+
    Ñåûj+  ã                   óü   € ^ RI Ht ^RIHtHt ^RIHtHtHtHtH	t	H
t
HtHtHt RR.t ! R R]4      tRR] R	]
 R
] R
] R
] R
]	 R2,           ]n        RR R llt]P$                  ! RR7      R,           ]n        R# )é    )ÚTensor)ÚAdamÚadam)	Ú_capturable_docÚ_differentiable_docÚ_foreach_docÚ_functional_api_docÚ
_fused_docÚ_maximize_docÚ_params_docÚParamsTÚ	StateDictÚAdamWÚadamwc                   ór   a a€ ] tR t^t oRRRRRRRRRRR/V3R lV 3R	 lllltV3R
 lV 3R lltRtVtV ;t# )r   FÚmaximizeÚforeachNÚ
capturableÚdifferentiableÚfusedc                óÂ   <€ V ^8„  d   QhRS[ RS[S[,          RS[S[S[,          S[S[,          3,          RS[RS[RS[RS[RS[R	,          R
S[RS[RS[R	,          RR	/# )é   ÚparamsÚlrÚbetasÚepsÚweight_decayÚamsgradr   r   Nr   r   r   Úreturn)r   Úfloatr   ÚtupleÚbool)ÚformatÚ__classdict__s   "€ÚMC:\Ai_Coding\apps\studio\ComfyUI\.venv\Lib\site-packages\torch\optim\adamw.pyÚ__annotate__ÚAdamW.__annotate__   s¡   ø€ ÷ 
ñ 
áð
ñ ‘F�Nð
ñ ‘U™V•^¡U©V¥^Ð3Õ4ð	
ñ
 ð
ñ ð
ñ ð
ñ ð
ñ ˜•ð
ñ ð
ñ ð
ñ �d�{ð
ð 
ñ
ó    c               ó>   <€ \         SV `  VVVVVVVVV	V
VR R7       R# )T)r   r   r   r   r   Údecoupled_weight_decayN)ÚsuperÚ__init__)Úselfr   r   r   r   r   r   r   r   r   r   r   Ú	__class__s   &&&&&&&$$$$$€r%   r,   ÚAdamW.__init__   s;   ø€ ô 	‰ÑØØØØØØØØØ!Ø)ØØ#'ð 	ö 	
r(   c                ó$   <€ V ^8„  d   QhRS[ RR/# )r   Ústater   N)r   )r#   r$   s   "€r%   r&   r'   6   s   ø€ ÷ 3ñ 3¡)ð 3°ñ 3r(   c                óV   <€ \         SV `  V4       V P                   F  pR VR&   K
  	  R# )Tr*   N)r+   Ú__setstate__Úparam_groups)r-   r1   Úgroupr.   s   && €r%   r3   ÚAdamW.__setstate__6   s+   ø€ Ü‰Ñ˜UÔ#Ø×&Ô&ˆEØ.2ˆEÐ*Ó+ó 'r(   © )gü©ñÒMbP?)gÍÌÌÌÌÌì?g+‡ÙÎ÷ï?g:Œ0âŽyE>g{®Gáz„?F)	Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__r,   r3   Ú__static_attributes__Ú__classdictcell__Ú__classcell__)r.   r$   s   @@r%   r   r      sQ   ù‡ € ñ
ð ð
ð  $ð
ð !ð
ð  %ð
ð "÷
ö 
÷B3÷ 3ó 3r(   a­  Implements AdamW algorithm, where weight decay does not accumulate in the momentum nor variance.

    .. 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)},
                \: \epsilon \text{ (epsilon)}                                                    \\
            &\hspace{13mm}      \lambda \text{(weight decay)},  \: \textit{amsgrad},
                \: \textit{maximize}                                                             \\
            &\textbf{initialize} : m_0 \leftarrow 0 \text{ (first moment)}, v_0 \leftarrow 0
                \text{ ( second moment)}, \: v_0^{max}\leftarrow 0                        \\[-1.ex]
            &\rule{110mm}{0.4pt}                                                                 \\
            &\textbf{for} \: t=1 \: \textbf{to} \: \ldots \: \textbf{do}                         \\

