Converting Streamlit application to exe file

link管理

链接快照平台

输入网页链接，自动生成快照
标签化管理网页链接

相关文章推荐

面冷心慈的熊猫 · 请问，用 Electron ...· 1 月前 ·

傲视众生的鞭炮 · electron调用DLL开发记录 | ...· 1 月前 ·

多情的松球 · 常见打印质量问题和解决办法 | Bambu ...· 1 周前 ·

温文尔雅的作业本 · 国家知识产权局动态信息 ...· 1 年前 ·

老实的火腿肠 · file.transferto系统找不到路径-掘金· 1 年前 ·

温柔的打火机 · 中英版+英文版汪汪队立大功最新全集1-78季 ...· 2 年前 ·

慈祥的热水瓶 · 如何利用Microsoft ...· 2 年前 ·

Home

Blog

Open Source

Pricing

Contact

</div><div class="navbar-right"><a href="https://www.cloud.ploomber.io/register.html" class="btn btn-primary cloud-signup">Get Started</a></div></div></nav></div></div></div></header><main><section class="section-sm pt-0"><div class="container"><div class="row justify-content-center" data-aos="fade-in"><div class="col-lg-10 mb-5"><img loading="lazy" decoding="async" class="w-100 rounded-lg" src="https://ploomber.io/images/blog/streamlit_exe-header_hu71128c8bf15a88fa2928d6a7e75ccf75_102697_214d7df072442c80e0c51313bb25874c.webp" alt="blog-post" onerror="this.onerror=null,this.src='/images/blog/streamlit_exe-header_hu71128c8bf15a88fa2928d6a7e75ccf75_102697_1020x0_resize_box_2.png'"/></div><div class="col-lg-8"><div class="mb-5"><h1>Converting Streamlit application to exe file</h1><div class="d-flex align-items-center card-author mt-4"><div class="user-image mr-3"><img loading="lazy" decoding="async" class="rounded-circle" src="https://ploomber.io/images/author/neelasha_huddc521a98d51824b33698360ab1c28c4_1987942_6357e5611b6468584180ea9de5abc917.webp" alt="author image" onerror="this.onerror=null,this.src='/images/author/neelasha_huddc521a98d51824b33698360ab1c28c4_1987942_70x70_fill_box_smart1_2.png'"/></div><div class="user-info"><p class="font-weight-500 mb-0 h5">Neelasha Sen</p><span>Mar 07, 2023 - 4 Min read</span></div></div></div><aside class="sticky w-20-l mt4-l d-none d-md-none d-xl-block"><div><div id="TableOfContents" class="nav toc-top"><div><div class="socialheader"><h5>Follow us:</h5></div><ul><div class="twitter"><a href="https://twitter.com/ploomber?ref_src=twsrc%5Etfw" class="twitter-follow-button" data-show-count="false">Follow @ploomber</a> <div class="g-ytsubscribe" data-channelid="UCaIS5BMlmeNQE4-Gn0xTDXQ" data-layout="default" data-count="hidden"/></div></ul></div><div><div><h5>Share:</h5></div><div class="share"><a href="https://twitter.com/share?url=https://ploomber.io/blog/streamlit_exe/&amp;text=Converting%20Streamlit%20application%20to%20exe%20file" class="twitter-share-button" aria-label="share on Twitter"/></div></div><div>Join our <a href="https://ploomber.io/community/" class="docs-toc-title">community!</a></div></div></div></aside><div class="content contentstickycorrection"><h2 id="converting-streamlit-application-to-exe-file">Converting Streamlit application to exe file</h2><blockquote><p>Looking to host or export Streamlit applications? Do it for free with <a href="https://platform.ploomber.io/accounts/signup/">Ploomber Platform!</a></p></blockquote><p><code>Streamlit</code> is an open souce framework for converting data scripts into shareable web applications. It allows data scientists and machine learning engineers to build intuitive interfaces without having to know web development. In this tutorial we’ll learn some approaches of converting a Streamlit application to a double-click executable file for Windows / Linux / Mac platforms.</p><p>The main motivation behind this is that users should be able to bundle the Streamlit application and all of its dependencies into a single package. This bundled package can be easily shared with other users who do not have Streamlit or even Python installed in their machines.</p><p>Let’s create a simple streamlit application. Create a file <code>streamlit_app.py</code> with the below code:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-python" data-lang="python"><span style="color:#f92672">import</span> streamlit <span style="color:#66d9ef">as</span> st x <span style="color:#f92672">=</span> st<span style="color:#f92672">.</span>slider(<span style="color:#e6db74">"Select a value"</span>) st<span style="color:#f92672">.</span>write(x, <span style="color:#e6db74">"squared is"</span>, x <span style="color:#f92672">*</span> x) </code></pre></div><p>This application can be run from the terminal using the below command:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">streamlit run streamlit_app.py </code></pre></div><p>Let’s see two frameworks for converting an application to a desktop application: <code>Nativefier</code> and <code>Stlite</code>. Both of these packages are based on <a href="https://www.electronjs.org/">Electron</a>. Electron is a platform that easily enables writing cross-platform desktop applications using JavaScript, HTML, and CSS. Today, some of the most popular desktop apps are entirely written in Electron such as Atom, Visual Studio Code, Slack, etc.