o V h%1@sdZddlZddlZddlZddlZddlZddlZddlm Z m Z ddl m Z m Z mZmZmZmZmZddlmZGdddZdS)a5 Module provides functionalities for hyperparameter tuning of the Ultralytics YOLO models for object detection, instance segmentation, image classification, pose estimation, and multi-object tracking. Hyperparameter tuning is the process of systematically searching for the optimal set of hyperparameters that yield the best model performance. This is particularly crucial in deep learning models like YOLO, where small changes in hyperparameters can lead to significant differences in model accuracy and efficiency. Examples: Tune hyperparameters for YOLO11n on COCO8 at imgsz=640 and epochs=30 for 300 tuning iterations. >>> from ultralytics import YOLO >>> model = YOLO("yolo11n.pt") >>> model.tune(data="coco8.yaml", epochs=10, iterations=300, optimizer="AdamW", plots=False, save=False, val=False) N)get_cfg get_save_dir) DEFAULT_CFGLOGGER callbackscolorstrremove_colorstr yaml_print yaml_save)plot_tune_resultsc@s2eZdZdZedfddZdd d Zdd dZdS)Tunerap A class for hyperparameter tuning of YOLO models. The class evolves YOLO model hyperparameters over a given number of iterations by mutating them according to the search space and retraining the model to evaluate their performance. Attributes: space (dict): Hyperparameter search space containing bounds and scaling factors for mutation. tune_dir (Path): Directory where evolution logs and results will be saved. tune_csv (Path): Path to the CSV file where evolution logs are saved. args (dict): Configuration arguments for the tuning process. callbacks (list): Callback functions to be executed during tuning. prefix (str): Prefix string for logging messages. Methods: _mutate: Mutates the given hyperparameters within the specified bounds. __call__: Executes the hyperparameter evolution across multiple iterations. Examples: Tune hyperparameters for YOLO11n on COCO8 at imgsz=640 and epochs=30 for 300 tuning iterations. >>> from ultralytics import YOLO >>> model = YOLO("yolo11n.pt") >>> model.tune( ... data="coco8.yaml", epochs=10, iterations=300, optimizer="AdamW", plots=False, save=False, val=False ... ) Tune with custom search space. >>> model.tune(space={key1: val1, key2: val2}) # custom search space dictionary NcCs$|ddpCiddddddd d d d d dddddddddddddddddddddd d d!d!d!d!d!d!d"|_t|d#|_|jj|j_t|j|jjpYd$d%|_d&\|j_|j_|j_|jd'|_ |pst |_ t d(|_ t |t|j d)|jd*|j d+dS),z Initialize the Tuner with configurations. Args: args (dict): Configuration for hyperparameter evolution. _callbacks (list, optional): Callback functions to be executed during tuning. spaceNlr0)gh㈵>皙?lrf)g-C6?rmomentum)gffffff?g\(\?333333? weight_decay)gMbP? warmup_epochs)rg@warmup_momentum)rgffffff?box)?g4@cls)皙?g@dfl)g?g@hsv_h)rrhsv_s)rg?hsv_vdegrees)rgF@ translatescaleshear)rg$@ perspective)rr)flipudfliplrbgrmosaicmixup copy_paste) overridestune)name)NFFztune_results.csvzTuner: z*Initialized Tuner instance with 'tune_dir=z' uT💡 Learn about tuning at https://docs.ultralytics.com/guides/hyperparameter-tuning)popr rargsresumeexist_okrr,tune_dirtune_csvrget_default_callbacksrprefixadd_integration_callbacksrinfo)selfr. _callbacksr9L/var/www/vscode/kcb/lib/python3.10/site-packages/ultralytics/engine/tuner.py__init__=sn          zTuner.__init__single皙?rc sjrtjjdddddddf}t|t}t| d|dddfdddfd}|dksHtdkrVtjt ||d dn|d kri| |d d| tj}| t ttd d jD}tj} t| tdkr||| |k|| ||dd dtdksfddtjD} n fddjD} jD]#\} t| | d| | <t| | d| | <t| | d| | <q| S)a Mutate hyperparameters based on bounds and scaling factors specified in `self.space`. Args: parent (str): Parent selection method: 'single' or 'weighted'. n (int): Number of parents to consider. mutation (float): Probability of a parameter mutation in any given iteration. sigma (float): Standard deviation for Gaussian random number generator. Returns: (dict): A dictionary containing mutated hyperparameters. ,ndmin delimiterskiprowsNrgư>r<)weightsweightedcSs$g|]}t|dkr|dndqS)r?r)len).0vr9r9r: s$z!Tuner._mutate..rg@cs*i|]\}}|t|d|qSrAfloatrJik)rKxr9r: s*z!Tuner._mutate..csi|] }|tj|qSr9)getattrr.rJrR)r7r9r:rTsr=)r2existsnploadtxtminrIargsortrandomchoicesrangereshapesumseedinttimearrayr valuesonesallrandnclip enumeratekeysitemsmaxround) r7parentnmutationsigmafitnesswrgnghyprRr9)r7rKrSr:_mutatels2 (   4 z Tuner._mutate Tcst}d\}}|jdjdddd}|jr:tj|jdddd jd}t |j d |jd |dd t ||D]}| t |j d |dd|di} it |j} tt| } | d} zs z!Tuner.__call__..)checkzbest.ptzlast.pt train_metricsztraining failedz5training failure for hyperparameter tuning iteration  Nrsrr=csg|]}|qSr9r9rV) mutated_hypr9r:rLsz"Tuner.__call__..azutf-8)encodingcSsi|] \}}|t|dqS)r=)rnrr9r9r:rTsz"Tuner.__call__..z*.pt) ignore_errorsu iterations complete ✅ (z.2fzs) zResults saved to boldz Best fitness=z observed at iteration zBest fitness metrics are zBest fitness model is z0Best fitness hyperparameters are printed below. cs&i|]\}}|t|dfqSrMrNrP)best_idxrSr9r:rTs&zbest_hyperparameters.yamlz# )dataheader)1rcr1mkdirr2rWrXrYshaperr6r4r^ryvarsr.rr __import__ executablerl subprocessrun returncodetorchload Exceptionerrorgetrnr rkjoinlistopenwritemapstrargmaxglobshutilcopy2rmtreer rrjr rreplacer )r7model iterationscleanupt0 best_save_dir best_metricsstartrQmetrics train_argssave_dir weights_dirlaunchcmd return_code ckpt_fileerslog_rowheadersfbest_is_currentckptrrr9)rrrSr:__call__s  $& & $,  & zTuner.__call__)r<r=r>r)NrzT)__name__ __module__ __qualname____doc__rr;ryrr9r9r9r:r s  /.r )rr\rrrcnumpyrXrultralytics.cfgrrultralytics.utilsrrrrrr r ultralytics.utils.plottingr r r9r9r9r:s$