Buckets:

HuggingFaceDocBuilder's picture
download
raw
11.6 kB
import"../chunks/DsnmJJEf.js";import{i as J,h as K,C as Q,H as n,D as t,E as V,s as W}from"../chunks/DT0OpeMJ.js";import{p as X,o as Y,s as a,f as Z,a as U,b as $,c as e,d as C,r as i,n as aa}from"../chunks/Bb-LL0eD.js";const ta='{"title":"AdEMAMix","local":"ademamix","sections":[{"title":"AdEMAMix","local":"api-class ][ bitsandbytes.optim.AdEMAMix","sections":[],"depth":2},{"title":"AdEMAMix8bit","local":"bitsandbytes.optim.AdEMAMix8bit","sections":[],"depth":2},{"title":"AdEMAMix32bit","local":"bitsandbytes.optim.AdEMAMix32bit","sections":[],"depth":2},{"title":"PagedAdEMAMix","local":"bitsandbytes.optim.PagedAdEMAMix","sections":[],"depth":2},{"title":"PagedAdEMAMix8bit","local":"bitsandbytes.optim.PagedAdEMAMix8bit","sections":[],"depth":2},{"title":"PagedAdEMAMix32bit","local":"bitsandbytes.optim.PagedAdEMAMix32bit","sections":[],"depth":2}],"depth":1}';var ea=C('<meta name="hf:doc:metadata"/>'),ia=C('<p></p> <!> <!> <p><a href="https://hf.co/papers/2409.03137" rel="nofollow">AdEMAMix</a> is a variant of the <code>Adam</code> optimizer.</p> <p>bitsandbytes also supports paged optimizers which take advantage of CUDAs unified memory to transfer memory from the GPU to the CPU when GPU memory is exhausted.</p> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div></div> <!> <p></p>',1);function sa(D,G){X(G,!1),Y(()=>{new URLSearchParams(window.location.search).get("fw")}),J();var r=ia();K("6ykfaz",T=>{var F=ea();W(F,"content",ta),U(T,F)});var p=a(Z(r),2);Q(p,{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"});var v=a(p,2);n(v,{title:"AdEMAMix",local:"ademamix",headingTag:"h1"});var _=a(v,6);n(_,{title:"AdEMAMix",local:"api-class ][ bitsandbytes.optim.AdEMAMix",headingTag:"h2"});var l=a(_,2),y=e(l);t(y,{name:"class bitsandbytes.optim.AdEMAMix",anchor:"bitsandbytes.optim.AdEMAMix",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L107",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": typing.Literal[8, 32] = 32"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}]});var g=a(y,2),k=e(g);t(k,{name:"__init__",anchor:"bitsandbytes.optim.AdEMAMix.__init__",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L108",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": typing.Literal[8, 32] = 32"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}]}),i(g),i(l);var c=a(l,2);n(c,{title:"AdEMAMix8bit",local:"bitsandbytes.optim.AdEMAMix8bit",headingTag:"h2"});var o=a(c,2),h=e(o);t(h,{name:"class bitsandbytes.optim.AdEMAMix8bit",anchor:"bitsandbytes.optim.AdEMAMix8bit",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L270",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}]});var M=a(h,2),H=e(M);t(H,{name:"__init__",anchor:"bitsandbytes.optim.AdEMAMix8bit.__init__",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L271",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}]}),i(M),i(o);var A=a(o,2);n(A,{title:"AdEMAMix32bit",local:"bitsandbytes.optim.AdEMAMix32bit",headingTag:"h2"});var s=a(A,2),f=e(s);t(f,{name:"class bitsandbytes.optim.AdEMAMix32bit",anchor:"bitsandbytes.optim.AdEMAMix32bit",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L355",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}]});var x=a(f,2),S=e(x);t(S,{name:"__init__",anchor:"bitsandbytes.optim.AdEMAMix32bit.__init__",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L356",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}]}),i(x),i(s);var u=a(s,2);n(u,{title:"PagedAdEMAMix",local:"bitsandbytes.optim.PagedAdEMAMix",headingTag:"h2"});var d=a(u,2),E=e(d);t(E,{name:"class bitsandbytes.optim.PagedAdEMAMix",anchor:"bitsandbytes.optim.PagedAdEMAMix",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L326",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": typing.Literal[8, 32] = 32"},{name:"min_8bit_size",val:": int = 4096"}]});var P=a(E,2),R=e(P);t(R,{name:"__init__",anchor:"bitsandbytes.optim.PagedAdEMAMix.__init__",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L327",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": typing.Literal[8, 32] = 32"},{name:"min_8bit_size",val:": int = 4096"}]}),i(P),i(d);var O=a(d,2);n(O,{title:"PagedAdEMAMix8bit",local:"bitsandbytes.optim.PagedAdEMAMix8bit",headingTag:"h2"});var b=a(O,2),N=e(b);t(N,{name:"class bitsandbytes.optim.PagedAdEMAMix8bit",anchor:"bitsandbytes.optim.PagedAdEMAMix8bit",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L299",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"}]});var w=a(N,2),j=e(w);t(j,{name:"__init__",anchor:"bitsandbytes.optim.PagedAdEMAMix8bit.__init__",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L300",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"}]}),i(w),i(b);var L=a(b,2);n(L,{title:"PagedAdEMAMix32bit",local:"bitsandbytes.optim.PagedAdEMAMix32bit",headingTag:"h2"});var m=a(L,2),z=e(m);t(z,{name:"class bitsandbytes.optim.PagedAdEMAMix32bit",anchor:"bitsandbytes.optim.PagedAdEMAMix32bit",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L386",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"}]});var I=a(z,2),q=e(I);t(q,{name:"__init__",anchor:"bitsandbytes.optim.PagedAdEMAMix32bit.__init__",source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_2008/bitsandbytes/optim/ademamix.py#L387",parameters:[{name:"params",val:": Iterable"},{name:"lr",val:": float = 0.001"},{name:"betas",val:": tuple = (0.9, 0.999, 0.9999)"},{name:"alpha",val:": float = 5.0"},{name:"t_alpha",val:": typing.Optional[int] = None"},{name:"t_beta3",val:": typing.Optional[int] = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"min_8bit_size",val:": int = 4096"}]}),i(I),i(m);var B=a(m,2);V(B,{source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/main/docs/source/reference/optim/ademamix.mdx"}),aa(2),U(D,r),$()}export{sa as component};

Xet Storage Details

Size:
11.6 kB
·
Xet hash:
7fb8ea3deb7df4b040662a3d582570fb3c5652fecc90abb25a0bd09d9d74ddb3

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.