</p><h3 id="nativefier-approach">Nativefier Approach</h3><p><a href="https://github.com/nativefier/nativefier">Nativefier</a> is a command-line tool that easily creates a desktop app for any web site with minimal configuration. Apps are wrapped by Electron in an OS executable (.app, .exe, etc.) for use on Windows, OSX and Linux. Note that converting Streamlit applications using Nativefier requires users to deploy the application to <a href="https://share.streamlit.io/">Streamlit Share</a>.</p><p>Deploy your application in Streamlit Share and copy the URL of the deployed app as shown below:</p><p><img src="https://ploomber.io/images/blog/streamlit_exe/share.png" alt="img"/></p><p>Install <code>nativefier</code> by running the below command:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">npm install -g nativefier </code></pre></div><p>Now convert your Streamlit application like so:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">nativefier --name <span style="color:#e6db74">'&lt;app.exe name&gt;'</span> <span style="color:#e6db74">'&lt;streamlit sharing website url&gt;'</span> --platform &lt;<span style="color:#e6db74">'windows'</span> or <span style="color:#e6db74">'mac'</span> or <span style="color:#e6db74">'linux'</span>&gt; </code></pre></div><p>This will create the exe file in the current directory.</p><p><strong>Package versions</strong></p><p><code>Streamlit</code> 1.19.0</p><p><code>Nativefier</code> 50.0.1</p><h3 id="stlite-approach">Stlite Approach</h3><p>Stlite is a <a href="https://webassembly.org/">WebAssembly</a> port of Streamlit that runs on Pyodide runtime. <a href="https://pyodide.org/en/stable/">Pyodide</a> is an experimental project from Mozilla to create a full Python data science stack that runs entirely in the browser. A Streamlit application can be converted to an exe using <a href="https://github.com/whitphx/stlite/blob/main/packages/desktop/README.md">Stlite desktop</a>. This approach does not require you to deploy the Streamlit application to Streamlit share.</p><p>First create a <code>package.json</code> file to start a new NPM Project. You may edit the <code>name</code> field.</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-python" data-lang="python">{ <span style="color:#e6db74">"name"</span>: <span style="color:#e6db74">"streamlit_app_exe"</span>, <span style="color:#e6db74">"version"</span>: <span style="color:#e6db74">"0.1.0"</span>, <span style="color:#e6db74">"main"</span>: <span style="color:#e6db74">"./build/electron/main.js"</span>, <span style="color:#e6db74">"scripts"</span>: { <span style="color:#e6db74">"dump"</span>: <span style="color:#e6db74">"dump-stlite-desktop-artifacts"</span>, <span style="color:#e6db74">"serve"</span>: <span style="color:#e6db74">'NODE_ENV="production" electron .'</span>, <span style="color:#e6db74">"pack"</span>: <span style="color:#e6db74">"electron-builder --dir"</span>, <span style="color:#e6db74">"dist"</span>: <span style="color:#e6db74">"electron-builder"</span>, <span style="color:#e6db74">"postinstall"</span>: <span style="color:#e6db74">"electron-builder install-app-deps"</span>, <span style="color:#e6db74">"build"</span>: {<span style="color:#e6db74">"files"</span>: [<span style="color:#e6db74">"build/**/*"</span>], <span style="color:#e6db74">"directories"</span>: {<span style="color:#e6db74">"buildResources"</span>: <span style="color:#e6db74">"assets"</span>}}, <span style="color:#e6db74">"devDependencies"</span>: { <span style="color:#e6db74">"@stlite/desktop"</span>: <span style="color:#e6db74">"^0.25.0"</span>, <span style="color:#e6db74">"electron"</span>: <span style="color:#e6db74">"23.1.1"</span>, <span style="color:#e6db74">"electron-builder"</span>: <span style="color:#e6db74">"^23.6.0"</span>, </code></pre></div><p>Install Npm as follows:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">npm install </code></pre></div><p>Then create a directory to contain the application files, e.g., <code>streamlit_app</code>. So the path of the main app file looks like <code>streamlit_app</code>/<code>streamlit_app.py</code></p><p>Run the <code>dump</code> command to create the <code>./build</code> folder.</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">npm run dump streamlit_app </code></pre></div><p>The <code>dist</code> command bundles the <code>./build</code> directory created in the step above into application files (.app, .exe, .dmg etc.) in the ./dist directory.