<?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 Anouar Bakouch on Medium]]></title>
        <description><![CDATA[Stories by Anouar Bakouch on Medium]]></description>
        <link>https://medium.com/@anouar_bakouch?source=rss-32e75440c3c7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*0C_WEUJ-zVnVVBva6M5KGQ.jpeg</url>
            <title>Stories by Anouar Bakouch on Medium</title>
            <link>https://medium.com/@anouar_bakouch?source=rss-32e75440c3c7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 19:34:38 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@anouar_bakouch/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[Gradient Descent… In the Pocket]]></title>
            <link>https://medium.com/@anouar_bakouch/gradient-descent-in-the-pocket-a157b5e1cc77?source=rss-32e75440c3c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/a157b5e1cc77</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[mathematics]]></category>
            <dc:creator><![CDATA[Anouar Bakouch]]></dc:creator>
            <pubDate>Wed, 18 Feb 2026 23:51:01 GMT</pubDate>
            <atom:updated>2026-02-18T23:51:01.150Z</atom:updated>
            <content:encoded><![CDATA[<p>Mathematics is the magic behind the universe, and if someone asks me why?</p><p>I would reply: Mathematics lets us <strong>model</strong> and explain everything even things that physics <strong>cannot</strong> explain yet. Here I will explain what gradient descent is in machine learning, it is an easy and <strong>fun</strong> concept.</p><p>During the training of a machine learning model, we tend to evaluate it at the end based on its accuracy and loss (the two main ones).</p><p>Of course, during training, the model is allowed to <strong>make as many wrong guesses</strong> as it wants, <strong>because that is its</strong> training. The same thing can be said for a student preparing for an exam.</p><p>I say that mistakes are inevitable and are the things that lead to the truth and gradient descent proves it:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/735/1*mpaFbZlUteP8vDgcOma2tg.png" /><figcaption>Brilliant course : Gradient descent</figcaption></figure><p>A model (neural network) <strong>has</strong> two weights, <strong>w1</strong> and <strong>w2</strong>, for simplicity. Changing (updating) these weights during training <strong>leads</strong> to minimizing loss.</p><p>As you see in the image, at the beginning we have a high loss. Think of it like being lost on a foggy mountain your goal is to reach the valley at the bottom.</p><p>The graph represents an <strong>infinite number of possibilities</strong> because the weights can be any decimal value it&#39;s impossible to check them all one by one!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/667/1*0YjQBw9z5VDsYiHvT2IMVA.png" /><figcaption>Brilliant course : Gradient descent</figcaption></figure><p>In fact, as the image shows, the model might have already explored over <strong>1500 possibilities</strong> just to find this path.</p><p>The graph represents <strong>thousands of possibilities</strong>, so to make it faster and in order not to <strong>have to</strong> navigate all of those possibilities we go through what we call <strong>gradient</strong> descent.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/650/1*uuN6N1ZE0P-9eYBlSeJwhQ.png" /><figcaption>Brilliant course</figcaption></figure><p>Notice the jump in progress: we went from exploring <strong>1515</strong> possibilities to <strong>2693</strong>.</p><p>By following the gradient, the model finally finds the bottom of the valley. If you look at the small graph in the top-left of the second image, the curve now perfectly hugs the data points! The loss has been conquered.</p><p>This is about calculating derivatives to get the orientation and values for the next weights.</p><p>This principle is called <strong>gradient descent</strong>. It’s simple enough that it can be encapsulated in a computer program and automated.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*KWwxnQcriXGukzlbQ2HZCQ.png" /><figcaption>Brilliant course</figcaption></figure><p>Note that it can be found after 323 iterations too !</p><p>All the deep nets in use in today’s technology are trained by some flavor of it.</p><p>You must be happy now , gradient descent is now in the pocket.</p><p>Anouar</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a157b5e1cc77" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Virtualization vs. Containerization: Key Differences Explained]]></title>
            <link>https://medium.com/@anouar_bakouch/virtualization-vs-containerization-key-differences-explained-702a1da9c2aa?source=rss-32e75440c3c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/702a1da9c2aa</guid>
            <category><![CDATA[docker]]></category>
            <category><![CDATA[virtualization]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[containers]]></category>
            <dc:creator><![CDATA[Anouar Bakouch]]></dc:creator>
            <pubDate>Sat, 30 Nov 2024 02:40:25 GMT</pubDate>
            <atom:updated>2024-11-30T02:40:25.256Z</atom:updated>
