o h@@sddlmZmZmZmZddlZddlmZddlmZm Z m Z m Z m Z m Z mZmZmZmZmZmZmZddgZGdddeZd d ed e d ed ed e d e_deedeedeedeedeededededededededefddZdeedeedeedeedeededededededededefddZe ed  d#deedeedeedeedeeded!eedededededededefd"dZdS)$)AnycastOptionalUnionN)Tensor) _capturable_doc_default_to_fused_or_foreach_differentiable_doc_disable_dynamo_if_unsupported _foreach_doc!_get_capturable_supported_devices_get_scalar_dtype _maximize_doc _params_doc_use_grad_for_differentiable _view_as_real OptimizerParamsTAdadeltaadadeltacseZdZ     dddddded eeefd ed ed ed eedededeffddZ fddZ de e e fdeedeedeedeedeef ddZed ddZZS)!r??ư>rNF) capturablemaximizedifferentiableparamslrrhoeps weight_decayforeachrrrc st|tr|dkrtdd|kstd|d|kr$dks,ntd|d|ks7td|d|ksBtd|t|||||||| d } t|| dS) NrzTensor lr must be 1-elementgzInvalid learning rate: rzInvalid rho value: zInvalid epsilon value: zInvalid weight_decay value: )rrr r!rrr"r) isinstancernumel ValueErrordictsuper__init__) selfrrrr r!r"rrrdefaults __class__h/var/www/html/construction_image-detection-poc/venv/lib/python3.10/site-packages/torch/optim/adadelta.pyr(s*  zAdadelta.__init__cst||jD]S}|dd|dd|dd|dd|dD]4}|j|g}t|dkr[t|ds[t |d}|drQtj |t |j d ntj |t d |d<q'q dS) Nr"rFrrrrstepdtypedevicer1) r' __setstate__ param_groups setdefaultstategetlentorch is_tensorfloattensorrr2)r)r7grouppp_statestep_valr+r-r.r4@s&         zAdadelta.__setstate__r>params_with_gradgrads square_avgs acc_deltas state_stepsc Csd}|dD]n}|jdurq|t|O}|||jjr"td||j|j|} t| dkr_|drAtjdt |j dntjdt d| d <tj |tj d | d <tj |tj d | d <|| d || d || d q|S) NFrz*Adadelta does not support sparse gradientsrrr-r0r3r/) memory_format square_avg acc_delta) gradr: is_complexappend is_sparse RuntimeErrorr7r9zerosrr2 zeros_likepreserve_format) r)r>rBrCrDrErF has_complexr?r7r-r-r. _init_groupSs2        zAdadelta._init_groupcCs|d}|dur!t |}Wdn1swY|jD]J}g}g}g}g}g}|d|d|d|d|d|d|d|d f\} } } } } }}}|||||||}t|||||| | | | | ||||d q$|S) zPerform a single optimization step. Args: closure (Callable, optional): A closure that reevaluates the model and returns the loss. Nrrr r!r"rrr) rrr r!r"rrrrR) _cuda_graph_capture_health_checkr: enable_gradr5rSr)r)closurelossr>rBrCrDrErFrrr r!r"rrrrRr-r-r.r/~sd    z Adadelta.step)rrrrNN)__name__ __module__ __qualname__rrr<rrboolr(r4r&strrlistrSrr/ __classcell__r-r-r+r.rsZ      $  +aImplements Adadelta algorithm. .. math:: \begin{aligned} &\rule{110mm}{0.4pt} \\ &\textbf{input} : \gamma \text{ (lr)}, \: \theta_0 \text{ (params)}, \: f(\theta) \text{ (objective)}, \: \rho \text{ (decay)}, \: \lambda \text{ (weight decay)} \\ &\textbf{initialize} : v_0 \leftarrow 0 \: \text{ (square avg)}, \: u_0 \leftarrow 0 \: \text{ (accumulate variables)} \\[-1.ex] &\rule{110mm}{0.4pt} \\ &\textbf{for} \: t=1 \: \textbf{to} \: \ldots \: \textbf{do} \\ &\hspace{5mm}g_t \leftarrow \nabla_{\theta} f_t (\theta_{t-1}) \\ &\hspace{5mm}if \: \lambda \neq 0 \\ &\hspace{10mm} g_t \leftarrow g_t + \lambda \theta_{t-1} \\ &\hspace{5mm} v_t \leftarrow v_{t-1} \rho + g^2_t (1 - \rho) \\ &\hspace{5mm}\Delta x_t \leftarrow \frac{\sqrt{u_{t-1} + \epsilon }}{ \sqrt{v_t + \epsilon} }g_t \hspace{21mm} \\ &\hspace{5mm} u_t \leftarrow u_{t-1} \rho + \Delta x^2_t (1 - \rho) \\ &\hspace{5mm}\theta_t \leftarrow \theta_{t-1} - \gamma \Delta x_t \\ &\rule{110mm}{0.4pt} \\[-1.ex] &\bf{return} \: \theta_t \\[-1.ex] &\rule{110mm}{0.4pt} \\[-1.ex] \end{aligned} For further details regarding the algorithm we refer to `ADADELTA: An Adaptive Learning Rate Method`_. z Args: ar lr (float, Tensor, optional): coefficient that scale delta before it is applied to the parameters (default: 1.0) rho (float, optional): coefficient used for computing a running average of squared gradients (default: 0.9). A higher value of `rho` will result in a slower average, which can be helpful for preventing oscillations in the learning process. eps (float, optional): term added to the denominator to improve numerical stability (default: 1e-6). weight_decay (float, optional): weight decay (L2 penalty) (default: 0) z zd .. _ADADELTA\: An Adaptive Learning Rate Method: https://arxiv.org/abs/1212.5701 rrCrDrErFrrr r!rrrrRcsLtjs"| r"tddtfddt||Ds"Jddt|||||D]y\} }}}}|d7}| s9|n| }|dkrG|j| |d }t| r[t|}t|}t|}| |j ||d|d || }|| }| r|| }| | || |j ||d|d t| rt|}| j|| d q*dS) NF supports_xlac30|]\}}|jj|jjko|jjvVqdSrXr2type.0r?r/capturable_supported_devicesr-r.    z*_single_tensor_adadelta..IIf capturable=True, params and state_steps must be on supported devices: .rralphavalue)r:compiler is_compilingr allzipaddrK view_as_realmul_addcmul_sqrt_clonediv_view_as_complexadd_)rrCrDrErFrrr r!rrrrRparamrJrHrIr/stddeltar-rgr._single_tensor_adadeltas>         rcs.| rJdtjs(| r(tddtfddt||Ds(Jddt|dkr0dSt|||||g} | D]\\}}}}}}t t t |}t t t |}t t t |}t t t |}t t t |}| rst ||||tjs|djrtj|tjd d d d d nt|d | rt|}|dkr| rtj|||d ntj|||d }t||tj|||d |dt||}t|t||}t|t||t||t||tj|||d |d| r t|tj r t|| t||q>tj||| d q>dS)Nz#_foreach ops don't support autogradFr`c3rbrXrcrergr-r.riArjz)_multi_tensor_adadelta..rkrlrrcpu)r2rmrro)r:rqrrr rsrtr9r"_group_tensors_by_device_and_dtypevaluesrr^rris_cpu _foreach_add_r= _foreach_neg _foreach_add _foreach_mul__foreach_addcmul__foreach_sqrt_ _foreach_div_r#)rrCrDrErFrrr r!rrrrRgrouped_tensorsdevice_params_ device_grads_device_square_avgs_device_acc_deltas_device_state_steps__ device_params device_gradsdevice_square_avgsdevice_acc_deltasdevice_state_stepsrdeltasr-rgr._multi_tensor_adadelta*s|                 r)single_tensor_fnFr"c Cstjstdd|Dstd|durt||dd\}}|r*tjr*td|r4tjs4t}nt }||||||| | | | | |||d dS) zvFunctional API that performs Adadelta algorithm computation. See :class:`~torch.optim.Adadelta` for details. css|] }t|tjVqdSrX)r#r:r)rftr-r-r.ris  zadadelta..zPAPI has changed, `state_steps` argument must contain a list of singleton tensorsNF) use_fusedz6torch.jit.script not supported with foreach optimizers)rrr r!rrrrR) r:rqrrrsrNr jit is_scriptingrr)rrCrDrErFrr"rrRrrr r!rrfuncr-r-r.rs<  )FNFF)typingrrrrr:r optimizerrr r r r r rrrrrrr__all__r__doc__r^r<r\rrrr-r-r-r.s <&  5      6      d