<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by R. Kukuh on Medium]]></title>
        <description><![CDATA[Stories by R. Kukuh on Medium]]></description>
        <link>https://medium.com/@rkukuh?source=rss-b7563d2b13d7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*7mn9MA5BSuKbJYNEs7-HjQ.png</url>
            <title>Stories by R. Kukuh on Medium</title>
            <link>https://medium.com/@rkukuh?source=rss-b7563d2b13d7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 16:59:22 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@rkukuh/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Tokenization in Action]]></title>
            <link>https://medium.com/learn-ai-with-rkukuh/tokenization-in-action-6db1107d66c3?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/6db1107d66c3</guid>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[google-colab]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Tue, 26 Aug 2025 01:38:54 GMT</pubDate>
            <atom:updated>2025-08-26T01:38:54.057Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zfwVxiI_BvOSeIjYP6Q4fA.png" /></figure><p>This article is the 2nd chapter of <a href="https://medium.com/learn-ai-with-rkukuh/understanding-llm-series-87d6de7519a8">Understanding LLM series</a>.</p><h3>Setup</h3><h4>Load an LLM</h4><p>Code:</p><pre>from transformers import AutoModelForCausalLM, AutoTokenizer<br><br># Load model and tokenizer<br>model = AutoModelForCausalLM.from_pretrained(<br>    &quot;microsoft/Phi-3-mini-4k-instruct&quot;,<br>    device_map=&quot;cuda&quot;,<br>    torch_dtype=&quot;auto&quot;,<br>    trust_remote_code=True,<br>)<br><br>tokenizer = AutoTokenizer.from_pretrained(&quot;microsoft/Phi-3-mini-4k-instruct&quot;)</pre><p>Output:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ldkcOF1_8LwUmsfbGpMFtQ.png" /></figure><h4>Prompt example</h4><p>Code:</p><pre>prompt = &quot;Write a one-liner poetry about sunset.&lt;|assistant|&gt;&quot;<br><br># Tokenize the input prompt<br>input_ids = tokenizer(prompt, return_tensors=&quot;pt&quot;).input_ids.to(&quot;cuda&quot;)<br><br># Generate the text<br>generation_output = model.generate(<br>  input_ids=input_ids,<br>  max_new_tokens=100,<br>  use_cache=False<br>)<br><br># Print the output<br>print(tokenizer.decode(generation_output[0]))</pre><p>Output:</p><pre>Write a one-liner poetry about sunset.&lt;|assistant|&gt; &quot;As daylight wanes, the sun dips low,<br>A fiery orb in the sky&#39;s grand show.<br>Horizon ablaze with hues of gold,<br>Night&#39;s embrace, the day&#39;s tale told.&quot;&lt;|endoftext|&gt;</pre><h3>Tokenization</h3><h4>Print decoded input</h4><p>Code:</p><pre>print(input_ids)</pre><p>Output:</p><pre>tensor([[14350,   263,   697, 29899,  1915,   261, 22309,  1048,  6575,   842,<br>         29889, 32001]], device=&#39;cuda:0&#39;)</pre><p>Code:</p><pre>for id in input_ids[0]:<br>   print(tokenizer.decode(id))</pre><p>Output:</p><pre>Write<br>a<br>one<br>-<br>lin<br>er<br>poetry<br>about<br>sun<br>set<br>.<br>&lt;|assistant|&gt;</pre><h4>Print decoded output</h4><p>Code:</p><pre>generation_output</pre><p>Output:</p><pre>tensor([[14350,   263,   697, 29899,  1915,   261, 22309,  1048,  6575,   842,<br>         29889, 32001,   376,  2887,  2462,  4366,   281, 17297, 29892,   278,<br>          6575,   652,   567,  4482, 29892,    13, 29909, 21334, 29891, 19528,<br>           297,   278, 14744, 29915, 29879,  4595,  1510, 29889,    13, 17241,<br>         18162,   633,   433,   911,   411,   298,  1041,   310,  7684, 29892,<br>            13, 29940,   523, 29915, 29879,   953, 13842, 29892,   278,  2462,<br>         29915, 29879, 17694,  5429,  1213, 32000]], device=&#39;cuda:0&#39;)</pre><p>Code:</p><pre>print(tokenizer.decode(13842))<br>print(tokenizer.decode(29892))<br>print(tokenizer.decode([278, 2462]))</pre><p>Output:</p><pre>brace<br>,<br>the day</pre><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/7730cdcc01bdf862ef1ddc4ab0b9aa77/href">https://medium.com/media/7730cdcc01bdf862ef1ddc4ab0b9aa77/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6db1107d66c3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ai-with-rkukuh/tokenization-in-action-6db1107d66c3">Tokenization in Action</a> was originally published in <a href="https://medium.com/learn-ai-with-rkukuh">Learn AI with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Comparing Tokenizers]]></title>
