U 0-eeo@sddlZddlZddlZddlmZddlmZmZmZm Z m Z ddl Z ddl m Z ddlmZddlmZmZmZmZmZmZmZgZereejereejereejereejereeje eZ dd Z!d d Z"Gd d d Z#Gddde#Z$Gddde#Z%Gddde#Z&Gddde#Z'Gddde#Z(e'e&e(e$e%dZ)dee e*ee#fe e*ej+fdddZ,dS)Nwraps)AnyDictListOptionalUnion) get_logger) PartialState) LoggerTypeis_aim_availableis_comet_ml_availableis_mlflow_availableis_tensorboard_availableis_wandb_availablelistifycstfdd}|S)a Decorator to selectively run the decorated function on the main process only based on the `main_process_only` attribute in a class. Checks at function execution rather than initialization time, not triggering the initialization of the `PartialState`. cs8t|ddr$t|f||S|f||SdS)Nmain_process_onlyF)getattrr on_main_process)selfargskwargsfunctionT/var/www/html/Darija-Ai-Train/env/lib/python3.8/site-packages/accelerate/tracking.pyexecute_on_main_processDs z0on_main_process..execute_on_main_processr)rrrrrr;s rcCstS)z@Returns a list of all supported available trackers in the system)_available_trackersrrrrget_available_trackersNsrc@sHeZdZdZdZdddZedddZeee d d d Z d d Z dS)GeneralTrackera` A base Tracker class to be used for all logging integration implementations. Each function should take in `**kwargs` that will automatically be passed in from a base dictionary provided to [`Accelerator`]. Should implement `name`, `requires_logging_directory`, and `tracker` properties such that: `name` (`str`): String representation of the tracker class name, such as "TensorBoard" `requires_logging_directory` (`bool`): Whether the logger requires a directory to store their logs. `tracker` (`object`): Should return internal tracking mechanism used by a tracker class (such as the `run` for wandb) Implementations can also include a `main_process_only` (`bool`) attribute to toggle if relevent logging, init, and other functions should occur on the main process or across all processes (by default will use `True`) TFcCs|sd}t|ds|d7}t|ds@t|dkr8|d7}|d7}dt|krht|dkr`|d7}|d 7}t|dkrtd |dS) Nnamez`name`requires_logging_directoryrz, z`requires_logging_directory`trackerz `tracker`zThe implementation for this tracker class is missing the following required attributes. Please define them in the class definition: )hasattrlendirNotImplementedError)rZ_blankerrrrr__init__fs       zGeneralTracker.__init__valuescCsdS)a Logs `values` as hyperparameters for the run. Implementations should use the experiment configuration functionality of a tracking API. Args: values (Dictionary `str` to `bool`, `str`, `float` or `int`): Values to be stored as initial hyperparameters as key-value pairs. The values need to have type `bool`, `str`, `float`, `int`, or `None`. Nrrr,rrrstore_init_configuration|s z'GeneralTracker.store_init_configurationr,stepcKsdS)a Logs `values` to the current run. Base `log` implementations of a tracking API should go in here, along with special behavior for the `step parameter. Args: values (Dictionary `str` to `str`, `float`, or `int`): Values to be logged as key-value pairs. The values need to have type `str`, `float`, or `int`. step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. Nrrr,r0rrrrlogs zGeneralTracker.logcCsdS)z Should run any finalizing functions within the tracking API. If the API should not have one, just don't overwrite that method. NrrrrrfinishszGeneralTracker.finishN)F) __name__ __module__ __qualname____doc__rr*dictr.rintr2r4rrrrr Ss    r cseZdZdZdZdZeeeee j fdfdd Z e ddZ eed d d Zedeeed ddZeeeed ddZeddZZS)TensorBoardTrackera A `Tracker` class that supports `tensorboard`. Should be initialized at the start of your script. Args: run_name (`str`): The name of the experiment run logging_dir (`str`, `os.PathLike`): Location for TensorBoard logs to be stored. kwargs: Additional key word arguments passed along to the `tensorboard.SummaryWriter.