<!DOCTYPE html>
<html lang="en">
<head>
<link rel="stylesheet" href="https://unpkg.com/latex.css/style.min.css" />
<link rel="stylesheet" href="/assets/main.css" />
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>id="polynomial-regression-using-tensorflow-2x">Polynomial Regression Using TensorFlow 2.x</title>
<meta name="og:site_name" content="Navan Chauhan" />
<link rel="canonical" href="https://web.navan.dev/posts/2024-03-21-Polynomial-Regression-in-TensorFlow-2.html" />
<meta name="twitter:url" content="https://web.navan.dev/posts/2024-03-21-Polynomial-Regression-in-TensorFlow-2.html />
<meta name="og:url" content="https://web.navan.dev/posts/2024-03-21-Polynomial-Regression-in-TensorFlow-2.html" />
<meta name="twitter:title" content="id="polynomial-regression-using-tensorflow-2x">Polynomial Regression Using TensorFlow 2.x" />
<meta name="og:title" content="id="polynomial-regression-using-tensorflow-2x">Polynomial Regression Using TensorFlow 2.x" />
<meta name="description" content="Predicting n-th degree polynomials using TensorFlow 2.x" />
<meta name="twitter:description" content="Predicting n-th degree polynomials using TensorFlow 2.x" />
<meta name="og:description" content="Predicting n-th degree polynomials using TensorFlow 2.x" />
<meta name="twitter:card" content="summary_large_image" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<link rel="shortcut icon" href="/images/favicon.png" type="image/png" />
<link rel="alternate" href="/feed.rss" type="application/rss+xml" title="Subscribe to Navan Chauhan" />
<meta name="twitter:image" content="https://web.navan.dev/images/opengraph/posts/2024-03-21-Polynomial-Regression-in-TensorFlow-2.png" />
<meta name="og:image" content="https://web.navan.dev/images/opengraph/posts/2024-03-21-Polynomial-Regression-in-TensorFlow-2.png" />
<meta name="google-site-verification" content="LVeSZxz-QskhbEjHxOi7-BM5dDxTg53x2TwrjFxfL0k" />
<script data-goatcounter="https://navanchauhan.goatcounter.com/count"
async src="//gc.zgo.at/count.js"></script>
<script defer data-domain="web.navan.dev" src="https://plausible.io/js/plausible.js"></script>
<link rel="manifest" href="/manifest.json" />
</head>
<body>
<center><nav style="display: block;">
|
<a href="/">home</a> |
<a href="/about/">about/links</a> |
<a href="/posts/">posts</a> |
<a href="/3D-Designs/">3D designs</a> |
<!--<a href="/publications/">publications</a> |-->
<!--<a href="/repo/">iOS repo</a> |-->
<a href="/feed.rss">RSS Feed</a> |
</nav>
</center>
<main>
<h1 id="polynomial-regression-using-tensorflow-2x">Polynomial Regression Using TensorFlow 2.x</h1>
<p>I have a similar post titled <a rel="noopener" target="_blank" href="/posts/2019-12-16-TensorFlow-Polynomial-Regression.html">Polynomial Regression Using Tensorflow</a> that used <code>tensorflow.compat.v1</code> (Which still works as of TF 2.16). But, I thought it would be nicer to redo it with newer TF versions. </p>
<p>I will be skipping all the introductions about polynomial regression and jumping straight to the code. Personally, I prefer using <code>scikit-learn</code> for this task.</p>
<h2 id="position-vs-salary-dataset">Position vs Salary Dataset</h2>
<p>Again, we will be using https://drive.google.com/file/d/1tNL4jxZEfpaP4oflfSn6pIHJX7Pachm9/view (Salary vs Position Dataset)</p>
<p>If you are in a Python Notebook environment like Kaggle or Google Colaboratory, you can simply run:</p>
<div class="codehilite">
<pre><span></span><code><span class="nt">!wget</span><span class="na"> --no-check-certificate 'https</span><span class="p">:</span><span class="nc">//docs.google.com/uc?export</span><span class="o">=</span><span class="l">download&id=1tNL4jxZEfpaP4oflfSn6pIHJX7Pachm9' -O data.csv</span>
</code></pre>
</div>
<h2 id="code">Code</h2>
<p>If you just want to copy-paste the code, scroll to the bottom for the entire snippet. Here I will try and walk through setting up code for a 3rd-degree (cubic) polynomial</p>
<h3 id="imports">Imports</h3>
<div class="codehilite">
<pre><span></span><code><span class="kn">import</span> <span class="nn">pandas</span> <span class="k">as</span> <span class="nn">pd</span>
<span class="kn">import</span> <span class="nn">tensorflow</span> <span class="k">as</span> <span class="nn">tf</span>
<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
</code></pre>
</div>
<h3 id="reading-the-dataset">Reading the Dataset</h3>
<div class="codehilite">
<pre><span></span><code><span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s2">"data.csv"</span><span class="p">)</span>
</code></pre>
</div>
<h3 id="variables-and-constants">Variables and Constants</h3>
<p>Here, we initialize the X and Y values as constants, since they are not going to change. The coefficients are defined as variables.</p>
<div class="codehilite">
<pre><span></span><code><span class="n">X</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="s2">"Level"</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="n">Y</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="s2">"Salary"</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="n">coefficients</span> <span class="o">=</span> <span class="p">[</span><span class="n">tf</span><span class="o">.</span><span class="n">Variable</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randn</span><span class="p">()</span> <span class="o">*</span> <span class="mf">0.01</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
</code></pre>
</div>
<p>Here, <code>X</code> and <code>Y</code> are the values from our dataset. We initialize the coefficients for the equations as small random values.</p>