</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">npm run dist </code></pre></div><p><strong>Package versions</strong></p><p><code>Streamlit</code> 1.19.0</p><p><code>Stlite desktop</code> 0.25.0</p><h3 id="pyinstaller-approach">PyInstaller Approach</h3><p><a href="https://pyinstaller.org/en/stable/">PyInstaller</a> bundles a Python application and all its dependencies into a single package. The user can run the packaged app without installing a Python interpreter or any modules.</p><p>Let’s see how you can bundle a Streamlit application using PyInstaller.</p><p>Create a wrapper code <code>run.py</code> to run the main application.</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-python" data-lang="python"><span style="color:#f92672">import</span> streamlit <span style="color:#f92672">import</span> streamlit.web.cli <span style="color:#66d9ef">as</span> stcli <span style="color:#f92672">import</span> os<span style="color:#f92672">,</span> sys <span style="color:#66d9ef">def</span> <span style="color:#a6e22e">resolve_path</span>(path): resolved_path <span style="color:#f92672">=</span> os<span style="color:#f92672">.</span>path<span style="color:#f92672">.</span>abspath(os<span style="color:#f92672">.</span>path<span style="color:#f92672">.</span>join(os<span style="color:#f92672">.</span>getcwd(), path)) <span style="color:#66d9ef">return</span> resolved_path <span style="color:#66d9ef">if</span> __name__ <span style="color:#f92672">==</span> <span style="color:#e6db74">"__main__"</span>: sys<span style="color:#f92672">.</span>argv <span style="color:#f92672">=</span> [ <span style="color:#e6db74">"streamlit"</span>, <span style="color:#e6db74">"run"</span>, resolve_path(<span style="color:#e6db74">"streamlit_app.py"</span>), <span style="color:#e6db74">"--global.developmentMode=false"</span>, sys<span style="color:#f92672">.</span>exit(stcli<span style="color:#f92672">.</span>main()) </code></pre></div><p>Create a hooks file <code>./hooks/hook-streamlit.py</code>:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-python" data-lang="python"><span style="color:#f92672">from</span> PyInstaller.utils.hooks <span style="color:#f92672">import</span> copy_metadata datas <span style="color:#f92672">=</span> copy_metadata(<span style="color:#e6db74">"streamlit"</span>) </code></pre></div><p>Now call PyInstaller as below:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-bash" data-lang="bash">pyinstaller --onefile --additional-hooks-dir<span style="color:#f92672">=</span>./hooks run.py --clean </code></pre></div><p>This will generate <code>build</code> and <code>dist</code> folders and a <code>run.spec</code> file. Edit the <code>run.spec</code> file to ensure paths are set properly as below:</p><div class="highlight"><pre style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4"><code class="language-python" data-lang="python"><span style="color:#f92672">from</span> PyInstaller.utils.hooks <span style="color:#f92672">import</span> collect_data_files <span style="color:#f92672">from</span> PyInstaller.utils.hooks <span style="color:#f92672">import</span> copy_metadata datas <span style="color:#f92672">=</span> [(<span style="color:#e6db74">"{$YOURPYTHONENV}/site-packages/streamlit/runtime"</span>, <span style="color:#e6db74">"./streamlit/runtime"</span>)] datas <span style="color:#f92672">+=</span> collect_data_files(<span style="color:#e6db74">"streamlit"</span>) datas <span style="color:#f92672">+=</span> copy_metadata(<span style="color:#e6db74">"streamlit"</span>) block_cipher <span style="color:#f92672">=</span> <span style="color:#66d9ef">None</span> a <span style="color:#f92672">=</span> Analysis( [<span style="color:#e6db74">"run.py"</span>], pathex<span style="color:#f92672">=</span>[<span style="color:#e6db74">"."</span>], binaries<span style="color:#f92672">=</span>[], datas<span style="color:#f92672">=</span>datas, hiddenimports<span style="color:#f92672">=</span>[], hookspath<span style="color:#f92672">=</span>[], hooksconfig<span style="color:#f92672">=</span>{}, runtime_hooks<span style="color:#f92672">=</span>[], excludes<span style="color:#f92672">=</span>[], win_no_prefer_redirects<span style="color:#f92672">=</span><span style="color:#66d9ef">False</span>, win_private_assemblies<span style="color:#f92672">=</span><span style="color:#66d9ef">False</span>, cipher<span style="color:#f92672">=</span>block_cipher, noarchive<span style="color:#f92672">=</span><span style="color:#66d9ef">False</span>,