__init__` method. tensorboardTrun_name logging_dirc szddlm}Wntk r,ddl}YnXt||_tj |||_ |j |j f||_ t d|jd|j t ddS)Nr)r<z Initialized TensorBoard project z logging to aMake sure to log any initial configurations with `self.store_init_configuration` before training!)Z torch.utilsr<ModuleNotFoundErrorZ tensorboardXsuperr*r>ospathjoinr?Z SummaryWriterwriterloggerdebug)rr>r?rr< __class__rrr*s zTensorBoardTracker.__init__cCs|jSNrFr3rrrr$szTensorBoardTracker.trackerr+c Cs|jj|id|jt}tj|jt|}tj |ddt tj|dd<}zt ||Wn$t j jk rtdYnXW5QRXtddS) a Logs `values` as hyperparameters for the run. Should be run at the beginning of your experiment. Stores the hyperparameters in a yaml file for future use. Args: values (Dictionary `str` to `bool`, `str`, `float` or `int`): Values to be stored as initial hyperparameters as key-value pairs. The values need to have type `bool`, `str`, `float`, `int`, or `None`. )Z metric_dictT)exist_okz hparams.ymlwz-Serialization to store hyperparameters failedzQStored initial configuration hyperparameters to TensorBoard and hparams yaml fileN)rFZ add_hparamsflushtimerCrDrEr?strmakedirsopenyamldumpZ representerZRepresenterErrorrGerrorrH)rr,Zproject_run_namedir_nameoutfilerrrr.s   z+TensorBoardTracker.store_init_configurationNr/cKst|}|D]|\}}t|ttfrB|jj||fd|i|qt|trh|jj||fd|i|qt|t r|jj ||fd|i|q|j t ddS)a Logs `values` to the current run. Args: values (Dictionary `str` to `str`, `float`, `int` or `dict` of `str` to `float`/`int`): Values to be logged as key-value pairs. The values need to have type `str`, `float`, `int` or `dict` of `str` to `float`/`int`. step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. kwargs: Additional key word arguments passed along to either `SummaryWriter.add_scaler`, `SummaryWriter.add_text`, or `SummaryWriter.add_scalers` method based on the contents of `values`. global_stepz"Successfully logged to TensorBoardN)ritems isinstancer:floatrFZ add_scalarrQadd_textr9Z add_scalarsrOrGrHrr,r0rkvrrrr2s   zTensorBoardTracker.logcKs:|D]"\}}|jj||fd|i|qtddS)a Logs `images` to the current run. Args: values (Dictionary `str` to `List` of `np.ndarray` or `PIL.Image`): Values to be logged as key-value pairs. The values need to have type `List` of `np.ndarray` or step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. kwargs: Additional key word arguments passed along to the `SummaryWriter.add_image` method. rYz)Successfully logged images to TensorBoardN)rZrFZ add_imagesrGrHr^rrr log_imagess zTensorBoardTracker.log_imagescCs|jtddS)z- Closes `TensorBoard` writer zTensorBoard writer closedN)rFcloserGrHr3rrrr4s zTensorBoardTracker.finish)N)r5r6r7r8r"r#rrQrrCPathLiker*propertyr$r9r.rr:r2rar4 __classcell__rrrIrr;s   r;cseZdZdZdZdZdZeedfdd Z e ddZ ee d d d Z ede eed ddZede eed ddZedeeeeeeeeedddZeddZZS) WandBTrackera A `Tracker` class that supports `wandb`. Should be initialized at the start of your script. Args: run_name (`str`): The name of the experiment run. kwargs: Additional key word arguments passed along to the `wandb.init` method. wandbFr>c sPt||_ddl}|jfd|ji||_td|jtddS)NrprojectzInitialized WandB project r@)rBr*r>rginitrunrGrH)rr>rrgrIrrr*s zWandBTracker.