9iO$6SrSSKJrJrJr SSKrSSKJr SSKJrJ r J r J r J r J r JrJrJrJrJrJrJrJr SS/r"S S\5rS S \S \S \ S \S \ S3 -\lS\\S\\S\\S\\S\\S\\S\S\S\S\S\S\S\S\S\S\4 SjrS\\S\\S\\S\\S\\S\\S\S\S\S\S\S\S\S\S\S\4 S jr\ "\S!9S$S\\S\\S\\S\\S\\S\\S"\\S\S\S\S\S\S\S\S\S\S\4"S#jj5rg)%z)Implementation for the RMSprop algorithm.)castOptionalUnionN)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 _to_scalar_use_grad_for_differentiable _view_as_real OptimizerParamsTRMSproprmspropc^\rSrSrSS\S\\\4S\S\S\S\S\S \S \ \S \S \4U4S jjjr U4Sjr Sr \ SSj5rSrU=r$)rparamslralphaeps weight_decaymomentumcentered capturableforeachmaximizedifferentiablec n>[U[5(aUR5S:wa [S5eSU::d[SU35eSU::d[SU35eSU::d[SU35eSU::d[SU35eSU::d[SU35eUUUUUUUU U U S . n [T U]X5 g) NrzTensor lr must be 1-elementgzInvalid learning rate: zInvalid epsilon value: zInvalid momentum value: zInvalid weight_decay value: zInvalid alpha value: ) rrrrr rr!r"r#r$) isinstancernumel ValueErrorsuper__init__)selfrrrrrrr r!r"r#r$defaults __class__s S/var/www/html/land-doc-ocr/venv/lib/python3.13/site-packages/torch/optim/rmsprop.pyr*RMSprop.__init__s b& ! !bhhjAo:; ;by6rd;< <cz6se<= =h7zBC Cl";L>JK Ke|4UG<= =  ($ ,   *c>[TU]U5 URGH)nURSS5 URSS5 URSS5 URSS5 URSS5 URSS5 US HnURR U/5n[ U5S:wdM0[R"US 5(aMP[US 5nUS(a([R"U[5URS 9O[R"U[5S 9US 'M GM, g) Nrrr Fr"r#r$r!rstepdtypedevicer4) r) __setstate__ param_groups setdefaultstategetlentorch is_tensorfloattensorrr5)r+r:grouppp_statestep_valr-s r.r7RMSprop.__setstate__Hs U#&&E   Z +   Z /   Y -   Z /   -u 5   \5 18_**..B/w<1$U__WV_-M-M$WV_5H !. $,=,?#\\(:K:MN FO %'r0cSnUSGHn U RcMU[R"U 5-nURU 5 U RR(a [ S5eURU R5 UR U n [U 5S:XaUS(a([R"S[5U RS9O[R"S[5S9U S '[R"U [RS 9U S 'US S:a&[R"U [RS 9U S 'US(a&[R"U [RS 9U S'URU S 5 URU S 5 US S:aURU S 5 US(dGMURU S5 GM U$)NFrz)RMSprop does not support sparse gradientsrr!r3r6r2) memory_format square_avgrmomentum_bufferr grad_avg) gradr= is_complexappend is_sparse RuntimeErrorr:r<zerosrr5 zeros_likepreserve_format) r+rAparams_with_gradgrads square_avgsmomentum_buffer_list grad_avgs state_steps has_complexrBr:s r. _init_groupRMSprop._init_group]s xAvv~ 5++A. .K  # #A &vv"#NOO LL JJqME5zQ\*KK*;*=ahhOR/@/BCf ',&6&6U%:%:'l#$q(/4/?/?)>)>0E+,$(-(8(8)>)>)E*%   u\2 3   uV} -Z 1$$++E2C,DEZ    z!23I!Lr0cvUR5 SnUb%[R"5 U"5nSSS5 URH]n/n/n/n/n/n/n UR UUUUUUU 5n [ UUUUUU USUSUSUSUSUSUSUS US US U S 9 M_ U$!,(df  N}=f) zPerform a single optimization step. Args: closure (Callable, optional): A closure that reevaluates the model and returns the loss. Nrrrrrr r"r#r$r!) rrrrrr r"r#r$r!rZ) _cuda_graph_capture_health_checkr= enable_gradr8r[r) r+closurelossrArTrUrVrXrWrYrZs r.r2 RMSprop.steps  --/  ""$y%&&E-/ "$E(*K&(I13 (*K** $K  $;Gn%L">2z*z*i(z*$%56 .'