;iRSSKJr "SS\5r"SS\5r"SS\5rg) )Optimizer1Statec@^\rSrSrSU4SjjrSrU=r$)Adagradc >SU::d[SU35eSU::d[SU35eSU::d[SU35eUS:wa [S5eUS:wa [S5e[T U] SUUSUUUUU U U 5 g ) a Base Adagrad optimizer. Arguments: params (`torch.tensor`): The input parameters to optimize. lr (`float`, defaults to 1e-2): The learning rate. lr_decay (`int`, defaults to 0): The learning rate decay. weight_decay (`float`, defaults to 0.0): The weight decay value for the optimizer. initial_accumulator_value (`int`, defaults to 0): The initial momemtum values. eps (`float`, defaults to 1e-10): The epsilon value prevents division by zero in the optimizer. optim_bits (`int`, defaults to 32): The number of bits of the optimizer state. args (`object`, defaults to `None`): An object with additional arguments. min_8bit_size (`int`, defaults to 4096): The minimum number of elements of the parameter tensors for 8-bit optimization. percentile_clipping (`int`, defaults to 100): Adapts clipping threshold automatically by tracking the last 100 gradient norms and clipping the gradient at a certain percentile to improve stability. block_wise (`bool`, defaults to `True`): Whether to independently quantize each block of tensors to reduce outlier effects and improve stability. Invalid learning rate: Invalid weight_decay value: Invalid epsilon value: /Initial accumulator value != 0.0 not supported!Lr Decay != 0.0 not supported!adagradrrN ValueErrorsuper__init__ selfparamslrlr_decay weight_decayinitial_accumulator_valueeps optim_bitsargs min_8bit_sizepercentile_clipping block_wise __class__s j/home/dmtnaga/Documents/work/airagagent/rag_env/lib/python3.13/site-packages/bitsandbytes/optim/adagrad.pyrAdagrad.__init__ sRby6rd;< <l";L>JK Kcz6se<= = $ +NO O s?=> >             {Gz?rrr绽|= NdT__name__ __module__ __qualname____firstlineno__r__static_attributes__ __classcell__r!s@r"rr- "#  ? ? r$rc@^\rSrSrSU4SjjrSrU=r$) Adagrad8bitKc >SU::d[SU35eSU::d[SU35eSU::d[SU35eUS:wa [S5eUS:wa [S5eU (de[T U] SUUSUUS UU U U 5 g ) a 8-bit Adagrad optimizer. Arguments: params (`torch.tensor`): The input parameters to optimize. lr (`float`, defaults to 1e-2): The learning rate. lr_decay (`int`, defaults to 0): The learning rate decay. weight_decay (`float`, defaults to 0.0): The weight decay value for the optimizer. initial_accumulator_value (`int`, defaults to 0): The initial momemtum values. eps (`float`, defaults to 1e-10): The epsilon value prevents division by zero in the optimizer. optim_bits (`int`, defaults to 8): The number of bits of the optimizer state. args (`object`, defaults to `None`): An object with additional arguments. min_8bit_size (`int`, defaults to 4096): The minimum number of elements of the parameter tensors for 8-bit optimization. percentile_clipping (`int`, defaults to 100): Adapts clipping threshold automatically by tracking the last 100 gradient norms and clipping the gradient at a certain percentile to improve stability. block_wise (`bool`, defaults to `True`): Whether to independently quantize each block of tensors to reduce outlier effects and improve stability. rr r r r r rrrNrrs r"rAdagrad8bit.__init__LsRby6rd;< <l";L>JK Kcz6se<= = $ +NO O s?=> >z            r$r%) r'rrrr(rNr*r+Tr,r3s@r"r6r6Ks/ "#  @ @ r$r6c@^\rSrSrSU4SjjrSrU=r$) Adagrad32bitc >SU::d[SU35eSU::d[SU35eSU::d[SU35eUS:wa [S5eUS:wa [S5e[T U] SUUSUUS UU U U 5 g ) a 32-bit Adagrad optimizer. Arguments: params (`torch.tensor`): The input parameters to optimize. lr (`float`, defaults to 1e-2): The learning rate. lr_decay (`int`, defaults to 0): The learning rate decay. weight_decay (`float`, defaults to 0.0): The weight decay value for the optimizer. initial_accumulator_value (`int`, defaults to 0): The initial momemtum values. eps (`float`, defaults to 1e-10): The epsilon value prevents division by zero in the optimizer. optim_bits (`int`, defaults to 32): The number of bits of the optimizer state. args (`object`, defaults to `None`): An object with additional arguments. min_8bit_size (`int`, defaults to 4096): The minimum number of elements of the parameter tensors for 8-bit optimization. percentile_clipping (`int`, defaults to 100): Adapts clipping threshold automatically by tracking the last 100 gradient norms and clipping the gradient at a certain percentile to improve stability. block_wise (`bool`, defaults to `True`): Whether to independently quantize each block of tensors to reduce outlier effects and improve stability. rr r r r r rrr)Nrrs r"rAdagrad32bit.__init__sRby6rd;< <l";L>JK Kcz6se<= = $ +NO O s?=> >             r$r%r&r,r3s@r"r;r;r4r$r;N)bitsandbytes.optim.optimizerrrr6r;r%r$r"r@s6 9@ o@ FA /A H@ ?@ r$