            <link>https://medium.com/learn-ai-with-rkukuh/comparing-tokenizers-510f32e39728?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/510f32e39728</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[google-colab]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Mon, 25 Aug 2025 13:07:23 GMT</pubDate>
            <atom:updated>2025-08-26T02:00:55.282Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oaNYvDqTm_Gt7nqMkm2nfQ.png" /></figure><p>This article is the 3rd chapter of <a href="https://medium.com/learn-ai-with-rkukuh/understanding-llm-series-87d6de7519a8">Understanding LLM series</a>.</p><p>We’ll compare eight tokenizers to see which perform best to tokenize: <br>English word, capitalized word, other languages, emoji, mathematic operation, and code.</p><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/fb523f9c341aa503f4e8761c0b4c1786/href">https://medium.com/media/fb523f9c341aa503f4e8761c0b4c1786/href</a></iframe><h3>Tokenizer Properties</h3><p>The embedded Google Colab above is the demo of how trained tokenizers work. As you can see, they showed a number of ways of tokenization.</p><p>But what actually determines their tokenization behaviour?</p><p>There are three factor of LLM design choices that determine how the tokenizer break down text:</p><ol><li>Tokenization Methods</li><li>Tokenizer Parameters</li><li>Tokenizer Target Domain</li></ol><h3>Tokenization Methods</h3><p>There are four notable methods of tokenization: Word tokens, Subword tokens, Character tokens, and Byte tokens as shown in the picture below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*G0jqi2dsSZ1VmK-d-8WP8A.png" /></figure><h3>Tokenizer Parameters</h3><p>Once the tokenization method is chosen, it’s time to decide the parameters of the tokenizer:</p><ul><li>Vocabulary size (50K? 100K?)</li><li>Special tokens (beginning/end of token? padding token?)</li><li>Capitalization</li></ul><h3>Tokenizer Target Domain</h3><p>Tokenizer behaviour will be different based on the dataset it was trained on. From the code example in the Google Colab above, we saw how each tokenizer behave on the code (programming text).</p><p>This is the example of tokenization that is not code-specific:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iJjuJgx7J1UcB00iUaMLPg.png" /></figure><p>And this is the result of tokenization from the one that is code-specific:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bAw_W5R5cYgWza2k4928dw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=510f32e39728" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ai-with-rkukuh/comparing-tokenizers-510f32e39728">Comparing Tokenizers</a> was originally published in <a href="https://medium.com/learn-ai-with-rkukuh">Learn AI with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Supervised Learning]]></title>
            <link>https://medium.com/learn-ml-with-rkukuh/supervised-learning-84ba0cbd4f6d?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/84ba0cbd4f6d</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[k-nearest-neighbours]]></category>
            <category><![CDATA[supervised-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Fri, 22 Aug 2025 16:56:13 GMT</pubDate>
            <atom:updated>2025-08-22T16:56:13.267Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VGyEPWaRIYRsa-nHbKn3WA.png" /></figure><p>This is the 2nd chapter of “<a href="https://medium.com/learn-ml-with-rkukuh/applied-machine-learning-4ad93ebf2d71">Applied Machine Learning</a>” series.</p><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/350fabcf4be8b585227341b2c90ac7bd/href">https://medium.com/media/350fabcf4be8b585227341b2c90ac7bd/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=84ba0cbd4f6d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ml-with-rkukuh/supervised-learning-84ba0cbd4f6d">Supervised Learning</a> was originally published in <a href="https://medium.com/learn-ml-with-rkukuh">Learn ML with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Finding and Trying Our First LLM]]></title>
            <link>https://medium.com/learn-ai-with-rkukuh/finding-and-trying-our-first-llm-deb269513ca7?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/deb269513ca7</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[google-colab]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Fri, 22 Aug 2025 00:51:51 GMT</pubDate>