<p>These coefficients are evaluated by Tensorflow's <code>tf.math.poyval</code> function which returns the n-th order polynomial based on how many coefficients are passed. Since our list of coefficients contains 4 different variables, it will be evaluated as:</p>
<pre><code>y = (x**3)*coefficients[3] + (x**2)*coefficients[2] + (x**1)*coefficients[1] (x**0)*coefficients[0]
</code></pre>
<p>Which is equivalent to the general cubic equation:</p>
<script src="https://cdn.jsdelivr.net/npm/mathjax@4.0.0-beta.4/tex-mml-chtml.js" id="MathJax-script"></script>
<script src="https://cdn.jsdelivr.net/npm/mathjax@4.0.0-beta.4/input/tex/extensions/noerrors.js" charset="UTF-8"></script>
<math xmlns="http://www.w3.org/1998/Math/MathML" display="block"><mrow><mi>y</mi><mo>=</mo><mi>a</mi><msup><mi>x</mi><mn>3</mn></msup><mo>+</mo><mi>b</mi><msup><mi>x</mi><mn>2</mn></msup><mo>+</mo><mi>c</mi><mi>x</mi><mo>+</mo><mi>d</mi></mrow></math>
<h3 id="optimizer-selection-training">Optimizer Selection & Training</h3>
<div class="codehilite">
<pre><span></span><code><span class="n">optimizer</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">keras</span><span class="o">.</span><span class="n">optimizers</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span><span class="n">learning_rate</span><span class="o">=</span><span class="mf">0.3</span><span class="p">)</span>
<span class="n">num_epochs</span> <span class="o">=</span> <span class="mi">10_000</span>
<span class="k">for</span> <span class="n">epoch</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_epochs</span><span class="p">):</span>
<span class="k">with</span> <span class="n">tf</span><span class="o">.</span><span class="n">GradientTape</span><span class="p">()</span> <span class="k">as</span> <span class="n">tape</span><span class="p">:</span>
<span class="n">y_pred</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">polyval</span><span class="p">(</span><span class="n">coefficients</span><span class="p">,</span> <span class="n">X</span><span class="p">)</span>
<span class="n">loss</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">tf</span><span class="o">.</span><span class="n">square</span><span class="p">(</span><span class="n">y</span> <span class="o">-</span> <span class="n">y_pred</span><span class="p">))</span>
<span class="n">grads</span> <span class="o">=</span> <span class="n">tape</span><span class="o">.</span><span class="n">gradient</span><span class="p">(</span><span class="n">loss</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">)</span>
<span class="n">optimizer</span><span class="o">.</span><span class="n">apply_gradients</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">grads</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">))</span>
<span class="k">if</span> <span class="p">(</span><span class="n">epoch</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="mi">1000</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">"Epoch: </span><span class="si">{</span><span class="n">epoch</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">, Loss: </span><span class="si">{</span><span class="n">loss</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="si">}</span><span class="s2">"</span>
</code></pre>
</div>
<p>In TensorFlow 1, we would have been using <code>tf.Session</code> instead. </p>
<p>Here we are using <code>GradientTape()</code> instead, to keep track of the loss evaluation and coefficients. This is crucial, as our optimizer needs these gradients to be able to optimize our coefficients. </p>
<p>Our loss function is Mean Squared Error (MSE):</p>
<math xmlns="http://www.w3.org/1998/Math/MathML" display="block"><mrow><mo>=</mo><mfrac><mrow><mn>1</mn></mrow><mrow><mi>n</mi></mrow></mfrac><msubsup><mo>∑</mo><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>n</mi></mrow></msubsup><mrow><mo stretchy="false">(</mo><mi>Y</mi><mi>_</mi><mi>i</mi><mo>−</mo><mover><mrow><mi>Y</mi><mi>_</mi><mi>i</mi></mrow><mo stretchy="false">^</mo></mover><msup><mo stretchy="false">)</mo><mn>2</mn></msup></mrow></mrow></math>
<p>Where <math xmlns="http://www.w3.org/1998/Math/MathML"><mover><msub><mi>Y</mi><mi>i</mi></msub><mo stretchy="false" style="math-style:normal;math-depth:0;">^</mo></mover></math> is the predicted value and <math xmlns="http://www.w3.org/1998/Math/MathML"><msub><mi>Y</mi><mi>i</mi></msub></math> is the actual value</p>
<h3 id="plotting-final-coefficients">Plotting Final Coefficients</h3>
<div class="codehilite">
<pre><span></span><code><span class="n">final_coefficients</span> <span class="o">=</span> <span class="p">[</span><span class="n">c</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">coefficients</span><span class="p">]</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Final Coefficients:"</span><span class="p">,</span> <span class="n">final_coefficients</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="s2">"Level"</span><span class="p">],</span> <span class="n">df</span><span class="p">[</span><span class="s2">"Salary"</span><span class="p">],</span> <span class="n">label</span><span class="o">=</span><span class="s2">"Original Data"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="s2">"Level"</span><span class="p">],[</span><span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">polyval</span><span class="p">(</span><span class="n">final_coefficients</span><span class="p">,</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">))</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">df</span><span class="p">[</span><span class="s2">"Level"</span><span class="p">]])</span>