__init__cCs|jSrK)rkr3rrrr$(szWandBTracker.trackerr+cCs&ddl}|jj|ddtddS)u Logs `values` as hyperparameters for the run. Should be run at the beginning of your experiment. Args: values (Dictionary `str` to `bool`, `str`, `float` or `int`): Values to be stored as initial hyperparameters as key-value pairs. The values need to have type `bool`, `str`, `float`, `int`, or `None`. rNT)Zallow_val_changez5Stored initial configuration hyperparameters to WandB)rgconfigupdaterGrH)rr,rgrrrr.,s z%WandBTracker.store_init_configurationNr/cKs&|jj|fd|i|tddS)a, Logs `values` to the current run. Args: values (Dictionary `str` to `str`, `float`, `int` or `dict` of `str` to `float`/`int`): Values to be logged as key-value pairs. The values need to have type `str`, `float`, `int` or `dict` of `str` to `float`/`int`. step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. kwargs: Additional key word arguments passed along to the `wandb.log` method. r0zSuccessfully logged to WandBN)rkr2rGrHr1rrrr2;szWandBTracker.logc sPddl|D]0\}}|j|fdd|Difd|i|qtddS)a Logs `images` to the current run. Args: values (Dictionary `str` to `List` of `np.ndarray` or `PIL.Image`): Values to be logged as key-value pairs. The values need to have type `List` of `np.ndarray` or step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. kwargs: Additional key word arguments passed along to the `wandb.log` method. rNcsg|]}|qSr)ZImage).0imagergrr \sz+WandBTracker.log_images..r0z#Successfully logged images to WandB)rgrZr2rGrHr^rrqrraLs *zWandBTracker.log_images) table_namecolumnsdata dataframer0c Ks6ddl}||j|||di}|j|fd|i|dS)a Log a Table containing any object type (text, image, audio, video, molecule, html, etc). Can be defined either with `columns` and `data` or with `dataframe`. Args: table_name (`str`): The name to give to the logged table on the wandb workspace columns (List of `str`'s *optional*): The name of the columns on the table data (List of List of Any data type *optional*): The data to be logged in the table dataframe (Any data type *optional*): The data to be logged in the table step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. rN)rtrurvr0)rgTabler2) rrsrtrurvr0rrgr,rrr log_table_szWandBTracker.log_tablecCs|jtddS)z' Closes `wandb` writer zWandB run closedN)rkr4rGrHr3rrrr4~s zWandBTracker.finish)N)N)NNNN)r5r6r7r8r"r#rrrQr*rdr$r9r.rr:r2rarrrxr4rerrrIrrf s8    rfcsveZdZdZdZdZeedfdd Ze ddZ ee d d d Z ede e ed ddZeddZZS)CometMLTrackeraZ A `Tracker` class that supports `comet_ml`. Should be initialized at the start of your script. API keys must be stored in a Comet config file. Args: run_name (`str`): The name of the experiment run. kwargs: Additional key word arguments passed along to the `Experiment.__init__` method. comet_mlFrhc sPt||_ddlm}|fd|i||_td|jtddS)Nr) Experiment project_namezInitialized CometML project r@)rBr*r>rzr{rFrGrH)rr>rr{rIrrr*s  zCometMLTracker.