Buckets:

rtrm's picture
download
raw
15.4 kB
import{s as kt,n as qt,o as Rt}from"../chunks/scheduler.852ec091.js";import{S as Bt,i as Jt,g as p,s as i,r as s,A as Kt,h as v,f as a,c as l,j as g,u as m,x as jt,k as f,y as $,a as n,v as r,d as o,t as d,w as b}from"../chunks/index.28275fd3.js";import{D as _}from"../chunks/Docstring.f41d6f21.js";import{H as J,E as Qt}from"../chunks/index.6348f961.js";function Wt(Dt){let y,et,tt,nt,E,it,w,Vt='<a href="https://hf.co/papers/2409.03137" rel="nofollow">AdEMAMix</a> is a variant of the <code>Adam</code> optimizer.',lt,P,Tt="bitsandbytes also supports paged optimizers which take advantage of CUDAs unified memory to transfer memory from the GPU to the CPU when GPU memory is exhausted.",st,O,mt,u,I,Pt,K,N,rt,L,ot,h,z,Ot,Q,D,dt,V,bt,c,T,It,W,C,pt,F,vt,A,U,Nt,X,H,gt,G,ft,M,S,Lt,Y,j,$t,k,_t,x,q,zt,Z,R,yt,B,ut,at,ht;return E=new J({props:{title:"AdEMAMix",local:"ademamix",headingTag:"h1"}}),O=new J({props:{title:"AdEMAMix",local:"api-class ][ bitsandbytes.optim.AdEMAMix",headingTag:"h2"}}),I=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_1539/bitsandbytes/optim/ademamix.py#L107"}}),N=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_1539/bitsandbytes/optim/ademamix.py#L108"}}),L=new J({props:{title:"AdEMAMix8bit",local:"bitsandbytes.optim.AdEMAMix8bit",headingTag:"h2"}}),z=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_1539/bitsandbytes/optim/ademamix.py#L274"}}),D=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_1539/bitsandbytes/optim/ademamix.py#L275"}}),V=new J({props:{title:"AdEMAMix32bit",local:"bitsandbytes.optim.AdEMAMix32bit",headingTag:"h2"}}),T=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_1539/bitsandbytes/optim/ademamix.py#L359"}}),C=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_1539/bitsandbytes/optim/ademamix.py#L360"}}),F=new J({props:{title:"PagedAdEMAMix",local:"bitsandbytes.optim.PagedAdEMAMix",headingTag:"h2"}}),U=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_1539/bitsandbytes/optim/ademamix.py#L330"}}),H=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_1539/bitsandbytes/optim/ademamix.py#L331"}}),G=new J({props:{title:"PagedAdEMAMix8bit",local:"bitsandbytes.optim.PagedAdEMAMix8bit",headingTag:"h2"}}),S=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_1539/bitsandbytes/optim/ademamix.py#L303"}}),j=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_1539/bitsandbytes/optim/ademamix.py#L304"}}),k=new J({props:{title:"PagedAdEMAMix32bit",local:"bitsandbytes.optim.PagedAdEMAMix32bit",headingTag:"h2"}}),q=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_1539/bitsandbytes/optim/ademamix.py#L392"}}),R=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_1539/bitsandbytes/optim/ademamix.py#L393"}}),B=new Qt({props:{source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/main/docs/source/reference/optim/ademamix.mdx"}}),{c(){y=p("meta"),et=i(),tt=p("p"),nt=i(),s(E.$$.fragment),it=i(),w=p("p"),w.innerHTML=Vt,lt=i(),P=p("p"),P.textContent=Tt,st=i(),s(O.$$.fragment),mt=i(),u=p("div"),s(I.$$.fragment),Pt=i(),K=p("div"),s(N.$$.fragment),rt=i(),s(L.$$.fragment),ot=i(),h=p("div"),s(z.$$.fragment),Ot=i(),Q=p("div"),s(D.$$.fragment),dt=i(),s(V.$$.fragment),bt=i(),c=p("div"),s(T.$$.fragment),It=i(),W=p("div"),s(C.$$.fragment),pt=i(),s(F.$$.fragment),vt=i(),A=p("div"),s(U.$$.fragment),Nt=i(),X=p("div"),s(H.$$.fragment),gt=i(),s(G.$$.fragment),ft=i(),M=p("div"),s(S.$$.fragment),Lt=i(),Y=p("div"),s(j.$$.fragment),$t=i(),s(k.$$.fragment),_t=i(),x=p("div"),s(q.$$.fragment),zt=i(),Z=p("div"),s(R.$$.fragment),yt=i(),s(B.$$.fragment),ut=i(),at=p("p"),this.h()},l(t){const e=Kt("svelte-u9bgzb",document.head);y=v(e,"META",{name:!0,content:!0}),e.forEach(a),et=l(t),tt=v(t,"P",{}),g(tt).forEach(a),nt=l(t),m(E.$$.fragment,t),it=l(t),w=v(t,"P",{"data-svelte-h":!0}),jt(w)!