# %'L S%$s B** B8rG) g{Gz?gGz?g:0yE>rrFFNFFN)__name__ __module__ __qualname____firstlineno__rrr?rboolrr*r7r[rr2__static_attributes__ __classcell__)r-s@r.rrs$( "&$'+'+ %- '+ '+  '+  '+'+'+'+$'+'+'+'+R*1f"4"4r0aj Implements RMSprop algorithm. .. math:: \begin{aligned} &\rule{110mm}{0.4pt} \\ &\textbf{input} : \alpha \text{ (alpha)}, \: \gamma \text{ (lr)}, \: \theta_0 \text{ (params)}, \: f(\theta) \text{ (objective)} \\ &\hspace{13mm} \lambda \text{ (weight decay)},\: \mu \text{ (momentum)}, \: centered, \: \epsilon \text{ (epsilon)} \\ &\textbf{initialize} : v_0 \leftarrow 0 \text{ (square average)}, \: \textbf{b}_0 \leftarrow 0 \text{ (buffer)}, \: g^{ave}_0 \leftarrow 0 \\[-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 \alpha v_{t-1} + (1 - \alpha) g^2_t \hspace{8mm} \\ &\hspace{5mm} \tilde{v_t} \leftarrow v_t \\ &\hspace{5mm}if \: centered \\ &\hspace{10mm} g^{ave}_t \leftarrow g^{ave}_{t-1} \alpha + (1-\alpha) g_t \\ &\hspace{10mm} \tilde{v_t} \leftarrow \tilde{v_t} - \big(g^{ave}_{t} \big)^2 \\ &\hspace{5mm}if \: \mu > 0 \\ &\hspace{10mm} \textbf{b}_t\leftarrow \mu \textbf{b}_{t-1} + g_t/ \big(\sqrt{\tilde{v_t}} + \epsilon \big) \\ &\hspace{10mm} \theta_t \leftarrow \theta_{t-1} - \gamma \textbf{b}_t \\ &\hspace{5mm} else \\ &\hspace{10mm}\theta_t \leftarrow \theta_{t-1} - \gamma g_t/ \big(\sqrt{\tilde{v_t}} + \epsilon \big) \hspace{3mm} \\ &\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 `lecture notes `_ by G. Hinton. and centered version `Generating Sequences With Recurrent Neural Networks `_. The implementation here takes the square root of the gradient average before adding epsilon (note that TensorFlow interchanges these two operations). The effective learning rate is thus :math:`\gamma/(\sqrt{v} + \epsilon)` where :math:`\gamma` is the scheduled learning rate and :math:`v` is the weighted moving average of the squared gradient. z Args: a0 lr (float, Tensor, optional): learning rate (default: 1e-2) alpha (float, optional): smoothing constant (default: 0.99) 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) momentum (float, optional): momentum factor (default: 0) centered (bool, optional) : if ``True``, compute the centered RMSProp, the gradient is normalized by an estimation of its variance z z rrUrVrXrWrYrrrrrr r#r$r!rZc [RR5(d [U5n[ U5GHMunnUUn[R R 5(ddU(a][5nURRURR:XaURRU;d SUS35eUUnU (dUOU*nUUnUS- nU S:waURUU S9n[R"U5nU(aB[R"U5n[R"U5n[R"U5nURU5RUUSU- S9 U (aWUUnU(a[R"U5nURUSU- 5 UR!UUSS9R#5nOUR%5nU (aURU5nOUR'U5nU S:aWUUnU(a[R"U5nURU 5R)UU5 UR'UU*S9 GM;UR)UUU*S9 