Buckets:
| import{s as Rt,n as Bt,o as Jt}from"../chunks/scheduler.8a2cc2fa.js";import{S as Kt,i as Qt,e as b,s as i,c as s,h as Wt,a as f,d as a,b as l,f as v,g as m,j as qt,k as g,l as $,m as n,n as r,t as o,o as d,p}from"../chunks/index.7079e750.js";import{C as Xt,H as K,E as Yt}from"../chunks/MermaidChart.svelte_svelte_type_style_lang.c6103146.js";import{D as _}from"../chunks/Docstring.5b942aec.js";function Zt(Tt){let y,nt,at,it,E,lt,w,st,P,Ct='<a href="https://hf.co/papers/2409.03137" rel="nofollow">AdEMAMix</a> is a variant of the <code>Adam</code> optimizer.',mt,O,Ft="bitsandbytes also supports paged optimizers which take advantage of CUDAs unified memory to transfer memory from the GPU to the CPU when GPU memory is exhausted.",rt,I,ot,u,N,It,Q,L,dt,z,pt,h,D,Nt,W,V,bt,T,ft,c,C,Lt,X,F,vt,U,gt,M,H,zt,Y,S,$t,G,_t,A,j,Dt,Z,k,yt,q,ut,x,R,Vt,tt,B,ht,J,ct,et,Mt;return E=new Xt({props:{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"}}),w=new K({props:{title:"AdEMAMix",local:"ademamix",headingTag:"h1"}}),I=new K({props:{title:"AdEMAMix",local:"api-class ][ bitsandbytes.optim.AdEMAMix",headingTag:"h2"}}),N=new _({props:{name:"class bitsandbytes.optim.AdEMAMix",anchor:"bitsandbytes.optim.AdEMAMix",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L107"}}),L=new _({props:{name:"__init__",anchor:"bitsandbytes.optim.AdEMAMix.__init__",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L108"}}),z=new K({props:{title:"AdEMAMix8bit",local:"bitsandbytes.optim.AdEMAMix8bit",headingTag:"h2"}}),D=new _({props:{name:"class bitsandbytes.optim.AdEMAMix8bit",anchor:"bitsandbytes.optim.AdEMAMix8bit",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L270"}}),V=new _({props:{name:"__init__",anchor:"bitsandbytes.optim.AdEMAMix8bit.__init__",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L271"}}),T=new K({props:{title:"AdEMAMix32bit",local:"bitsandbytes.optim.AdEMAMix32bit",headingTag:"h2"}}),C=new _({props:{name:"class bitsandbytes.optim.AdEMAMix32bit",anchor:"bitsandbytes.optim.AdEMAMix32bit",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L355"}}),F=new _({props:{name:"__init__",anchor:"bitsandbytes.optim.AdEMAMix32bit.__init__",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L356"}}),U=new K({props:{title:"PagedAdEMAMix",local:"bitsandbytes.optim.PagedAdEMAMix",headingTag:"h2"}}),H=new _({props:{name:"class bitsandbytes.optim.PagedAdEMAMix",anchor:"bitsandbytes.optim.PagedAdEMAMix",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L326"}}),S=new _({props:{name:"__init__",anchor:"bitsandbytes.optim.PagedAdEMAMix.__init__",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L327"}}),G=new K({props:{title:"PagedAdEMAMix8bit",local:"bitsandbytes.optim.PagedAdEMAMix8bit",headingTag:"h2"}}),j=new _({props:{name:"class bitsandbytes.optim.PagedAdEMAMix8bit",anchor:"bitsandbytes.optim.PagedAdEMAMix8bit",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L299"}}),k=new _({props:{name:"__init__",anchor:"bitsandbytes.optim.PagedAdEMAMix8bit.