            <atom:updated>2025-08-22T02:39:31.690Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m1TSR-WAJRkKrZYY-FQVvw.png" /></figure><p>This article is the first chapter of <a href="https://medium.com/learn-ai-with-rkukuh/understanding-llm-series-87d6de7519a8">Understanding LLM series</a>.</p><p>Three simple steps:</p><ol><li>Get Access Token</li><li>Setup Google Colab</li><li>Download and Run Chosen LLM</li></ol><h3>Step 1. Get Access Token</h3><h4>1.1 Get HuggingFace access token</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uyM7mQXM9tS9mjZzZCZxIg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U5wnLbEueRhOhAZlLyE90g.png" /></figure><h4>1.2 Set HuggingFace token to Google Colab</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/818/1*uncrA1d0EoCTIK8mKi9m9w.png" /></figure><h3>Step 2. Setup Google Colab</h3><h4>Chose the right GPU</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/990/1*t3o1tmhxLRQJTrz_nIaWhQ.png" /></figure><h3>Step 3. Download and Run Chosen LLM</h3><h4>3.1 Downloading the chosen LLM</h4><pre>from transformers import AutoModelForCausalLM, AutoTokenizer<br><br># Load model and tokenizer<br>model = AutoModelForCausalLM.from_pretrained(<br>    &quot;microsoft/Phi-3-mini-4k-instruct&quot;,<br>    device_map=&quot;c<br>uda&quot;,<br>    torch_dtype=&quot;auto&quot;,<br>    trust_remote_code=False,<br>)<br><br>tokenizer = AutoTokenizer.from_pretrained(&quot;microsoft/Phi-3-mini-4k-instruct&quot;)</pre><p>Run the code above, and you’ll get something similar to mine:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WmM3yKVJ7j1z167CfRArXg.png" /></figure><h4>3.3 Configure the transformer’s pipeline</h4><pre>from transformers import pipeline<br><br># Create a pipeline<br>generator = pipeline(<br>    &quot;text-generation&quot;,<br>    model=model,<br>    tokenizer=tokenizer,<br>    return_full_text=False,<br>    max_new_tokens=500,<br>    do_sample=False<br>)</pre><h4>3.4 Use chosen LLM to generate text</h4><pre># The prompt (user input / query)<br>messages = [<br>    {&quot;role&quot;: &quot;user&quot;, &quot;content&quot;: &quot;Create one-liner quote about wealth.&quot;}<br>]<br><br># Generate output<br>output = generator(messages)<br><br>print(output[0][&quot;generated_text&quot;])</pre><p>Here’s what I got in my side:</p><pre>&quot;Wealth is not about having more, but about having enough to share.&quot;</pre><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/1b94cef01c6ddbe4001ee743ea6a844d/href">https://medium.com/media/1b94cef01c6ddbe4001ee743ea6a844d/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=deb269513ca7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ai-with-rkukuh/finding-and-trying-our-first-llm-deb269513ca7">Finding and Trying Our First LLM</a> was originally published in <a href="https://medium.com/learn-ai-with-rkukuh">Learn AI with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Series: Understanding LLM]]></title>
            <link>https://medium.com/learn-ai-with-rkukuh/understanding-llm-series-87d6de7519a8?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/87d6de7519a8</guid>
            <category><![CDATA[large-language-models]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Fri, 22 Aug 2025 00:35:11 GMT</pubDate>
            <atom:updated>2025-08-26T01:39:22.070Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IXuzjlu2_1Z1NXlakXMAvQ.png" /></figure><p>This article is just a table of contents of this series</p><p>Chapter 1: <a href="https://medium.com/learn-ai-with-rkukuh/finding-and-trying-our-first-llm-deb269513ca7">Finding and Trying Our First LLM</a></p><p>Chapter 2: <a href="https://medium.com/learn-ai-with-rkukuh/tokenization-in-action-6db1107d66c3">Tokenization in Action</a></p><p>Chapter 3: <a href="https://medium.com/learn-ai-with-rkukuh/comparing-tokenizers-510f32e39728">Comparing Tokenizers</a></p><p>Chapter 4: Embeddings</p><p>Chapter 5: LLM, Under-the-hood</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=87d6de7519a8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ai-with-rkukuh/understanding-llm-series-87d6de7519a8">Series: Understanding LLM</a> was originally published in <a href="https://medium.com/learn-ai-with-rkukuh">Learn AI with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unsupervised Learning]]></title>