GMP g)NIIf capturable=True, params and state_steps must be on supported devices: .rrrvalue)r=jit is_scriptingr enumeratecompiler is_compilingr r5typeaddrM view_as_realmul_addcmul_lerp_addcmulsqrt_sqrtadd_addcdiv_)rrUrVrXrWrYrrrrrr r#r$r!rZiparamr2capturable_supported_devicesrLrIis_complex_paramrKavgbufs r._single_tensor_rmspropr s3& 99 ! ! # # ^f%51~~~**,,+L+N ( !!T[[%5%55LL%%)EE \\x[yyz{  F Qx#t$ ^    1 88E86D ++E2 &&u-E%%d+D++J7J''d!e)'D  |H --h7 NN4U +$$Xxr$BHHJC//#C ''#,C((3-C a<&q)C((- HHX  ' 'c 2 JJs2#J & NN4RCN 0g&r0c ^![U5S:XagU (aS5e[RR5(d?U(a8[ 5m![ U!4Sj[ X555(d ST!S35e[U5n[R"XX#XE/5nUR5GHuunnnnnnn[[[U5n[[[U5n[[[U5n[[[U5nU(amUU/nU S:a([[[U5nURU5 U (a([[[U5nURU5 [U/UQ76 U (a[R "U5n[RR5(d>USR"(a*[R$"U[R&"SSS9SS 9 O[R$"US 5 U S:wa4U (a[R$"UUU S 9 O[R("UUU S 9n[R*"UU5 [R,"UUUS U- S 9 U (aw[[[U5n[R."UUS U- 5 [R0"UUUS S 9n[R2"U5 [R$"UU5 O-[R4"U5n[R$"UU5 U S:a[[[U5n[R*"UU 5 [R6"UUU5 U(aQ[9U[R5(a2[R:"UU*5n [R$"UU 5 GM-[R$"UUU*S 9 GMGU(aR[9U[R5(a3[R<"UU*5 [R6"UUU5 GM[R6"UUUU*S 9 GM g) Nrz#_foreach ops don't support autogradc3># UHTupURRURR:H=(a URRT;v MV g7frc)r5rw).0rBr2rs r. (_multi_tensor_rmsprop..psN 4 HHMMT[[-- - > !== >3sAArlrmg?cpu)r5rnrrorq)r<r=rurvr allziprr"_group_tensors_by_device_and_dtypevaluesrlistrrNr _foreach_negis_cpu _foreach_add_r@ _foreach_add _foreach_mul__foreach_addcmul__foreach_lerp__foreach_addcmul_foreach_sqrt_ _foreach_sqrt_foreach_addcdiv_r& _foreach_mul _foreach_div_)"rrUrVrXrWrYrrrrrr r#r$r!rZgrouped_tensorsgrouped_params_grouped_grads_grouped_square_avgs_grouped_grad_avgs_grouped_momentum_buffer_list_grouped_state_steps__grouped_params grouped_gradsgrouped_square_avgsgrouped_state_stepsstate_and_gradsgrouped_momentum_buffer_listgrouped_grad_avgsr momentum_lrrs" @r._multi_tensor_rmsproprUs& 6{aDDD  >> & & ( (Z'H'J$ v3    XXtWuuv w    BBBB  0DRO  " " $     ) d6lO<T&\>: "4<1EF"4<1EF ,.ABO!|/3L"?0, &&'CD$(f7I$J!&&'89 . ;? ; !..}=M ~~**,,1DQ1G1N1N   #U\\#e%DC     3Q 7 1 ##M>V % 2 2!>!  /7   QY   $T&\3E F   !2M1u9 M((#%68IQSC   %   S )%%&9:C   S ) a<+/V ;, (    .s!5-8 1ell##[s')zPAPI has changed, `state_steps` argument must contain a list of singleton tensorsNF) use_fusedz6torch.jit.script not supported with foreach optimizers) rrrrrr r#r!r$rZ) r=rurvrrPr rrrsrr)rrUrVrXrWrYr"r#r$r!rZrrrrrr rfuncs r.rrs: >> & & ( (5-8522 ^  1 e 7599))++STTuyy--//$%   !%!r0)NFFFF)__doc__typingrrrr=r optimizerrr r r r r rrrrrrrr__all__rrr?rhrrrrGr0r.rs?0(( $ i gigV+X       Y<BI1 LI1 <I1fI1F| I1 v, I1 f I1 I1 I1 I1I1I1I1I1I1 !I1"#I1XAW LAW <AWfAWF| AW v, AW f AW AW AW AWAWAWAWAWAW !AW"#AWH 1GH# A LA <AfAF| A v, A f Ad^AAAAA A !A" #A$%A&'A()AIAr0