__init__",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L300"}}),q=new K({props:{title:"PagedAdEMAMix32bit",local:"bitsandbytes.optim.PagedAdEMAMix32bit",headingTag:"h2"}}),R=new _({props:{name:"class bitsandbytes.optim.PagedAdEMAMix32bit",anchor:"bitsandbytes.optim.PagedAdEMAMix32bit",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L386"}}),B=new _({props:{name:"__init__",anchor:"bitsandbytes.optim.PagedAdEMAMix32bit.__init__",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"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1920/bitsandbytes/optim/ademamix.py#L387"}}),J=new Yt({props:{source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/main/docs/source/reference/optim/ademamix.mdx"}}),{c(){y=b("meta"),nt=i(),at=b("p"),it=i(),s(E.$$.fragment),lt=i(),s(w.$$.fragment),st=i(),P=b("p"),P.innerHTML=Ct,mt=i(),O=b("p"),O.textContent=Ft,rt=i(),s(I.$$.fragment),ot=i(),u=b("div"),s(N.$$.fragment),It=i(),Q=b("div"),s(L.$$.fragment),dt=i(),s(z.$$.fragment),pt=i(),h=b("div"),s(D.$$.fragment),Nt=i(),W=b("div"),s(V.$$.fragment),bt=i(),s(T.$$.fragment),ft=i(),c=b("div"),s(C.$$.fragment),Lt=i(),X=b("div"),s(F.$$.fragment),vt=i(),s(U.$$.fragment),gt=i(),M=b("div"),s(H.$$.fragment),zt=i(),Y=b("div"),s(S.$$.fragment),$t=i(),s(G.$$.fragment),_t=i(),A=b("div"),s(j.$$.fragment),Dt=i(),Z=b("div"),s(k.$$.fragment),yt=i(),s(q.$$.fragment),ut=i(),x=b("div"),s(R.$$.fragment),Vt=i(),tt=b("div"),s(B.$$.fragment),ht=i(),s(J.$$.fragment),ct=i(),et=b("p"),this.h()},l(t){const e=Wt("svelte-u9bgzb",document.head);y=f(e,"META",{name:!0,content:!0}),e.forEach(a),nt=l(t),at=f(t,"P",{}),v(at).forEach(a),it=l(t),m(E.$$.fragment,t),lt=l(t),m(w.$$.fragment,t),st=l(t),P=f(t,"P",{"data-svelte-h":!0}),qt(P)!=="svelte-1holr5j"&&(P.innerHTML=Ct),mt=l(t),O=f(t,"P",{"data-svelte-h":!0}),qt(O)!=="svelte-qpasov"&&(O.textContent=Ft),rt=l(t),m(I.$$.fragment,t),ot=l(t),u=f(t,"DIV",{class:!0});var At=v(u);m(N.$$.fragment,At),It=l(At),Q=f(At,"DIV",{class:!0});var Ut=v(Q);m(L.$$.fragment,Ut),Ut.forEach(a),At.forEach(a),dt=l(t),m(z.$$.fragment,t),pt=l(t),h=f(t,"DIV",{class:!0});var xt=v(h);m(D.$$.fragment,xt),Nt=l(xt),W=f(xt,"DIV",{class:!0});var Ht=v(W);m(V.$$.fragment,Ht),Ht.forEach(a),xt.forEach(a),bt=l(t),m(T.$$.fragment,t),ft=l(t),c=f(t,"DIV",{class:!0});var Et=v(c);m(C.$$.fragment,Et),Lt=l(Et),X=f(Et,"DIV",{class:!0});var St=v(X);m(F.$$.fragment,St),St.forEach(a),Et.forEach(a),vt=l(t),m(U.$$.fragment,t),gt=l(t),M=f(t,"DIV",{class:!0});var wt=v(M);m(H.$$.fragment,wt),zt=l(wt),Y=f(wt,"DIV",{class:!0});var Gt=v(Y);m(S.$$.fragment,Gt),Gt.forEach(a),wt.forEach(a),$t=l(t),m(G.$$.fragment,t),_t=l(t),A=f(t,"DIV",{class:!0});var Pt=v(A);m(j.$$.fragment,Pt),Dt=l(Pt),Z=f(Pt,"DIV",{class:!0});var jt=v(Z);m(k.$$.fragment,jt),jt.forEach(a),Pt.forEach(a),yt=l(t),m(q.$$.fragment,t),ut=l(t),x=f(t,"DIV",{class:!0});var Ot=v(x);m(R.$$.fragment,Ot),Vt=l(Ot),tt=f(Ot,"DIV",{class:!0});var kt=v(tt);m(B.$$.fragment,kt),kt.forEach(a),Ot.forEach(a),ht=l(t),m(J.