Team Ai
Datasetpublic

Brunobkr/llama.cpp_AlgMor24_github

ΩFFFΣLLIa • llama.cpp • AlgMor24 ██████╗ ███████╗███████╗███████╗██╗ ██╗ ██╗ █████╗ ██╔═══██╗██╔════╝██╔════╝██╔════╝██║ ██║ ██║██╔══██╗ ██║ ██║█████╗ █████╗ █████╗ ██║ ██║ ██║███████║ ██║ ██║██╔══╝ ██╔══╝ ██╔══╝ ██║ ██║ ██║██╔══██║ ╚██████╔╝██║ ██║ ███████╗███████╗███████╗██║██║ ██║ ╚═════╝ ╚═╝ ╚═╝ ╚══════╝╚══════╝╚══════╝╚═╝╚═╝ ╚═╝ High-Performance LLM / VLM Inference & Autonomous Agentic Ecosystem… See the full description on the dataset page: https://huggingface.co/datasets/Brunobkr/llama.cpp_AlgMor24_github.

sourceHugging Faceupdated 2mo agoView on Hugging Face
0likes3.1kdownloads
log.js141 linesDownload Raw Back to src
1import {ticks} from "d3-array";2import {format, formatSpecifier} from "d3-format";3import nice from "./nice.js";4import {copy, transformer} from "./continuous.js";5import {initRange} from "./init.js";6 7function transformLog(x) {8  return Math.log(x);9}10 11function transformExp(x) {12  return Math.exp(x);13}14 15function transformLogn(x) {16  return -Math.log(-x);17}18 19function transformExpn(x) {20  return -Math.exp(-x);21}22 23function pow10(x) {24  return isFinite(x) ? +("1e" + x) : x < 0 ? 0 : x;25}26 27function powp(base) {28  return base === 10 ? pow1029      : base === Math.E ? Math.exp30      : x => Math.pow(base, x);31}32 33function logp(base) {34  return base === Math.E ? Math.log35      : base === 10 && Math.log1036      || base === 2 && Math.log237      || (base = Math.log(base), x => Math.log(x) / base);38}39 40function reflect(f) {41  return (x, k) => -f(-x, k);42}43 44export function loggish(transform) {45  const scale = transform(transformLog, transformExp);46  const domain = scale.domain;47  let base = 10;48  let logs;49  let pows;50 51  function rescale() {52    logs = logp(base), pows = powp(base);53    if (domain()[0] < 0) {54      logs = reflect(logs), pows = reflect(pows);55      transform(transformLogn, transformExpn);56    } else {57      transform(transformLog, transformExp);58    }59    return scale;60  }61 62  scale.base = function(_) {63    return arguments.length ? (base = +_, rescale()) : base;64  };65 66  scale.domain = function(_) {67    return arguments.length ? (domain(_), rescale()) : domain();68  };69 70  scale.ticks = count => {71    const d = domain();72    let u = d[0];73    let v = d[d.length - 1];74    const r = v < u;75 76    if (r) ([u, v] = [v, u]);77 78    let i = logs(u);79    let j = logs(v);80    let k;81    let t;82    const n = count == null ? 10 : +count;83    let z = [];84 85    if (!(base % 1) && j - i < n) {86      i = Math.floor(i), j = Math.ceil(j);87      if (u > 0) for (; i <= j; ++i) {88        for (k = 1; k < base; ++k) {89          t = i < 0 ? k / pows(-i) : k * pows(i);90          if (t < u) continue;91          if (t > v) break;92          z.push(t);93        }94      } else for (; i <= j; ++i) {95        for (k = base - 1; k >= 1; --k) {96          t = i > 0 ? k / pows(-i) : k * pows(i);97          if (t < u) continue;98          if (t > v) break;99          z.push(t);100        }101      }102      if (z.length * 2 < n) z = ticks(u, v, n);103    } else {104      z = ticks(i, j, Math.min(j - i, n)).map(pows);105    }106    return r ? z.reverse() : z;107  };108 109  scale.tickFormat = (count, specifier) => {110    if (count == null) count = 10;111    if (specifier == null) specifier = base === 10 ? "s" : ",";112    if (typeof specifier !== "function") {113      if (!(base % 1) && (specifier = formatSpecifier(specifier)).precision == null) specifier.trim = true;114      specifier = format(specifier);115    }116    if (count === Infinity) return specifier;117    const k = Math.max(1, base * count / scale.ticks().length); // TODO fast estimate?118    return d => {119      let i = d / pows(Math.round(logs(d)));120      if (i * base < base - 0.5) i *= base;121      return i <= k ? specifier(d) : "";122    };123  };124 125  scale.nice = () => {126    return domain(nice(domain(), {127      floor: x => pows(Math.floor(logs(x))),128      ceil: x => pows(Math.ceil(logs(x)))129    }));130  };131 132  return scale;133}134 135export default function log() {136  const scale = loggish(transformer()).domain([1, 10]);137  scale.copy = () => copy(scale, log()).base(scale.base());138  initRange.apply(scale, arguments);139  return scale;140}141 
Brunobkr/llama.cpp_AlgMor24_github · Team Ai