| <!--- |
| Licensed to the Apache Software Foundation (ASF) under one |
| or more contributor license agreements. See the NOTICE file |
| distributed with this work for additional information |
| regarding copyright ownership. The ASF licenses this file |
| to you under the Apache License, Version 2.0 (the |
| "License"); you may not use this file except in compliance |
| with the License. You may obtain a copy of the License at |
| |
| http://www.apache.org/licenses/LICENSE-2.0 |
| |
| Unless required by applicable law or agreed to in writing, |
| software distributed under the License is distributed on an |
| "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
| KIND, either express or implied. See the License for the |
| specific language governing permissions and limitations |
| under the License. |
| --> |
| --- |
| layout: page |
| --- |
| <div class="row"> |
| <div class="col-3 docs-side-bar"> |
| |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| |
| |
| |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| |
| <!-- resource-p --> |
| <!-- page --> |
| </ul> |
| </div> |
| <div class="col-9"> |
| <!--- Licensed to the Apache Software Foundation (ASF) under one --> |
| <!--- or more contributor license agreements. See the NOTICE file --> |
| <!--- distributed with this work for additional information --> |
| <!--- regarding copyright ownership. The ASF licenses this file --> |
| <!--- to you under the Apache License, Version 2.0 (the --> |
| <!--- "License"); you may not use this file except in compliance --> |
| <!--- with the License. You may obtain a copy of the License at --> |
| |
| <!--- http://www.apache.org/licenses/LICENSE-2.0 --> |
| |
| <!--- Unless required by applicable law or agreed to in writing, --> |
| <!--- software distributed under the License is distributed on an --> |
| <!--- "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY --> |
| <!--- KIND, either express or implied. See the License for the --> |
| <!--- specific language governing permissions and limitations --> |
| <!--- under the License. --> |
| |
| <h1 id="ndarray-api">NDArray API</h1> |
| |
| <p>The NDArray API contains tensor operations similar to <code class="highlighter-rouge">numpy.ndarray</code>. The syntax is also similar, except for some additional calls for dealing with I/O and multiple devices.</p> |
| |
| <p>Topics:</p> |
| |
| <ul> |
| <li><a href="#create-ndarray">Create NDArray</a></li> |
| <li><a href="#ndarray-operations">NDArray Operations</a></li> |
| <li><a href="/versions/master/api/clojure/docs/api/org.apache.clojure-mxnet.ndarray-api.html">NDArray API Reference</a></li> |
| </ul> |
| |
| <p>To follow along with this documentation, you can use this namespace with the needed requires:</p> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="nf">ns</span><span class="w"> </span><span class="n">docs.ndarray</span><span class="w"> |
| </span><span class="p">(</span><span class="no">:require</span><span class="w"> </span><span class="p">[</span><span class="n">org.apache.clojure-mxnet.ndarray</span><span class="w"> </span><span class="no">:as</span><span class="w"> </span><span class="n">ndarray</span><span class="p">]</span><span class="w"> |
| </span><span class="p">[</span><span class="n">org.apache.clojure-mxnet.context</span><span class="w"> </span><span class="no">:as</span><span class="w"> </span><span class="n">context</span><span class="p">]))</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <h2 id="create-ndarray">Create NDArray</h2> |
| |
| <p>Create <code class="highlighter-rouge">mxnet.ndarray</code> as follows:</p> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/zeros</span><span class="w"> </span><span class="p">[</span><span class="mi">100</span><span class="w"> </span><span class="mi">50</span><span class="p">]))</span><span class="w"> </span><span class="c1">;;all zero arrray of dimension 100 x 50</span><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">b</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/ones</span><span class="w"> </span><span class="p">[</span><span class="mi">256</span><span class="w"> </span><span class="mi">32</span><span class="w"> </span><span class="mi">128</span><span class="w"> </span><span class="mi">1</span><span class="p">]))</span><span class="w"> </span><span class="c1">;; all one array of dimension</span><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">c</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/array</span><span class="w"> </span><span class="p">[</span><span class="mi">1</span><span class="w"> </span><span class="mi">2</span><span class="w"> </span><span class="mi">3</span><span class="w"> </span><span class="mi">4</span><span class="w"> </span><span class="mi">5</span><span class="w"> </span><span class="mi">6</span><span class="p">]</span><span class="w"> </span><span class="p">[</span><span class="mi">2</span><span class="w"> </span><span class="mi">3</span><span class="p">]))</span><span class="w"> </span><span class="c1">;; array with contents of a shape 2 x 3</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <p>There are also ways to convert a NDArray to a vec or get the shape or the NDArray as an object or vec as follows:</p> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">c</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [1.0 2.0 3.0 4.0 5.0 6.0]</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/shape</span><span class="w"> </span><span class="n">c</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> #object[org.apache.mxnet.Shape 0x583c865 "(2,3)"]</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/shape-vec</span><span class="w"> </span><span class="n">c</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [2 3]</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <h2 id="ndarray-operations">NDArray Operations</h2> |
| |
| <p>There are some basic NDArray operations, like arithmetic and slice operations.</p> |
| |
| <h3 id="arithmetic-operations">Arithmetic Operations</h3> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/ones</span><span class="w"> </span><span class="p">[</span><span class="mi">1</span><span class="w"> </span><span class="mi">5</span><span class="p">]))</span><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">b</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/ones</span><span class="w"> </span><span class="p">[</span><span class="mi">1</span><span class="w"> </span><span class="mi">5</span><span class="p">]))</span><span class="w"> |
| </span><span class="p">(</span><span class="nb">-></span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/+</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="n">b</span><span class="p">)</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="p">))</span><span class="w"> </span><span class="c1">;=> [2.0 2.0 2.0 2.0 2.0]</span><span class="w"> |
| |
| </span><span class="c1">;; original ndarrays are unchanged</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">a</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [1.0 1.0 1.0 1.0 1.0]</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">b</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [1.0 1.0 1.0 1.0 1.0]</span><span class="w"> |