$$.fragment,t),ct=l(t),et=f(t,"P",{}),v(et).forEach(a),this.h()},h(){g(y,"name","hf:doc:metadata"),g(y,"content",ta),g(Q,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(u,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(W,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(h,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(X,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(c,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(Y,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(M,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(Z,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(A,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(tt,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),g(x,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8")},m(t,e){$(document.head,y),n(t,nt,e),n(t,at,e),n(t,it,e),r(E,t,e),n(t,lt,e),r(w,t,e),n(t,st,e),n(t,P,e),n(t,mt,e),n(t,O,e),n(t,rt,e),r(I,t,e),n(t,ot,e),n(t,u,e),r(N,u,null),$(u,It),$(u,Q),r(L,Q,null),n(t,dt,e),r(z,t,e),n(t,pt,e),n(t,h,e),r(D,h,null),$(h,Nt),$(h,W),r(V,W,null),n(t,bt,e),r(T,t,e),n(t,ft,e),n(t,c,e),r(C,c,null),$(c,Lt),$(c,X),r(F,X,null),n(t,vt,e),r(U,t,e),n(t,gt,e),n(t,M,e),r(H,M,null),$(M,zt),$(M,Y),r(S,Y,null),n(t,$t,e),r(G,t,e),n(t,_t,e),n(t,A,e),r(j,A,null),$(A,Dt),$(A,Z),r(k,Z,null),n(t,yt,e),r(q,t,e),n(t,ut,e),n(t,x,e),r(R,x,null),$(x,Vt),$(x,tt),r(B,tt,null),n(t,ht,e),r(J,t,e),n(t,ct,e),n(t,et,e),Mt=!0},p:Bt,i(t){Mt||(o(E.$$.fragment,t),o(w.$$.fragment,t),o(I.$$.fragment,t),o(N.$$.fragment,t),o(L.$$.fragment,t),o(z.$$.fragment,t),o(D.$$.fragment,t),o(V.$$.fragment,t),o(T.$$.fragment,t),o(C.$$.fragment,t),o(F.$$.fragment,t),o(U.$$.fragment,t),o(H.$$.fragment,t),o(S.$$.fragment,t),o(G.$$.fragment,t),o(j.$$.fragment,t),o(k.$$.fragment,t),o(q.$$.fragment,t),o(R.$$.fragment,t),o(B.$$.fragment,t),o(J.$$.fragment,t),Mt=!0)},o(t){d(E.$$.fragment,t),d(w.$$.fragment,t),d(I.$$.fragment,t),d(N.$$.fragment,t),d(L.$$.fragment,t),d(z.$$.fragment,t),d(D.$$.fragment,t),d(V.$$.fragment,t),d(T.$$.fragment,t),d(C.$$.fragment,t),d(F.$$.fragment,t),d(U.$$.fragment,t),d(H.$$.fragment,t),d(S.$$.fragment,t),d(G.$$.fragment,t),d(j.$$.fragment,t),d(k.$$.fragment,t),d(q.$$.fragment,t),d(R.$$.fragment,t),d(B.$$.fragment,t),d(J.$$.fragment,t),Mt=!1},d(t){t&&(a(nt),a(at),a(it),a(lt),a(st),a(P),a(mt),a(O),a(rt),a(ot),a(u),a(dt),a(pt),a(h),a(bt),a(ft),a(c),a(vt),a(gt),a(M),a($t),a(_t),a(A),a(yt),a(ut),a(x),a(ht),a(ct),a(et)),a(y),p(E,t),p(w,t),p(I,t),p(N),p(L),p(z,t),p(D),p(V),p(T,t),p(C),p(F),p(U,t),p(H),p(S),p(G,t),p(j),p(k),p(q,t),p(R),p(B),p(J,t)}}}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}';function aa(Tt){return Jt(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class sa extends Kt{constructor(y){super(),Qt(this,y,aa,Zt,Rt,{})}}export{sa as component}; | |
Xet Storage Details
- Size:
- 15.7 kB
- Xet hash:
- f4e6dda5f87728bfceff60529c59be440556cc26dfbcfb24fbe8cc7887f7bb11
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.