zhangyiss/mlx
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx.git synced 2025-09-17 09:18:12 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

docs

Browse Source
This commit is contained in:
Awni Hannun
2024-05-23 21:11:24 -07:00
committed by CircleCI Docs
parent 20469ce2d1
commit d7a78fbe2b
1837 changed files with 6038 additions and 255894 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
docs/build/html/install.html vendored
View File

@@ -8,7 +8,7 @@
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
<title>Build and Install &#8212; MLX 0.13.1 documentation</title>
<title>Build and Install &#8212; MLX 0.14.0 documentation</title>
@@ -36,7 +36,7 @@
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=5b4479735964841361fd" />
<script src="_static/vendor/fontawesome/6.1.2/js/all.min.js?digest=5b4479735964841361fd"></script>
<script src="_static/documentation_options.js?v=aec58b01"></script>
<script src="_static/documentation_options.js?v=8eac8a5b"></script>
<script src="_static/doctools.js?v=888ff710"></script>
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
<script src="_static/scripts/sphinx-book-theme.js?v=efea14e4"></script>
@@ -131,8 +131,8 @@
<img src="_static/mlx_logo.png" class="logo__image only-light" alt="MLX 0.13.1 documentation - Home"/>
<script>document.write(`<img src="_static/mlx_logo_dark.png" class="logo__image only-dark" alt="MLX 0.13.1 documentation - Home"/>`);</script>
<img src="_static/mlx_logo.png" class="logo__image only-light" alt="MLX 0.14.0 documentation - Home"/>
<script>document.write(`<img src="_static/mlx_logo_dark.png" class="logo__image only-dark" alt="MLX 0.14.0 documentation - Home"/>`);</script>
</a></div>
@@ -271,7 +271,6 @@
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.bitwise_or.html">mlx.core.bitwise_or</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.bitwise_xor.html">mlx.core.bitwise_xor</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.block_masked_mm.html">mlx.core.block_masked_mm</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.block_sparse_mm.html">mlx.core.block_sparse_mm</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.broadcast_to.html">mlx.core.broadcast_to</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.ceil.html">mlx.core.ceil</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.clip.html">mlx.core.clip</a></li>
@@ -305,6 +304,8 @@
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.floor.html">mlx.core.floor</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.floor_divide.html">mlx.core.floor_divide</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.full.html">mlx.core.full</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.gather_mm.html">mlx.core.gather_mm</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.gather_qmm.html">mlx.core.gather_qmm</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.greater.html">mlx.core.greater</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.greater_equal.html">mlx.core.greater_equal</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.identity.html">mlx.core.identity</a></li>
@@ -385,6 +386,7 @@
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.tensordot.html">mlx.core.tensordot</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.tile.html">mlx.core.tile</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.topk.html">mlx.core.topk</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.trace.html">mlx.core.trace</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.transpose.html">mlx.core.transpose</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.tri.html">mlx.core.tri</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/_autosummary/mlx.core.tril.html">mlx.core.tril</a></li>
@@ -971,6 +973,9 @@ should point to the path to the built metal library.</p>
<tr class="row-even"><td><p>MLX_BUILD_GGUF</p></td>
<td><p>ON</p></td>
</tr>
<tr class="row-odd"><td><p>MLX_METAL_JIT</p></td>
<td><p>OFF</p></td>
</tr>
</tbody>
</table>
<div class="admonition note">
@@ -997,11 +1002,19 @@ GGUF, you can do:</p>
<div class="highlight-shell notranslate"><div class="highlight"><pre><span></span>cmake<span class="w"> </span>..
<span class="w"> </span>-DCMAKE_BUILD_TYPE<span class="o">=</span>MinSizeRel<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>-DBUILD_SHARED_LIBS<span class="o">=</span>ON<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>-DMLX_BUILD_CPU<span class="o">=</span>ON<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>-DMLX_BUILD_CPU<span class="o">=</span>OFF<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>-DMLX_BUILD_SAFETENSORS<span class="o">=</span>OFF<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>-DMLX_BUILD_GGUF<span class="o">=</span>OFF
<span class="w"> </span>-DMLX_BUILD_GGUF<span class="o">=</span>OFF<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>-DMLX_METAL_JIT<span class="o">=</span>ON
</pre></div>
</div>
<p>THE <cite>MLX_METAL_JIT</cite> flag minimizes the size of the MLX Metal library which
contains pre-built GPU kernels. This substantially reduces the size of the
Metal library by run-time compiling kernels the first time they are used in MLX
on a given machine. Note run-time compilation incurs a cold-start cost which can
be anwywhere from a few hundred millisecond to a few seconds depending on the
application. Once a kernel is compiled, it will be cached by the system. The
Metal kernel cache persists accross reboots.</p>
</section>
</section>
<section id="id2">