=="svelte-1holr5j"&&(w.innerHTML=Vt),lt=l(t),P=v(t,"P",{"data-svelte-h":!0}),jt(P)!=="svelte-qpasov"&&(P.textContent=Tt),st=l(t),m(O.$$.fragment,t),mt=l(t),u=v(t,"DIV",{class:!0});var ct=g(u);m(I.$$.fragment,ct),Pt=l(ct),K=v(ct,"DIV",{class:!0});var Ct=g(K);m(N.$$.fragment,Ct),Ct.forEach(a),ct.forEach(a),rt=l(t),m(L.$$.fragment,t),ot=l(t),h=v(t,"DIV",{class:!0});var At=g(h);m(z.$$.fragment,At),Ot=l(At),Q=v(At,"DIV",{class:!0});var Ft=g(Q);m(D.$$.fragment,Ft),Ft.forEach(a),At.forEach(a),dt=l(t),m(V.$$.fragment,t),bt=l(t),c=v(t,"DIV",{class:!0});var Mt=g(c);m(T.$$.fragment,Mt),It=l(Mt),W=v(Mt,"DIV",{class:!0});var Ut=g(W);m(C.$$.fragment,Ut),Ut.forEach(a),Mt.forEach(a),pt=l(t),m(F.$$.fragment,t),vt=l(t),A=v(t,"DIV",{class:!0});var xt=g(A);m(U.$$.fragment,xt),Nt=l(xt),X=v(xt,"DIV",{class:!0});var Ht=g(X);m(H.$$.fragment,Ht),Ht.forEach(a),xt.forEach(a),gt=l(t),m(G.$$.fragment,t),ft=l(t),M=v(t,"DIV",{class:!0});var Et=g(M);m(S.$$.fragment,Et),Lt=l(Et),Y=v(Et,"DIV",{class:!0});var Gt=g(Y);m(j.$$.fragment,Gt),Gt.forEach(a),Et.forEach(a),$t=l(t),m(k.$$.fragment,t),_t=l(t),x=v(t,"DIV",{class:!0});var wt=g(x);m(q.$$.fragment,wt),zt=l(wt),Z=v(wt,"DIV",{class:!0});var St=g(Z);m(R.$$.fragment,St),St.forEach(a),wt.forEach(a),yt=l(t),m(B.$$.fragment,t),ut=l(t),at=v(t,"P",{}),g(at).forEach(a),this.h()},h(){f(y,"name","hf:doc:metadata"),f(y,"content",Xt),f(K,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(u,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(Q,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(h,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(W,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(c,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(X,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(A,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(Y,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(M,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(Z,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),f(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){$(document.head,y),n(t,et,e),n(t,tt,e),n(t,nt,e),r(E,t,e),n(t,it,e),n(t,w,e),n(t,lt,e),n(t,P,e),n(t,st,e),r(O,t,e),n(t,mt,e),n(t,u,e),r(I,u,null),$(u,Pt),$(u,K),r(N,K,null),n(t,rt,e),r(L,t,e),n(t,ot,e),n(t,h,e),r(z,h,null),$(h,Ot),$(h,Q),r(D,Q,null),n(t,dt,e),r(V,t,e),n(t,bt,e),n(t,c,e),r(T,c,null),$(c,It),$(c,W),r(C,W,null),n(t,pt,e),r(F,t,e),n(t,vt,e),n(t,A,e),r(U,A,null),$(A,Nt),$(A,X),r(H,X,null),n(t,gt,e),r(G,t,e),n(t,ft,e),n(t,M,e),r(S,M,null),$(M,Lt),$(M,Y),r(j,Y,null),n(t,$t,e),r(k,t,e),n(t,_t,e),n(t,x,e),r(q,x,null),$(x,zt),$(x,Z),r(R,Z,null),n(t,yt,e),r(B,t,e),n(t,ut,e),n(t,at,e),ht=!0},p:qt,i(t){ht||(o(E.$$.fragment,t),o(O.$$.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.$$.fragment,t),o(G.$$.fragment,t),o(S.$$.fragment,t),o(j.$$.fragment,t),o(k.$$.fragment,t),o(q.$$.fragment,t),o(R.$$.fragment,t),o(B.$$.fragment,t),ht=!0)},o(t){d(E.$$.fragment,t),d(O.$$.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.$$.fragment,t),d(G.$$.fragment,t),d(S.$$.fragment,t),d(j.$$.fragment,t),d(k.$$.fragment,t),d(q.$$.fragment,t),d(R.$$.fragment,t),d(B.$$.fragment,t),ht=!1},d(t){t&&(a(et),a(tt),a(nt),a(it),a(w),a(lt),a(P),a(st),a(mt),a(u),a(rt),a(ot),a(h),a(dt),a(bt),a(c),a(pt),a(vt),a(A),a(gt),a(ft),a(M),a($t),a(_t),a(x),a(yt),a(ut),a(at)),a(y),b(E,t),b(O,t),b(I),b(N),b(L,t),b(z),b(D),b(V,t),b(T),b(C),b(F,t),b(U),b(H),b(G,t),b(S),b(j),b(k,t),b(q),b(R),b(B,t)}}}const Xt='{"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 Yt(Dt){return Rt(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class na extends Bt{constructor(y){super(),Jt(this,y,Yt,Wt,kt,{})}}export{na as component};

Xet Storage Details

Size:
15.4 kB
·
Xet hash:
958f1afb7990e804289c77223448205f7cba39ef098ee23f254ecfaa254f38a2

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