            <link>https://medium.com/learn-ml-with-rkukuh/unsupervised-learning-8e725c23feef?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/8e725c23feef</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[k-means-clustering]]></category>
            <category><![CDATA[unsupervised-learning]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Tue, 19 Aug 2025 02:37:57 GMT</pubDate>
            <atom:updated>2025-08-26T06:29:56.483Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hi_oNe9av3HGpdSe7QSY3Q.png" /></figure><p>This is the first chapter of “<a href="https://medium.com/learn-ml-with-rkukuh/applied-machine-learning-4ad93ebf2d71">Applied Machine Learning</a>” series.</p><h4>Download Dataset</h4><p>Download the dataset used in this article/tutorial:</p><p><a href="https://drive.google.com/file/d/1BJ5K0kKnmHy2AzvpZXptwwrtN0mTth5P/view?usp=sharing">customers.csv</a></p><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/1f65bd98592c1b7984a460345bf2288c/href">https://medium.com/media/1f65bd98592c1b7984a460345bf2288c/href</a></iframe><h3>Note</h3><ul><li>k-means might be the most popular clustering algorithm, but it’s not the only one.</li><li>Among others, there is <a href="https://scikit-learn.org/stable/modules/generated/sklearn.cluster.DBSCAN.html">DBSCAN</a> (Density-based Spatial Clustering of Applications with Noise). DBSCAN doesn’t require the cluster count to be specified ahead of time and can also identifyy points fall outside the clusters aka outliers.</li><li>Another one is <a href="https://scikit-learn.org/stable/modules/generated/sklearn.cluster.AgglomerativeClustering.html">Agglomerative Clustering</a>, which clusters data points in a hierarchical manner.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8e725c23feef" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ml-with-rkukuh/unsupervised-learning-8e725c23feef">Unsupervised Learning</a> was originally published in <a href="https://medium.com/learn-ml-with-rkukuh">Learn ML with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Series: Applied Machine Learning]]></title>
            <link>https://medium.com/learn-ml-with-rkukuh/applied-machine-learning-4ad93ebf2d71?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/4ad93ebf2d71</guid>
            <category><![CDATA[unsupervised-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[supervised-learning]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Tue, 19 Aug 2025 02:26:04 GMT</pubDate>
            <atom:updated>2025-08-22T16:56:34.998Z</atom:updated>
            <content:encoded><![CDATA[<p>This page is just for Table of Content of this series.</p><h3>Table of Content</h3><p>Chapter 1: <a href="https://medium.com/learn-ml-with-rkukuh/unsupervised-learning-8e725c23feef">Unsupervised Learning</a></p><p>Chapter 2: <a href="https://medium.com/learn-ml-with-rkukuh/supervised-learning-84ba0cbd4f6d">Supervised Learning</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4ad93ebf2d71" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ml-with-rkukuh/applied-machine-learning-4ad93ebf2d71">Series: Applied Machine Learning</a> was originally published in <a href="https://medium.com/learn-ml-with-rkukuh">Learn ML with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Perceptron]]></title>
            <link>https://medium.com/learn-ml-with-rkukuh/perceptron-f0ff364ea447?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/f0ff364ea447</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Wed, 16 Jul 2025 07:53:16 GMT</pubDate>
            <atom:updated>2025-07-24T04:46:34.115Z</atom:updated>
            <content:encoded><![CDATA[<p>This is the 10th chapter 🎉 of “<a href="https://medium.com/learn-ai-ml-with-rkukuh/ml-algorithms-from-scratch-series-2d3ee293ad69">ML Algorithms From Scratch</a>” series.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K1IxhndeLS-hEk_UKYbzxA.png" /></figure><p>The simplest type of Artificial Neural Network (ANN) is perceptron algorithm. It is a model of a single neuron that can be used for two-class classification problems, and provides the foundation for later developing much larger networks.</p><h4>Required Dataset</h4><p>Download this dataset before proceeding</p><p><a href="https://drive.google.com/file/d/1SuqLvMJi-zw6jyt-YPm1ekP_cHIpF1Wo/view?usp=sharing">sonar.all-data.csv</a></p><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/f4af5143c10433e154c7038615c06bfc/href">https://medium.com/media/f4af5143c10433e154c7038615c06bfc/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f0ff364ea447" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ml-with-rkukuh/perceptron-f0ff364ea447">Perceptron</a> was originally published in <a href="https://medium.com/learn-ml-with-rkukuh">Learn ML with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Logistic Regression]]></title>
