Buckets:

rtrm's picture
download
raw
15 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 f,u as m,x as jt,k as $,y as _,a as n,v as r,d as o,t as d,w as b}from"../chunks/index.28275fd3.js";import{D as g}from"../chunks/Docstring.ee6c313e.js";import{H as J,E as Qt}from"../chunks/EditOnGithub.582011f0.js";function Wt(zt){let u,et,tt,nt,E,it,w,Dt='<a href="https://hf.co/papers/2409.03137" rel="nofollow">AdEMAMix</a> is a variant of the <code>Adam</code> optimizer.',lt,P,Vt="bitsandbytes also supports paged optimizers which take advantage of CUDAs unified memory to transfer memory from the GPU to the CPU when GPU memory is exhausted.",st,O,mt,y,I,Pt,K,N,rt,T,ot,h,L,Ot,Q,z,dt,D,bt,c,V,It,W,C,pt,F,vt,A,U,Nt,X,H,ft,G,$t,M,S,Tt,Y,j,_t,k,gt,x,q,Lt,Z,R,ut,B,yt,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 g({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:": Optional = None"},{name:"t_beta3",val:": Optional = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": Literal = 32"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1386/bitsandbytes/optim/ademamix.py#L106"}}),N=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": Literal = 32"},{name:"min_8bit_size",val:": int = 4096"},{name:"is_paged",val:": bool = False"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1386/bitsandbytes/optim/ademamix.py#L107"}}),T=new J({props:{title:"AdEMAMix8bit",local:"bitsandbytes.optim.AdEMAMix8bit",headingTag:"h2"}}),L=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L272"}}),z=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L273"}}),D=new J({props:{title:"AdEMAMix32bit",local:"bitsandbytes.optim.AdEMAMix32bit",headingTag:"h2"}}),V=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L357"}}),C=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L358"}}),F=new J({props:{title:"PagedAdEMAMix",local:"bitsandbytes.optim.PagedAdEMAMix",headingTag:"h2"}}),U=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": Literal = 32"},{name:"min_8bit_size",val:": int = 4096"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1386/bitsandbytes/optim/ademamix.py#L328"}}),H=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = None"},{name:"eps",val:": float = 1e-08"},{name:"weight_decay",val:": float = 0.01"},{name:"optim_bits",val:": Literal = 32"},{name:"min_8bit_size",val:": int = 4096"}],source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/vr_1386/bitsandbytes/optim/ademamix.py#L329"}}),G=new J({props:{title:"PagedAdEMAMix8bit",local:"bitsandbytes.optim.PagedAdEMAMix8bit",headingTag:"h2"}}),S=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L301"}}),j=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L302"}}),k=new J({props:{title:"PagedAdEMAMix32bit",local:"bitsandbytes.optim.PagedAdEMAMix32bit",headingTag:"h2"}}),q=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L390"}}),R=new g({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:": Optional = None"},{name:"t_beta3",val:": Optional = 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_1386/bitsandbytes/optim/ademamix.py#L391"}}),B=new Qt({props:{source:"https://github.com/bitsandbytes-foundation/bitsandbytes/blob/main/docs/source/reference/optim/ademamix.mdx"}}),{c(){u=p("meta"),et=i(),tt=p("p"),nt=i(),s(E.$$.fragment),it=i(),w=p("p"),w.innerHTML=Dt,lt=i(),P=p("p"),P.textContent=Vt,st=i(),s(O.$$.fragment),mt=i(),y=p("div"),s(I.$$.fragment),Pt=i(),K=p("div"),s(N.$$.fragment),rt=i(),s(T.$$.fragment),ot=i(),h=p("div"),s(L.$$.fragment),Ot=i(),Q=p("div"),s(z.$$.fragment),dt=i(),s(D.$$.fragment),bt=i(),c=p("div"),s(V.$$.fragment),It=i(),W=p("div"),s(C.$$.fragment),pt=i(),s(F.$$.fragment),vt=i(),A=p("div"),s(U.$$.fragment),Nt=i(),X=p("div"),s(H.$$.fragment),ft=i(),s(G.$$.fragment),$t=i(),M=p("div"),s(S.$$.fragment),Tt=i(),Y=p("div"),s(j.