hSSKJr SSKrSSKrSSKJr SSKJrJr SSK J r J r J r S SSjjr \ S- SS 4SS jjr\S :Xa \ "S 5 gg)) annotationsN)Path) IMG_FORMATSimg2label_paths) DATASETS_DIRLOGGERTQDMc [U5n[US35nUS- US- pTURSS9 URSS9 URSS9 UR5Vs/sHofR5(dMUPM nn[ SU55n[ U5SUS3n [ R"S US U S US S SU- S S3 5 UHn XJR- RSS9 XZR- RSS9 [U RS55n [R"U 5 [[ U 5U-5n U SU H3n [R"XU R- U R- 5 M5 XSH3n [R"XU R- U R- 5 M5 M [ R"SUS35 U$s snf)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_okc3h# UH(n[[URS555v M* g7f)*.*N)lenlistglob).0ds P/home/james-whalen/.local/lib/python3.13/site-packages/ultralytics/data/split.py )split_classify_dataset..Ks%DAs4u .//s02z classes, z imagesz Splitting z (z) into z.0%z train, z val...rNzSplit complete in u ✅)rmkdiriterdiris_dirsumrrinfonamerrrandomshuffleintshutilcopy2) source_dir train_ratio source_path split_path train_pathval_pathr class_dirs total_imagesstats class_dir image_files split_idximgs rsplit_classify_datasetr2 sfz"K V,-J%/e1Cd#d# NNDN!)002A2hhj!2JADDDL:z,w ?E KK*[ME7'+c9J(STWbSbcfRggnop nn $++T+: NN "))4)89>>%01 {#K(;67 z *C LL9>>9CHHD E+z*C LL7#((B C+  KK$ZL56 -Bs "G*>G*z coco8/images)g?g?gFc>[U5n[SURS555n[U5n[R "S5 [R "/SQXS9n/SQnUHDnURU- R5(dM'URU- R5 MF [R"SU3SU--5 [[XS5US 9HupU(a6[[[U 5/5S5R5(dMB[!URXh- S S S 9n U R#S U R%UR5R'53S-5 SSS5 M g!,(df  M=f)aJ 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) c3r# UH-oRSSR5[;dM)Uv M/ g7f)rN)suffixlowerr)rxs rrautosplit..ys,W/88AB<3E3E3G;3V11/s(7 7rr)rr)weightsk)zautosplit_train.txtzautosplit_val.txtzautosplit_test.txtzAutosplitting images from z!, using *.txt labeled images only)totalazutf-8)encodingz./ N)rsortedrglobrr seedchoicesparentexistsunlinkrrr ziprstropenwrite relative_toas_posix) pathr:annotated_onlyfilesnindicestxtr7ir1fs r autosplitrUcs<* :D Wdjj/W WE E A KKNnnY=G LC  KK!O # # % % [[1_ $ $ & KK,TF36Y\j6jjks7*!4os3xj&A!&D!E!L!L!N!NdkkCF*C'Ba"S__T[[9BBDEFMNCB5BBs ?F  F __main__ caltech101)g?)r%z str | Pathr&floatreturnr)rMrr:ztuple[float, float, float]rNboolrYNone) __future__rr r#pathlibrultralytics.data.utilsrrultralytics.utilsrrr r2rU__name__rrcsy# ?88Sn.*9 $O $O '$O$O $ON z<(rb