<span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span><span class="p">(</span><span class="s1">'Salary'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span><span class="p">(</span><span class="s1">'Position'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">title</span><span class="p">(</span><span class="s2">"Salary vs Position"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</code></pre>
</div>
<h2 id="code-snippet-for-a-polynomial-of-degree-n">Code Snippet for a Polynomial of Degree N</h2>
<h3 id="using-gradient-tape">Using Gradient Tape</h3>
<p>This should work regardless of the Keras backend version (2 or 3)</p>
<div class="codehilite">
<pre><span></span><code><span class="kn">import</span> <span class="nn">tensorflow</span> <span class="k">as</span> <span class="nn">tf</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="kn">import</span> <span class="nn">pandas</span> <span class="k">as</span> <span class="nn">pd</span>
<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
<span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s2">"data.csv"</span><span class="p">)</span>
<span class="c1">############################</span>
<span class="c1">## Change Parameters Here ##</span>
<span class="c1">############################</span>
<span class="n">x_column</span> <span class="o">=</span> <span class="s2">"Level"</span> <span class="c1">#</span>
<span class="n">y_column</span> <span class="o">=</span> <span class="s2">"Salary"</span> <span class="c1">#</span>
<span class="n">degree</span> <span class="o">=</span> <span class="mi">2</span> <span class="c1">#</span>
<span class="n">learning_rate</span> <span class="o">=</span> <span class="mf">0.3</span> <span class="c1">#</span>
<span class="n">num_epochs</span> <span class="o">=</span> <span class="mi">25_000</span> <span class="c1">#</span>
<span class="c1">############################</span>
<span class="n">X</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="n">x_column</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="n">Y</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="n">y_column</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="n">coefficients</span> <span class="o">=</span> <span class="p">[</span><span class="n">tf</span><span class="o">.</span><span class="n">Variable</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randn</span><span class="p">()</span> <span class="o">*</span> <span class="mf">0.01</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">degree</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)]</span>
<span class="n">optimizer</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">keras</span><span class="o">.</span><span class="n">optimizers</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span><span class="n">learning_rate</span><span class="o">=</span><span class="n">learning_rate</span><span class="p">)</span>
<span class="k">for</span> <span class="n">epoch</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_epochs</span><span class="p">):</span>
<span class="k">with</span> <span class="n">tf</span><span class="o">.</span><span class="n">GradientTape</span><span class="p">()</span> <span class="k">as</span> <span class="n">tape</span><span class="p">:</span>
<span class="n">y_pred</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">polyval</span><span class="p">(</span><span class="n">coefficients</span><span class="p">,</span> <span class="n">X</span><span class="p">)</span>
<span class="n">loss</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">tf</span><span class="o">.</span><span class="n">square</span><span class="p">(</span><span class="n">Y</span> <span class="o">-</span> <span class="n">y_pred</span><span class="p">))</span>
<span class="n">grads</span> <span class="o">=</span> <span class="n">tape</span><span class="o">.</span><span class="n">gradient</span><span class="p">(</span><span class="n">loss</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">)</span>
<span class="n">optimizer</span><span class="o">.</span><span class="n">apply_gradients</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">grads</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">))</span>
<span class="k">if</span> <span class="p">(</span><span class="n">epoch</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="mi">1000</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">"Epoch: </span><span class="si">{</span><span class="n">epoch</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">, Loss: </span><span class="si">{</span><span class="n">loss</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="si">}</span><span class="s2">"</span><span class="p">)</span>
<span class="n">final_coefficients</span> <span class="o">=</span> <span class="p">[</span><span class="n">c</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">coefficients</span><span class="p">]</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Final Coefficients:"</span><span class="p">,</span> <span class="n">final_coefficients</span><span class="p">)</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Final Equation:"</span><span class="p">,</span> <span class="n">end</span><span class="o">=</span><span class="s2">" "</span><span class="p">)</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">degree</span><span class="o">+</span><span class="mi">1</span><span class="p">):</span>