            <link>https://medium.com/learn-ml-with-rkukuh/logistic-regression-a78c1e79ef5c?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/a78c1e79ef5c</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[google-colab]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Tue, 15 Jul 2025 10:41:38 GMT</pubDate>
            <atom:updated>2025-07-24T04:34:47.712Z</atom:updated>
            <content:encoded><![CDATA[<p>This is the 9th chapter of “<a href="https://medium.com/learn-ai-ml-with-rkukuh/ml-algorithms-from-scratch-series-2d3ee293ad69">ML Algorithms From Scratch</a>” series.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0nRpiLHvZVMr9-vdAFyKYA.png" /></figure><p>Logistic regression is an efficient and reliable linear classification algorithm for two-class problems.</p><p>In this tutorial we’ll learn how to implement logistic regression with stochastic gradient descent from scratch by:</p><ul><li>Making predictions with logistic regression model.</li><li>Estimating coefficients using stochastic gradient descent.</li><li>Applying logistic regression to a real-world prediction problem.</li></ul><h4>Required Dataset</h4><p>Download the dataset used in this article.</p><p><a href="https://drive.google.com/file/d/1VFvnE3L7Mhi8sdbsJQ5LIE6aAAipcsTK/view?usp=sharing">pima-indians-diabetes.csv</a></p><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/98201b6b186505459b5b6a3bc59b4727/href">https://medium.com/media/98201b6b186505459b5b6a3bc59b4727/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a78c1e79ef5c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ml-with-rkukuh/logistic-regression-a78c1e79ef5c">Logistic Regression</a> was originally published in <a href="https://medium.com/learn-ml-with-rkukuh">Learn ML with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Multivariate Linear Regression]]></title>
            <link>https://medium.com/learn-ml-with-rkukuh/multivariate-linear-regression-b7aff7d82d98?source=rss-b7563d2b13d7------2</link>
            <guid isPermaLink="false">https://medium.com/p/b7aff7d82d98</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[R. Kukuh]]></dc:creator>
            <pubDate>Mon, 14 Jul 2025 06:31:12 GMT</pubDate>
            <atom:updated>2025-07-24T04:26:13.233Z</atom:updated>
            <content:encoded><![CDATA[<p>This is the 8th chapter of “<a href="https://medium.com/learn-ai-ml-with-rkukuh/ml-algorithms-from-scratch-series-2d3ee293ad69">ML Algorithms From Scratch</a>” series.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cLy2VaaPnkWLXCNsZiS8bg.png" /></figure><p>Optimization is the core of many ML algorithms. Optimization itself is used to find a good set of ML model parameters given a training dataset.</p><p>The most common ML optimization algorithm is stochastic gradient descent. After completing this tutorial we should know how to:</p><ul><li>Estimate linear regression coefficients using stochastic gradient descent.</li><li>Make predictions for multivariate linear regression.</li><li>Implement linear regression with stochastic gradient descent to make predictions on new data.</li></ul><h4>Required Dataset</h4><p>Download the required dataset to work with here:</p><p><a href="https://drive.google.com/file/d/1ezt4-gDGDSlv1LZ0jklwfL7vERrij91S/view?usp=sharing">winequality-white.csv</a></p><h3>Shared Google Colab</h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/aa822e86a4fdbf008db7da7009a3eb73/href">https://medium.com/media/aa822e86a4fdbf008db7da7009a3eb73/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b7aff7d82d98" width="1" height="1" alt=""><hr><p><a href="https://medium.com/learn-ml-with-rkukuh/multivariate-linear-regression-b7aff7d82d98">Multivariate Linear Regression</a> was originally published in <a href="https://medium.com/learn-ml-with-rkukuh">Learn ML with RKukuh</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>