o )jZ@sddlZddlmZddlmZddlmZmZmZm Z m Z m Z ddl Z ddl mZddl mZddlmZmZddlmZmZmZdd lmZdd lmZd d lmZeZGd ddee jjZdS)N)deepcopy)partial)AnyCallableDictListOptionalUnion)version)nn) DataParallelDistributedDataParallel)save_checkpointsave_configurationsave_pretrained)func_receive_dict_inputs) get_logger)Modelc seZdZdZdfdd ZdeeeffddZ  d d d Z deeeffd d Z ddZ ddZ ddZ deedddefdeeejfdeeeefdedeedef ddZddZZS)! TorchModelz' Base model interface for pytorch Ncs,tj|g|Ri|tjj|dSN)super__init__torchr Module)selfZ model_dirargskwargs __class__h/var/www/html/Deteccion_Ine/venv/lib/python3.10/site-packages/modelscope/models/base/base_torch_model.pyrszTorchModel.__init__returncOs<t|jr||j|dfi|S||j|i|S)Nr)rforward postprocessrrrr r r!__call__s zTorchModel.__call__Tparamsc Cst|ttfr |j}tj|dddd}|dur3||vr+d|vr+d}td|d||vr3||}td|jj d |d |d t | D]\}}| d r`|||d d<| |qJ|j||dtd|S)NcSs|Srr )Zstoragelocr r r!/sz-TorchModel._load_pretrained..T)Z map_locationZ weights_onlyr'z Loading: z does not exist, use params.zLoading z model from z, with param key: [z].zmodule.)strictzload model done.) isinstancer r modulerloadloggerinfor__name__ritems startswithpopZload_state_dict)rnetZ load_pathr+Z param_keyZload_netkvr r r!_load_pretrained&s4    zTorchModel._load_pretrainedcOstr)NotImplementedErrorr%r r r!r#DszTorchModel.forwardcCs |dS)z A method executed at the end of each model initialization, to execute code that needs the model's modules properly initialized (such as weight initialization). N) init_weightsrr r r! post_initGs zTorchModel.post_initcCs||jdSr)apply _init_weightsr;r r r!r:NszTorchModel.init_weightscCst|tjr|jjjddd|jdur|jjdSdSt|tjr?|jjjddd|j dur=|jj|j dSdSt|tj rT|jj|jj ddSdS)zInitialize the weightsgg{Gz?)meanZstdNg?) r,r ZLinearweightdataZnormal_ZbiasZzero_Z EmbeddingZ padding_idxZ LayerNormZfill_)rr-r r r!r>Rs      zTorchModel._init_weightsF)Z with_meta target_foldersave_checkpoint_names save_functionconfigsave_config_functioncKsH|dur t|dr |j}t||||fi||dur"|||dSdS)asave the pretrained model, its configuration and other related files to a directory, so that it can be re-loaded Args: target_folder (Union[str, os.PathLike]): Directory to which to save. Will be created if it doesn't exist. save_checkpoint_names (Union[str, List[str]]): The checkpoint names to be saved in the target_folder save_function (Callable, optional): The function to use to save the state dictionary. config (Optional[dict], optional): The config for the configuration.json, might not be identical with model.config save_config_function (Callble, optional): The function to use to save the configuration. Ncfg)hasattrrGr)rrBrCrDrErFrr r r!rbszTorchModel.save_pretrainedcKs@ttjtdkrtj|fi|Stdtjd|S)zCompile torch model with torch>=2.0 Args: kwargs: backend: The backend param of torch.compile mode: The mode param of torch.compile z 2.0.0.devzFTorch compiling needs torch version >= 2.0.0, your torch version is : z, returns original model)r parser __version__compiler/warning)rrr r r!rKs  zTorchModel.compiler)Tr')r1 __module__ __qualname____doc__rrstrrr&r8r#r<r:r>rrrr osPathLikerrrdictrrK __classcell__r r rr!rs:    %r) rQcopyr functoolsrtypingrrrrrr r packagingr r Ztorch.nn.parallelr r Zmodelscope.utils.checkpointrrrZmodelscope.utils.file_utilsrZmodelscope.utils.loggerrZ base_modelrr/rrr r r r!s