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">final_coefficients</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="si">}</span><span class="s2"> * x^</span><span class="si">{</span><span class="n">degree</span><span class="o">-</span><span class="n">i</span><span class="si">}</span><span class="s2">"</span><span class="p">,</span> <span class="n">end</span><span class="o">=</span><span class="s2">" + "</span> <span class="k">if</span> <span class="n">i</span> <span class="o"><</span> <span class="n">degree</span> <span class="k">else</span> <span class="s2">"</span><span class="se">\n</span><span class="s2">"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">Y</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="s2">"Original Data"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">X</span><span class="p">,[</span><span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">polyval</span><span class="p">(</span><span class="n">final_coefficients</span><span class="p">,</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">))</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">df</span><span class="p">[</span><span class="n">x_column</span><span class="p">]]),</span> <span class="n">label</span><span class="o">=</span><span class="s2">"Our Poynomial"</span>
<span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span><span class="p">(</span><span class="n">y_column</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span><span class="p">(</span><span class="n">x_column</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">title</span><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">x_column</span><span class="si">}</span><span class="s2"> vs </span><span class="si">{</span><span class="n">y_column</span><span class="si">}</span><span class="s2">"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">legend</span><span class="p">()</span>
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</code></pre>
</div>
<h3 id="without-gradient-tape">Without Gradient Tape</h3>
<p>This relies on the Optimizer's <code>minimize</code> function and uses the <code>var_list</code> parameter to update the variables.</p>
<p>This will not work with Keras 3 backend in TF 2.16.0 and above unless you switch to the legacy backend.</p>
<div class="codehilite">
<pre><span></span><code><span class="kn">import</span> <span class="nn">tensorflow</span> <span class="k">as</span> <span class="nn">tf</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="kn">import</span> <span class="nn">pandas</span> <span class="k">as</span> <span class="nn">pd</span>
<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
<span class="n">df</span> <span class="o">=</span> <span class="n">pd</span><span class="o">.</span><span class="n">read_csv</span><span class="p">(</span><span class="s2">"data.csv"</span><span class="p">)</span>
<span class="c1">############################</span>
<span class="c1">## Change Parameters Here ##</span>
<span class="c1">############################</span>
<span class="n">x_column</span> <span class="o">=</span> <span class="s2">"Level"</span> <span class="c1">#</span>
<span class="n">y_column</span> <span class="o">=</span> <span class="s2">"Salary"</span> <span class="c1">#</span>
<span class="n">degree</span> <span class="o">=</span> <span class="mi">2</span> <span class="c1">#</span>
<span class="n">learning_rate</span> <span class="o">=</span> <span class="mf">0.3</span> <span class="c1">#</span>
<span class="n">num_epochs</span> <span class="o">=</span> <span class="mi">25_000</span> <span class="c1">#</span>
<span class="c1">############################</span>
<span class="n">X</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="n">x_column</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="n">Y</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">df</span><span class="p">[</span><span class="n">y_column</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="n">coefficients</span> <span class="o">=</span> <span class="p">[</span><span class="n">tf</span><span class="o">.</span><span class="n">Variable</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randn</span><span class="p">()</span> <span class="o">*</span> <span class="mf">0.01</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">degree</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)]</span>
<span class="n">optimizer</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">keras</span><span class="o">.</span><span class="n">optimizers</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span><span class="n">learning_rate</span><span class="o">=</span><span class="n">learning_rate</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">loss_function</span><span class="p">():</span>
<span class="n">pred_y</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">polyval</span><span class="p">(</span><span class="n">coefficients</span><span class="p">,</span> <span class="n">X</span><span class="p">)</span>
<span class="k">return</span> <span class="n">tf</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">tf</span><span class="o">.</span><span class="n">square</span><span class="p">(</span><span class="n">pred_y</span> <span class="o">-</span> <span class="n">Y</span><span class="p">))</span>
<span class="k">for</span> <span class="n">epoch</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_epochs</span><span class="p">):</span>
