústanza.models.constituencyr   r   r   r   r%   r   r   r   r   Ú<module>   s    