__init__cCs|jSrKrLr3rrrr$szCometMLTracker.trackerr+cCs|j|tddS)rlz7Stored initial configuration hyperparameters to CometMLN)rFZlog_parametersrGrHr-rrrr.s z'CometMLTracker.store_init_configurationNr/cKs|dk r|j||D]r\}}t|ttfrN|jj||fd|i|qt|trl|jj||f|qt|t r|jj |fd|i|qt ddS)a Logs `values` to the current run. Args: values (Dictionary `str` to `str`, `float`, `int` or `dict` of `str` to `float`/`int`): Values to be logged as key-value pairs. The values need to have type `str`, `float`, `int` or `dict` of `str` to `float`/`int`. step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. kwargs: Additional key word arguments passed along to either `Experiment.log_metric`, `Experiment.log_other`, or `Experiment.log_metrics` method based on the contents of `values`. Nr0zSuccessfully logged to CometML) rFZset_steprZr[r:r\Z log_metricrQZ log_otherr9 log_metricsrGrHr^rrrr2s   zCometMLTracker.logcCs|jtddS)z* Closes `comet-ml` writer zCometML run closedN)rFendrGrHr3rrrr4s zCometMLTracker.finish)N)r5r6r7r8r"r#rrQr*rdr$r9r.rr:r2r4rerrrIrrys    ryc@s~eZdZdZdZdZedeee ee j fdddZ e dd Zeed d d Zeeeed ddZeddZdS) AimTrackera A `Tracker` class that supports `aim`. Should be initialized at the start of your script. Args: run_name (`str`): The name of the experiment run. kwargs: Additional key word arguments passed along to the `Run.__init__` method. aimT.r=cKsP||_ddlm}|fd|i||_|j|j_td|jtddS)Nr)RunrepozInitialized Aim project r@)r>rrrFr"rGrH)rr>r?rrrrrr*s  zAimTracker.__init__cCs|jSrKrLr3rrrr$szAimTracker.trackerr+cCs||jd<dS) Logs `values` as hyperparameters for the run. Should be run at the beginning of your experiment. Args: values (`dict`): Values to be stored as initial hyperparameters as key-value pairs. ZhparamsNrLr-rrrr.s z#AimTracker.store_init_configurationr/cKs0|D]"\}}|jj|f||d|qdS)a} Logs `values` to the current run. Args: values (`dict`): Values to be logged as key-value pairs. step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. kwargs: Additional key word arguments passed along to the `Run.track` method. )r"r0N)rZrFtrack)rr,r0rkeyvaluerrrr2szAimTracker.logcCs|jdS)z% Closes `aim` writer N)rFrbr3rrrr4szAimTracker.finishN)r)r5r6r7r8r"r#rrQrrrCrcr*rdr$r9r.r:r2r4rrrrrs "   rc @seZdZdZdZdZedeee ee j feeee e ee fefeeeeeedddZedd Zeed d d Zeeeed ddZeddZdS) MLflowTrackeru A `Tracker` class that supports `mlflow`. Should be initialized at the start of your script. Args: experiment_name (`str`, *optional*): Name of the experiment. Environment variable MLFLOW_EXPERIMENT_NAME has priority over this argument. logging_dir (`str` or `os.PathLike`, defaults to `"."`): Location for mlflow logs to be stored. run_id (`str`, *optional*): If specified, get the run with the specified UUID and log parameters and metrics under that run. The run’s end time is unset and its status is set to running, but the run’s other attributes (source_version, source_type, etc.) are not changed. Environment variable MLFLOW_RUN_ID has priority over this argument. tags (`Dict[str, str]`, *optional*): An optional `dict` of `str` keys and values, or a `str` dump from a `dict`, to set as tags on the run. If a run is being resumed, these tags are set on the resumed run. If a new run is being created, these tags are set on the new run. Environment variable MLFLOW_TAGS has priority over this argument. nested_run (`bool`, *optional*, defaults to `False`): Controls whether run is nested in parent run. True creates a nested run. Environment variable MLFLOW_NESTED_RUN has priority over this argument. run_name (`str`, *optional*): Name of new run (stored as a mlflow.runName tag). Used only when `run_id` is unspecified. description (`str`, *optional*): An optional string that populates the description box of the run. If a run is being resumed, the description