$$.fragment),_t=i(),s(k.$$.fragment),gt=i(),x=p("div"),s(q.$$.fragment),Lt=i(),Z=p("div"),s(R.$$.fragment),ut=i(),s(B.$$.fragment),yt=i(),at=p("p"),this.h()},l(t){const e=Kt("svelte-u9bgzb",document.head);u=v(e,"META",{name:!0,content:!0}),e.forEach(a),et=l(t),tt=v(t,"P",{}),f(tt).forEach(a),nt=l(t),m(E.$$.fragment,t),it=l(t),w=v(t,"P",{"data-svelte-h":!0}),jt(w)!=="svelte-1holr5j"&&(w.innerHTML=Dt),lt=l(t),P=v(t,"P",{"data-svelte-h":!0}),jt(P)!=="svelte-qpasov"&&(P.textContent=Vt),st=l(t),m(O.$$.fragment,t),mt=l(t),y=v(t,"DIV",{class:!0});var ct=f(y);m(I.$$.fragment,ct),Pt=l(ct),K=v(ct,"DIV",{class:!0});var Ct=f(K);m(N.$$.fragment,Ct),Ct.forEach(a),ct.forEach(a),rt=l(t),m(T.$$.fragment,t),ot=l(t),h=v(t,"DIV",{class:!0});var At=f(h);m(L.$$.fragment,At),Ot=l(At),Q=v(At,"DIV",{class:!0});var Ft=f(Q);m(z.$$.fragment,Ft),Ft.forEach(a),At.forEach(a),dt=l(t),m(D.$$.fragment,t),bt=l(t),c=v(t,"DIV",{class:!0});var Mt=f(c);m(V.$$.fragment,Mt),It=l(Mt),W=v(Mt,"DIV",{class:!0});var Ut=f(W);m(C.$$.fragment,Ut),Ut.forEach(a),Mt.forEach(a),pt=l(t),m(F.$$.fragment,t),vt=l(t),A=v(t,"DIV",{class:!0});var xt=f(A);m(U.$$.fragment,xt),Nt=l(xt),X=v(xt,"DIV",{class:!0});var Ht=f(X);m(H.$$.fragment,Ht),Ht.forEach(a),xt.forEach(a),ft=l(t),m(G.$$.fragment,t),$t=l(t),M=v(t,"DIV",{class:!0});var Et=f(M);m(S.$$.fragment,Et),Tt=l(Et),Y=v(Et,"DIV",{class:!0});var Gt=f(Y);m(j.$$.fragment,Gt),Gt.forEach(a),Et.forEach(a),_t=l(t),m(k.$$.fragment,t),gt=l(t),x=v(t,"DIV",{class:!0});var wt=f(x);m(q.$$.fragment,wt),Lt=l(wt),Z=v(wt,"DIV",{class:!0});var St=f(Z);m(R.$$.fragment,St),St.forEach(a),wt.forEach(a),ut=l(t),m(B.$$.fragment,t),yt=l(t),at=v(t,"P",{}),f(at).forEach(a),this.h()},h(){$(u,"name","hf:doc:metadata"),$(u,"content",Xt),$(K,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(y,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(Q,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(h,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(W,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(c,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(X,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(A,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(Y,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(M,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(Z,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(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,u),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,y,e),r(I,y,null),_(y,Pt),_(y,K),r(N,K,null),n(t,rt,e),r(T,t,e),n(t,ot,e),n(t,h,e),r(L,h,null),_(h,Ot),_(h,Q),r(z,Q,null),n(t,dt,e),r(D,t,e),n(t,bt,e),n(t,c,e),r(V,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,ft,e),r(G,t,e),n(t,$t,e),n(t,M,e),r(S,M,null),_(M,Tt),_(M,Y),r(j,Y,null),n(t,_t,e),r(k,t,e),n(t,gt,e),n(t,x,e),r(q,x,null),_(x,Lt),_(x,Z),r(R,Z,null),n(t,ut,e),r(B,t,e),n(t,yt,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(T.$$.fragment,t),o(L.$$.fragment,t),o(z.$$.fragment,t),o(D.$$.fragment,t),o(V.$$.fragment,t),o(C.$$.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(T.$$.fragment,t),d(L.$$.fragment,t),d(z.$$.fragment,t),d(D.$$.fragment,t),d(V.$$.fragment,t),d(C.$$.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(y),a(rt),a(ot),a(h),a(dt),a(bt),a(c),a(pt),a(vt),a(A),a(ft),a($t),a(M),a(_t),a(gt),a(x),a(ut),a(yt),a(at)),a(u),b(E,t),b(O,t),b(I),b(N),b(T,t),b(L),b(z),b(D,t),b(V),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(zt){return Rt(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class na extends Bt{constructor(u){super(),Jt(this,u,Yt,Wt,kt,{})}}export{na as component};

Xet Storage Details

Size:
15 kB
·
Xet hash:
c7ee36bda7fbcd3947c428c3c63fe858af41c47278e9d79b4822540acf69eb4f

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