| |
| </span><span class="c1">;;inplace operators</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/+=</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="n">b</span><span class="p">)</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">a</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [2.0 2.0 2.0 2.0 2.0]</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <p>Other arithmetic operations are similar.</p> |
| |
| <h3 id="slice-operations">Slice Operations</h3> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/array</span><span class="w"> </span><span class="p">[</span><span class="mi">1</span><span class="w"> </span><span class="mi">2</span><span class="w"> </span><span class="mi">3</span><span class="w"> </span><span class="mi">4</span><span class="w"> </span><span class="mi">5</span><span class="w"> </span><span class="mi">6</span><span class="p">]</span><span class="w"> </span><span class="p">[</span><span class="mi">3</span><span class="w"> </span><span class="mi">2</span><span class="p">]))</span><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">a1</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/slice</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="mi">1</span><span class="p">))</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/shape-vec</span><span class="w"> </span><span class="n">a1</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [1 2]</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">a1</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [3.0 4.0]</span><span class="w"> |
| |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">a2</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/slice</span><span class="w"> </span><span class="n">a</span><span class="w"> </span><span class="mi">1</span><span class="w"> </span><span class="mi">3</span><span class="p">))</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/shape-vec</span><span class="w"> </span><span class="n">a2</span><span class="p">)</span><span class="w"> </span><span class="c1">;=>[2 2]</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">a2</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [3.0 4.0 5.0 6.0]</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <h3 id="dot-product">Dot Product</h3> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">arr1</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/array</span><span class="w"> </span><span class="p">[</span><span class="mi">1</span><span class="w"> </span><span class="mi">2</span><span class="p">]</span><span class="w"> </span><span class="p">[</span><span class="mi">1</span><span class="w"> </span><span class="mi">2</span><span class="p">]))</span><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">arr2</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/array</span><span class="w"> </span><span class="p">[</span><span class="mi">3</span><span class="w"> </span><span class="mi">4</span><span class="p">]</span><span class="w"> </span><span class="p">[</span><span class="mi">2</span><span class="w"> </span><span class="mi">1</span><span class="p">]))</span><span class="w"> |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">res</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/dot</span><span class="w"> </span><span class="n">arr1</span><span class="w"> </span><span class="n">arr2</span><span class="p">))</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/shape-vec</span><span class="w"> </span><span class="n">res</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [1 1]</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/->vec</span><span class="w"> </span><span class="n">res</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> [11.0]</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <h3 id="save-and-load-ndarray">Save and Load NDArray</h3> |
| |
| <p>You can use MXNet functions to save and load a list or dictionary of NDArrays from file systems, as follows:</p> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="nf">ndarray/save</span><span class="w"> </span><span class="s">"filename"</span><span class="w"> </span><span class="p">{</span><span class="s">"arr1"</span><span class="w"> </span><span class="n">arr1</span><span class="w"> </span><span class="s">"arr2"</span><span class="w"> </span><span class="n">arr2</span><span class="p">})</span><span class="w"> |
| </span><span class="c1">;; you can also do "s3://path" or "hdfs"</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <p>To load:</p> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">from-file</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/load</span><span class="w"> </span><span class="s">"filename"</span><span class="p">))</span><span class="w"> |
| </span><span class="n">from-file</span><span class="w"> |
| </span><span class="c1">;=>{"arr1" #object["org.apache.mxnet.NDArray@43d85753"], "arr2" #object["org.apache.mxnet.NDArray@5c93def4"]}</span><span class="w"> |
| </span></code></pre></div></div> |
| |
| <p>The good thing about using the <code class="highlighter-rouge">save</code> and <code class="highlighter-rouge">load</code> interface is that you can use the format across all <code class="highlighter-rouge">mxnet</code> language bindings. They also already support Amazon S3 and HDFS.</p> |
| |
| <h3 id="multi-device-support">Multi-Device Support</h3> |
| |
| <p>Device information is stored in the <code class="highlighter-rouge">mxnet.Context</code> structure. When creating NDArray in MXNet, you can use the context argument (the default is the CPU context) to create arrays on specific devices as follows:</p> |
| |
| <div class="language-clojure highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">cpu-a</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/zeros</span><span class="w"> </span><span class="p">[</span><span class="mi">100</span><span class="w"> </span><span class="mi">200</span><span class="p">]))</span><span class="w"> |
| </span><span class="p">(</span><span class="nf">ndarray/context</span><span class="w"> </span><span class="n">cpu-a</span><span class="p">)</span><span class="w"> </span><span class="c1">;=> #object[org.apache.mxnet.Context 0x3f376123 "cpu(0)"]</span><span class="w"> |
| |
| </span><span class="p">(</span><span class="k">def</span><span class="w"> </span><span class="n">gpu-b</span><span class="w"> </span><span class="p">(</span><span class="nf">ndarray/zeros</span><span class="w"> </span><span class="p">[</span><span class="mi">100</span><span class="w"> </span><span class="mi">200</span><span class="p">]</span><span class="w"> </span><span class="p">{</span><span class="no">:ctx</span><span class="w"> </span><span class="p">(</span><span class="nf">context/gpu</span><span class="w"> </span><span class="mi">0</span><span class="p">)}))</span><span class="w"> </span><span class="c1">;; to use with gpu</span><span class="w"> |
| |
| </span></code></pre></div></div> |
| |
| <h2 id="next-steps">Next Steps</h2> |
| <ul> |
| <li>See <a href="kvstore">KVStore API</a> for multi-GPU and multi-host distributed training.</li> |
| </ul> |
| |
| </div> |
| </div> |
