|h ddlZddlZddlmZddlmZmZddlmZm Z ddl m Z m Z m Z ddeeefdedefd Ze d z d d fd edeeeefdeddfdZedk(r edyy)N)Path)TupleUnion) IMG_FORMATSimg2label_paths) DATASETS_DIRLOGGERTQDM source_dir train_ratioreturnc t|}t|d}|dz |dz }}|jd|jd|jd|jDcgc]}|js|}}t d|D}t |d|d} t jd |d | d |d d d|z d d |D]} || jz jd|| jz jdt| jd} tj| tt | |z} | d| D]2} tj| || jz | jz 4| | dD]2} tj| || jz | jz 4t jd|d|Scc}w)u. Split classification dataset into train and val directories in a new directory. Creates a new directory '{source_dir}_split' with train/val subdirectories, preserving the original class structure with an 80/20 split by default. Directory structure: Before: caltech/ ├── class1/ │ ├── img1.jpg │ ├── img2.jpg │ └── ... ├── class2/ │ ├── img1.jpg │ └── ... └── ... After: caltech_split/ ├── train/ │ ├── class1/ │ │ ├── img1.jpg │ │ └── ... │ ├── class2/ │ │ ├── img1.jpg │ │ └── ... │ └── ... └── val/ ├── class1/ │ ├── img2.jpg │ └── ... ├── class2/ │ └── ... └── ... Args: source_dir (str | Path): Path to classification dataset root directory. train_ratio (float): Ratio for train split, between 0 and 1. Returns: (Path): Path to the created split directory. Examples: Split dataset with default 80/20 ratio >>> split_classify_dataset("path/to/caltech") Split with custom ratio >>> split_classify_dataset("path/to/caltech", 0.75) _splittrainvalT)exist_okc3bK|]'}tt|jd)yw)*.*N)lenlistglob).0ds U/var/www/html/test/engine/venv/lib/python3.12/site-packages/ultralytics/data/split.py z)split_classify_dataset..Js"DAs4u ./Ds-/z classes, z imagesz Splitting z (z) into z.0%z train, z val...rNzSplit complete in u ✅)rmkdiriterdiris_dirsumrr infonamerrrandomshuffleintshutilcopy2)r r source_path split_path train_pathval_pathr class_dirs total_imagesstats class_dir image_files split_idximgs rsplit_classify_datasetr3 sfz"K V,-J%/e1CJd#d# NNDN!)002Aahhj!AJADDDL:z,w ?E KK*[ME7'+c9J(STWbSbcfRggnopD inn $++T+: INN "))4)89>>%01 {#K(;67 z * FC LLj9>>9CHHD E Fyz* DC LLh7#((B C DD  KK$ZL56 -Bs ,G,G,z coco8/images)g?g?gFpathweightsannotated_onlyct|}td|jdD}t|}t j dt j gd||}gd}|D]=}|j|z js!|j|z j?tjd|d|zztt||| D]\}} |r0ttt| gdjs8t!|j||z d d 5} | j#d | j%|jj'dzdddy#1swYxYw)az Automatically split a dataset into train/val/test splits and save the resulting splits into autosplit_*.txt files. Args: path (Path): Path to images directory. weights (tuple): Train, validation, and test split fractions. annotated_only (bool): If True, only images with an associated txt file are used. Examples: Split images with default weights >>> from ultralytics.data.split import autosplit >>> autosplit() Split with custom weights and annotated images only >>> autosplit(path="path/to/images", weights=(0.8, 0.15, 0.05), annotated_only=True) c3hK|]*}|jddjtvs'|,yw)rN)suffixlowerr)rxs rrzautosplit..xs*W188AB<3E3E3G;3V1Ws(22rr)rr)r5k)zautosplit_train.txtzautosplit_val.txtzautosplit_test.txtzAutosplitting images from z!, using *.txt labeled images only)totalazutf-8)encodingz./ N)rsortedrglobrr#seedchoicesparentexistsunlinkr r!r ziprstropenwrite relative_toas_posix) r4r5r6filesnindicestxtr;ir2fs r autosplitrUbsY* :D Wdjj/W WE E A KKNnnY1=G LC ' KK!O # # % [[1_ $ $ &' KK,TF36Y\j6jjks7E*!4O3os3xj&A!&D!E!L!L!NdkkCF*C'B Oa"S__T[[9BBDEFMN O OO O Os 8?FF __main__ caltech101)g?)r#r&pathlibrtypingrrultralytics.data.utilsrrultralytics.utilsrr r rJfloatr3boolrU__name__rras ?88SuS$Y'7SeSVZSn.*9 $O $O 5%& '$O$O $ON z<(r`