            &\hspace{5mm}\textbf{if} \: \textit{maximize}:                                       \\
            &\hspace{10mm}g_t           \leftarrow   -\nabla_{\theta} f_t (\theta_{t-1})         \\
            &\hspace{5mm}\textbf{else}                                                           \\
            &\hspace{10mm}g_t           \leftarrow   \nabla_{\theta} f_t (\theta_{t-1})          \\
            &\hspace{5mm} \theta_t \leftarrow \theta_{t-1} - \gamma \lambda \theta_{t-1}         \\
            &\hspace{5mm}m_t           \leftarrow   \beta_1 m_{t-1} + (1 - \beta_1) g_t          \\
            &\hspace{5mm}v_t           \leftarrow   \beta_2 v_{t-1} + (1-\beta_2) g^2_t          \\
            &\hspace{5mm}\widehat{m_t} \leftarrow   m_t/\big(1-\beta_1^t \big)                   \\
            &\hspace{5mm}\textbf{if} \: amsgrad                                                  \\
            &\hspace{10mm} v_t^{max} \leftarrow \mathrm{max}(v_{t-1}^{max},v_t)                  \\
            &\hspace{10mm}\widehat{v_t} \leftarrow v_t^{max}/\big(1-\beta_2^t \big)              \\
            &\hspace{5mm}\textbf{else}                                                           \\
            &\hspace{10mm}\widehat{v_t} \leftarrow   v_t/\big(1-\beta_2^t \big)                  \\
            &\hspace{5mm}\theta_t \leftarrow \theta_t - \gamma \widehat{m_t}/
                \big(\sqrt{\widehat{v_t}} + \epsilon \big)                                       \\
            &\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 `Decoupled Weight Decay Regularization`_.
    z
    Args:
        ac  
        lr (float, Tensor, optional): learning rate (default: 1e-3). A tensor LR
            is not yet supported for all our implementations. Please use a float
            LR if you are not also specifying fused=True or capturable=True.
        betas (tuple[float | Tensor, float | Tensor], optional):
            coefficients used for computing running averages of gradient and
            its square. If a tensor is provided, must be 1-element. (default: (0.9, 0.999))
        eps (float, optional): term added to the denominator to improve
            numerical stability (default: 1e-8)
        weight_decay (float, optional): weight decay coefficient (default: 1e-2)
        amsgrad (bool, optional): whether to use the AMSGrad variant of this
            algorithm from the paper `On the Convergence of Adam and Beyond`_
            (default: False)
        z	
        a8  
    .. Note::
        A prototype implementation of Adam and AdamW for MPS supports `torch.float32` and `torch.float16`.
    .. _Decoupled Weight Decay Regularization:
        https://arxiv.org/abs/1711.05101
    .. _On the Convergence of Adam and Beyond:
        https://openreview.net/forum?id=ryQu7f-RZ

    Nc          *      ó
  € V ^8„  d   QhR\         \        ,          R\         \        ,          R\         \        ,          R\         \        ,          R\         \        ,          R\         \        ,          R\        R,          R	\        R
\        R\        R,          R\        R,          R\        R,          R\        R\        R\        \        ,          R\        \        ,          R\        \        ,          R\        R\        R\        RR/# )r   r   ÚgradsÚexp_avgsÚexp_avg_sqsÚmax_exp_avg_sqsÚstate_stepsr   Nr   r   r   Ú
grad_scaleÚ	found_infÚhas_complexr   Úbeta1Úbeta2r   r   r   r   r   )Úlistr   r"   r    )r#   s   "r%   r&   r&   ‚   s  € ÷ /ñ /Ü”�Lð/ä”�<ð/ô ”6�lð/ô ”f•ð	/ô
 œ&•\ð/ô ”f•ð/ô �D�[ð/ô ð/ô ð/ô �$�;ð/ô ˜•ð/ô ˜�}ð/ô ð/ô" ð#/ô$ ”6�>ð%/ô& ”6�>ð'/ô( 	”�ð)/ô* ð+/ô, 
ð-/ô. ð//ð0 
ñ1/r(   c          !      ó`   € \        V VVVVV3R VRVRVRV	RV
RVRVRVRVR	VR
VRVRVRVRR/  R# )r   r   r   r   rE   rF   rG   r   rH   rI   r   r   r   r   r*   TN)r   )r   r@   rA   rB   rC   rD   r   r   r   r   rE   rF   rG   r   rH   rI   r   r   r   r   s   &&&&&&&&&&&&&$$$$$$$r%   r   r   ‚   s´   € ô2 	ØØØØØØñð ðð ðð &ðð ðð ðð ðð  ðð ðð ðð  ð!ð" ð#ð$ "ð%ð& ð'ð( ð)ð*  $ô+r(   )Ú	optimizerz¦

.. note::
    With ``fused=True``, CUDA supports FP32 parameters and gradients with
    BF16 moment buffers. See :ref:`functional-adamw-bf16-state` for an example.
)NFFNNNF)Útorchr   r   r   rL   r   r   r   r	   r
   r   r   r   r   Ú__all__r   Ú__doc__r   r#   r7   r(   r%   Ú<module>rP      s¹   ðÝ ç ÷
÷ 
õ 
ð �GÐ
€ô%3ˆDô %3ðR$ðJ	à	ˆð 	ð 
ˆð 	Ø	ˆð 	Ø	Ðð 	Ø	Ðð 	Ø	ˆð ð'õK@ð „÷L/ðf ×Ò¨Ô1ðõð †r(   