biRddlmZGddeZGddeZGddeZy) )Optimizer1Statec4eZdZ dfd ZxZS)Adagradc d|kstd|d|kstd|d|kstd||dk7r td|dk7r tdt | d||d||||| | | y ) 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 U/home/cdr/jupyterlab/.venv/lib/python3.12/site-packages/bitsandbytes/optim/adagrad.pyrzAdagrad.__init__ sRby6rd;< <l";L>JK Kcz6se<= = $ +NO O s?=> >             {Gz?rrr绽|= NdT__name__ __module__ __qualname__r __classcell__r s@r!rr- "#  ? ? r"rc4eZdZ dfd ZxZS) Adagrad8bitc d|kstd|d|kstd|d|kstd||dk7r td|dk7r td| sJt | d||d||d || | | y ) 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. rrr r r r r rNrrs r!rzAdagrad8bit.__init__LsRby6rd;< <l";L>JK Kcz6se<= = $ +NO O s?=> >z            r") r$rrrr%r3Nr'r(Tr)r.s@r!r1r1Ks/ "#  @ @ r"r1c4eZdZ dfd ZxZS) Adagrad32bitc d|kstd|d|kstd|d|kstd||dk7r td|dk7r tdt | d||d||d || | | y ) 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. rrr r r r r rr&Nrrs r!rzAdagrad32bit.__init__sRby6rd;< <l";L>JK Kcz6se<= = $ +NO O s?=> >             r"r#r)r.s@r!r5r5r/r"r5N)bitsandbytes.optim.optimizerrrr1r5r"r!r9s6 9@ o@ FA /A H@ ?@ r"