is set on the resumed run. If a new run is being created, the description is set on the new run. mlflowFN)experiment_namer?run_idtags nested_runr> descriptionc Cstd|}td|}td|}t|tr8t|}td|}ddl}|jd|dd} t| dkrt| d krt d | dj } n|j |||d } |j || ||||d |_t d |t ddS)NZMLFLOW_EXPERIMENT_NAMEZ MLFLOW_RUN_IDZ MLFLOW_TAGSZMLFLOW_NESTED_RUNrzname = '')Z filter_stringr z?Multiple experiments with the same name found. Using first one.)r"Zartifact_locationr)r experiment_idr>nestedrrzInitialized mlflow experiment r@)rCgetenvr[rQjsonloadsrZsearch_experimentsr&rGwarningrZcreate_experimentZ start_run active_runrH) rrr?rrrr>rrZexpsrrrrr*:s:           zMLflowTracker.__init__cCs|jSrK)rr3rrrr$iszMLflowTracker.trackerr+c Csddl}t|D]H\}}tt||jjjkrt d|d|d|jjjd||=qt|}t dt||jjj D]$}| t ||||jjj qtddS)rrNz)Trainer is attempting to log a value of "z " for key "zJ" as a parameter. MLflow's log_param() only accepts values no longer than z) characters so we dropped this attribute.z6Stored initial configuration hyperparameters to MLflow)rlistrZr&rQutilsZ validationZMAX_PARAM_VAL_LENGTHrGrrangeZMAX_PARAMS_TAGS_PER_BATCHZ log_paramsr9rH)rr,rr"rZ values_listirrrr.ms  "z&MLflowTracker.store_init_configurationr/c Csti}|D]B\}}t|ttfr,|||<q td|dt|d|dq ddl}|j||dt ddS) a Logs `values` to the current run. Args: values (`dict`): Values to be logged as key-value pairs. step (`int`, *optional*): The run step. If included, the log will be affiliated with this step. z/MLflowTracker is attempting to log a value of "z " of type z for key "zc" as a metric. MLflow's log_metric() only accepts float and int types so we dropped this attribute.rN)r0zSuccessfully logged to mlflow) rZr[r:r\rGrtyperr}rH)rr,r0Zmetricsr_r`rrrrr2s  zMLflowTracker.logcCsddl}|dS)z, End the active MLflow run. rN)rZend_run)rrrrrr4szMLflowTracker.finish)NNNNFNN)r5r6r7r8r"r#rrQrrrCrcrrboolr*rdr$r9r.r:r2r4rrrrrs8. r)rrzrr<rg)log_withr?cCsg}|dk rt|ttfs"|g}d|ks4tj|krJdd|Dt}n|D]}|tkrtt|tst d|dttt|tr| |qNt|}||krN|tkrt t |}t |dr|dkrt d|d | |qNtd |d qN|S) a Takes in a list of potential tracker types and checks that: - The tracker wanted is available in that environment - Filters out repeats of tracker types - If `all` is in `log_with`, will return all trackers in the environment - If a tracker requires a `logging_dir`, ensures that `logging_dir` is not `None` Args: log_with (list of `str`, [`~utils.LoggerType`] or [`~tracking.GeneralTracker`], *optional*): A list of loggers to be setup for experiment tracking. Should be one or several of: - `"all"` - `"tensorboard"` - `"wandb"` - `"comet_ml"` - `"mlflow"` If `"all"` is selected, will pick up all available trackers in the environment and initialize them. Can also accept implementations of `GeneralTracker` for custom trackers, and can be combined with `"all"`. logging_dir (`str`, `os.PathLike`, *optional*): A path to a directory for storing logs of locally-compatible loggers. NallcSsg|]}tt|tr|qSr) issubclassrr )roorrrrrsz#filter_trackers..z Unsupported logging capability: z. Choose between r#zLogging with `z+` requires a `logging_dir` to be passed in.zTried adding logger z+, but package is unavailable in the system.)r[rtupler ALLrrrr ValueErrorappendLOGGER_TYPE_TO_CLASSrQrrGrH)rr?loggersZlog_typeZ tracker_initrrrfilter_trackerss.       r)N)-rrCrP functoolsrtypingrrrrrrTloggingr stater rr r rrrrrrrZ TENSORBOARDZWANDBZCOMETMLZAIMZMLFLOWr5rGrrr r;rfryrrrrQrcrrrrrsP   $      Jo{QC