<span class="n">optimizer</span><span class="o">.</span><span class="n">minimize</span><span class="p">(</span><span class="n">loss_function</span><span class="p">,</span> <span class="n">var_list</span><span class="o">=</span><span class="n">coefficients</span><span class="p">)</span>
<span class="k">if</span> <span class="p">(</span><span class="n">epoch</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="mi">1000</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="n">current_loss</span> <span class="o">=</span> <span class="n">loss_function</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">"Epoch </span><span class="si">{</span><span class="n">epoch</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">: Training Loss: </span><span class="si">{</span><span class="n">current_loss</span><span class="si">}</span><span class="s2">"</span><span class="p">)</span>
<span class="n">final_coefficients</span> <span class="o">=</span> <span class="n">coefficients</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Final Coefficients:"</span><span class="p">,</span> <span class="n">final_coefficients</span><span class="p">)</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Final Equation:"</span><span class="p">,</span> <span class="n">end</span><span class="o">=</span><span class="s2">" "</span><span class="p">)</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">degree</span><span class="o">+</span><span class="mi">1</span><span class="p">):</span>
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">final_coefficients</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="si">}</span><span class="s2"> * x^</span><span class="si">{</span><span class="n">degree</span><span class="o">-</span><span class="n">i</span><span class="si">}</span><span class="s2">"</span><span class="p">,</span> <span class="n">end</span><span class="o">=</span><span class="s2">" + "</span> <span class="k">if</span> <span class="n">i</span> <span class="o"><</span> <span class="n">degree</span> <span class="k">else</span> <span class="s2">"</span><span class="se">\n</span><span class="s2">"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">Y</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="s2">"Original Data"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">X</span><span class="p">,[</span><span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">polyval</span><span class="p">(</span><span class="n">final_coefficients</span><span class="p">,</span> <span class="n">tf</span><span class="o">.</span><span class="n">constant</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tf</span><span class="o">.</span><span class="n">float32</span><span class="p">))</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">df</span><span class="p">[</span><span class="n">x_column</span><span class="p">]],</span> <span class="n">label</span><span class="o">=</span><span class="s2">"Our Polynomial"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span><span class="p">(</span><span class="n">y_column</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span><span class="p">(</span><span class="n">x_column</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">legend</span><span class="p">()</span>
<span class="n">plt</span><span class="o">.</span><span class="n">title</span><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">x_column</span><span class="si">}</span><span class="s2"> vs </span><span class="si">{</span><span class="n">y_column</span><span class="si">}</span><span class="s2">"</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</code></pre>
</div>
<p>As always, remember to tweak the parameters and choose the correct model for the job. A polynomial regression model might not even be the best model for this particular dataset.</p>
<h2 id="further-programming">Further Programming</h2>
<p>How would you modify this code to use another type of nonlinear regression? Say, </p>
<math xmlns="http://www.w3.org/1998/Math/MathML" display="block"><mrow><mi>y</mi><mo>=</mo><mi>a</mi><msup><mi>b</mi><mi>x</mi></msup></mrow></math>
<p>Hint: Your loss calculation would be similar to:</p>
<div class="codehilite">
<pre><span></span><code><span class="n">bx</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="n">coefficients</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">X</span><span class="p">)</span>
<span class="n">pred_y</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">math</span><span class="o">.</span><span class="n">multiply</span><span class="p">(</span><span class="n">coefficients</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">bx</span><span class="p">)</span>
<span class="n">loss</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">(</span><span class="n">tf</span><span class="o">.</span><span class="n">square</span><span class="p">(</span><span class="n">pred_y</span> <span class="o">-</span> <span class="n">Y</span><span class="p">))</span>
</code></pre>
</div>
<blockquote>If you have scrolled this far, consider subscribing to my mailing list <a href="https://listmonk.navan.dev/subscription/form">here.</a> You can subscribe to either a specific type of post you are interested in, or subscribe to everything with the "Everything" list.</blockquote>
<script data-isso="https://comments.navan.dev/"
src="https://comments.navan.dev/js/embed.min.js"></script>
<section id="isso-thread">
<noscript>Javascript needs to be activated to view comments.</noscript>
</section>
</main>
<script src="assets/manup.min.js"></script>
<script src="/pwabuilder-sw-register.js"></script>
</body>
</html>