o ,hC @sddlmZmZmZddlZddlmZddlmZmZm Z m Z m Z m Z m Z mZmZmZmZmZmZmZmZddgZGdddeZd d ed e d ed e d ed e_deedeedeedeedeedededededededededefddZdeedeedeedeedeedededededededededefddZe ed   ! ! ! !d$deedeedeedeedeed"eedededededededededefd#dZdS)%)castOptionalUnionN)Tensor)_capturable_doc_default_to_fused_or_foreach_differentiable_doc_disable_dynamo_if_unsupported _foreach_doc!_get_capturable_supported_devices_get_scalar_dtype _get_value _maximize_doc _params_doc _to_scalar_use_grad_for_differentiable _view_as_real OptimizerParamsTAdamaxadamaxcseZdZ     dddddded eeefd eeefd ed ed ee de de de ffddZ fddZ ddZ e dddZZS)rMb`?g?g+?:0yE>rNF)maximizedifferentiable capturableparamslrbetaseps weight_decayforeachrrrc st|tr|dkrtdd|kstd|d|ks%td|d|dkr1dks;ntd|dd|dkrGdksQntd |dd|ks\td |t|||||||| d } t|| dS) NrzTensor lr must be 1-elementzInvalid learning rate: zInvalid epsilon value: r?z#Invalid beta parameter at index 0: z#Invalid beta parameter at index 1: zInvalid weight_decay value: )rr r!r"r#rrr) isinstancernumel ValueErrordictsuper__init__) selfrrr r!r"r#rrrdefaults __class__M/var/www/html/scripts/venv/lib/python3.10/site-packages/torch/optim/adamax.pyr+s.  zAdamax.__init__cst||jD]S}|dd|dd|dd|dd|dD]4}|j|g}t|dkr[t|ds[t |d}|drQtj |t |j d ntj |t d |d<q'q dS) Nr#rFrrrrstepdtypedevicer4) r* __setstate__ param_groups setdefaultstategetlentorch is_tensorfloattensorr r5)r,r:grouppp_statestep_valr.r0r1r7Ds&         zAdamax.__setstate__c Csd}|dD]n}|jdurq|t|O}|||jjr"td||j|j|} t| dkr_|drAtjdt |j dntj dt d | d <tj |tj d | d <tj |tj d | d <|| d || d || d q|S)NFrz(Adamax does not support sparse gradientsrrr0r3r$r6r2) memory_formatexp_avgexp_inf)gradr= is_complexappend is_sparse RuntimeErrorr:r<zerosr r5r@ zeros_likepreserve_format) r,rAparams_with_gradgradsexp_avgsexp_infs state_steps has_complexrBr:r0r0r1 _init_groupWs2        zAdamax._init_groupcCs|d}|dur!t |}Wdn1swY|jD]K}g}g}g}g}g}|d\} } |d} |d} |d} |d}|d}|d}|d }|||||||}t|||||| | | | | |||||d q$|S) zPerforms a single optimization step. Args: closure (Callable, optional): A closure that reevaluates the model and returns the loss. Nr r!rr"r#rrr) r!beta1beta2rr"r#rrrrU) _cuda_graph_capture_health_checkr= enable_gradr8rVr)r,closurelossrArPrQrRrSrTrWrXr!rr"r#rrrrUr0r0r1r2zsR    z Adamax.step)rrrrNN)__name__ __module__ __qualname__rrr?rtuplerboolr+r7rVrr2 __classcell__r0r0r.r1rsB       &#aImplements Adamax algorithm (a variant of Adam based on infinity norm). .. math:: \begin{aligned} &\rule{110mm}{0.4pt} \\ &\textbf{input} : \gamma \text{ (lr)}, \beta_1, \beta_2 \text{ (betas)},\theta_0 \text{ (params)},f(\theta) \text{ (objective)}, \: \lambda \text{ (weight decay)}, \\ &\hspace{13mm} \epsilon \text{ (epsilon)} \\ &\textbf{initialize} : m_0 \leftarrow 0 \text{ ( first moment)}, u_0 \leftarrow 0 \text{ ( infinity norm)} \\[-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}m_t \leftarrow \beta_1 m_{t-1} + (1 - \beta_1) g_t \\ &\hspace{5mm}u_t \leftarrow \mathrm{max}(\beta_2 u_{t-1}, |g_{t}|+\epsilon) \\ &\hspace{5mm}\theta_t \leftarrow \theta_{t-1} - \frac{\gamma m_t}{(1-\beta^t_1) u_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 `Adam: A Method for Stochastic Optimization`_. z Args: a lr (float, Tensor, optional): learning rate (default: 2e-3) betas (Tuple[float, float], optional): coefficients used for computing running averages of gradient and its square eps (float, optional): term added to the denominator to improve numerical stability (default: 1e-8) weight_decay (float, optional): weight decay (L2 penalty) (default: 0) z zd .. _Adam\: A Method for Stochastic Optimization: https://arxiv.org/abs/1412.6980 rrQrRrSrTr!rWrXrr"rrrrUc Cstjs t|}t|D]\}}||}| s|n| }||}||}||}tjsH| rHt}|jj |jj kr@|jj |vsHJd|d|d7}| dkrW|j || d}t |rpt |}t |}t |}t |}| |d|| stj|||||dn!t||d||dgd}|tj|ddd| r||d}||||}|||q d|t|}||}|j||| d q dS) NIIf capturable=True, params and state_steps must be on supported devices: .rralpha)outF)keepdim)value)r=jit is_scriptingr enumeratecompiler is_compilingr r5typeaddrI view_as_reallerp_maximummul_absadd_cat unsqueeze unsqueeze_copy_amaxdiv_addcdiv_r)rrQrRrSrTr!rWrXrr"rrrrUiparamrHrFrGstep_tcapturable_supported_devicesnorm_bufneg_bias_correctiondenombias_correctionclrr0r0r1_single_tensor_adamaxsV        "  rc s>| rJdt|dkrdStjs0| r0tddtfddt||Ds0Jddtt |||||g}| D]\\}}}}}}t t t |}t t t |}t t t |}t t t |}t t t |}| rwt||||| r~t|}tjs|djrtj|tjd d d d d nt|d | dkr| rtj||| d ntj||| d }t||d t||| s| dkrt|}nt|t||t||| rt|}t|d t|t||}t|||qBfdd|D}fdd|D}t||||qBdS)Nz#_foreach ops don't support autogradrF) supports_xlac3s0|]\}}|jj|jjko|jjvVqdSr])r5rp).0rBr2)rr0r1 Ls   z'_multi_tensor_adamax..rdrer%cpu)r5rfrcsg|] }dt|qS)rr)rr2)rWr0r1 sz(_multi_tensor_adamax..csg|] }t|dqS)r)rbc)rr0r1rs)r<r=rnror allziprr"_group_tensors_by_device_and_dtypevaluesrlistrr _foreach_negis_cpu _foreach_add_r@ _foreach_add_foreach_lerp_ _foreach_mul_ _foreach_abs _foreach_abs__foreach_maximum_ _foreach_pow _foreach_sub_ _foreach_div_ _foreach_mul_foreach_addcdiv_)rrQrRrSrTr!rWrXrr"rrrrUgrouped_tensorsgrouped_params_grouped_grads_grouped_exp_avgs_grouped_exp_infs_grouped_state_steps__grouped_params grouped_gradsgrouped_exp_avgsgrouped_exp_infsgrouped_state_stepsbias_correctionsr step_sizer0)rWrrr1_multi_tensor_adamax1s                   r)single_tensor_fnFr#c Cstjstdd|Dstd|durt||dd\}}|r*tjr*td|r4tjs4t}nt }||||||| | | | |||| |ddS) zrFunctional API that performs adamax algorithm computation. See :class:`~torch.optim.Adamax` for details. css|] }t|tjVqdSr])r&r=r)rtr0r0r1rs  zadamax..zPAPI has changed, `state_steps` argument must contain a list of singleton tensorsNF) use_fusedz6torch.jit.script not supported with foreach optimizers) r!rWrXrr"rrrUr) r=rnrorrLrrkrlrr)rrQrRrSrTr#rrrrUr!rWrXrr"rfuncr0r0r1rs>  )NFFFF)typingrrrr=r optimizerrrr r r r r rrrrrrrr__all__r__doc__rr?rbrrrr0r0r0r1s D    0      O      t