            <content:encoded><![CDATA[<p>Both Virtual Machines (VMs) and containers allow software to run safely side by side on the same physical machine without interference. While they achieve similar goals, there are key technical differences that make their use cases distinct.</p><h4>Virtualization</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/560/0*SKRqSztSguHMPN9_.png" /></figure><p>Virtualization allows software to run safely on the same physical machine by splitting resources like RAM, CPU, and Disk into separate, virtualized units. <br>For example, a 100GB hard disk can appear as four 25GB disks to different software. Both containers and Virtual Machines (VMs) use virtualization technologies.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/986/1*1_k84K0-u8XtJmNwXhTmgw.png" /></figure><h3>Containers vs Virtual Machines</h3><p>The primary difference between <strong>Virtual Machines</strong> (VMs) and <strong>containers </strong>lies in how they handle virtualization. <strong>Virtual Machines </strong>virtualize the entire system, including the hardware, by running a full operating system on top of the physical machine. This means each <strong>VM</strong> operates independently, with its own kernel and resources, providing strong isolation since each VM has its own complete system environment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/995/1*4L3A9SB1z05JJEd2qKuG-Q.png" /></figure><p>In contrast, <strong>containers</strong> virtualize at the operating system level, running as isolated processes within the host OS. Containers share the host’s kernel, which makes them more lightweight and efficient, but this also means they offer less separation compared to VMs. While <strong>containers</strong> are isolated from each other through namespaces and cgroups, they still rely on the host operating system, making them more vulnerable to potential security risks if the host is compromised.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/995/1*5Kwx3G4nSJSyOFbPn2bJ_g.png" /></figure><h3>Conclusion</h3><p>In conclusion, while both Virtual Machines and containers allow software to run on the same physical machine, VMs offer stronger isolation with higher resource use, while containers are more lightweight but offer less separation and may pose security risks. Choosing between them depends on your needs for performance, efficiency, and security.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=702a1da9c2aa" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PLSQL Essentials: Everything You Need to Know]]></title>
            <link>https://medium.com/@anouar_bakouch/plsql-essentials-everything-you-need-to-know-087fba903452?source=rss-32e75440c3c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/087fba903452</guid>
            <category><![CDATA[oracle-database]]></category>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[plsql]]></category>
            <category><![CDATA[relational-databases]]></category>
            <category><![CDATA[sql]]></category>
            <dc:creator><![CDATA[Anouar Bakouch]]></dc:creator>
            <pubDate>Sat, 30 Nov 2024 02:26:17 GMT</pubDate>
            <atom:updated>2024-11-30T02:26:17.313Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*cfrGDSMBfkBkNQsF.png" /></figure><p>PL SQL (Procedural Language/Structured Query Language) is a programming language designed specifically for use with Oracle databases. It enables developers to create custom procedures, functions, and triggers to automate tasks, generate reports, and manipulate data. With PL SQL, users can efficiently manage large datasets, enforce business rules, and perform complex operations with ease.</p><h3>Basics of PL SQL</h3><h4>Default Syntax of a block of code</h4><p>A typical PL SQL block follows a basic structure:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/616/1*w0_ox2sFOAl3uanMXGIQxw.png" /><figcaption>variabls / PL SQL</figcaption></figure><ul><li><strong>DECLARE</strong>: Used to declare variables and objects.</li><li><strong>BEGIN</strong>: Marks the start of the executable code block.</li><li><strong>END</strong>: Marks the conclusion of the block.</li></ul><p>Within the <strong>BEGIN</strong> and <strong>END</strong> section, you can include SQL statements such as SELECT, INSERT, UPDATE, and DELETE.</p><h4>Variables</h4><p>Variables in PL SQL temporarily store data during execution. They are declared within the <strong>DECLARE</strong> section and initialized using the := operator. Common data types include:</p><ul><li><strong>NUMBER</strong>: Stores numeric values.</li><li><strong>VARCHAR2</strong>: Stores variable-length character data.</li><li><strong>DATE</strong>: Stores date and time values.</li><li><strong>BOOLEAN</strong>: Stores true or false values.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/616/1*Y695U0MnGdVkJLpFtKHYkA.png" /><figcaption>variable declaration/initialization</figcaption></figure><p>Another interesting types of variables are :</p><ul><li><strong>Table</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1017/1*fvqbMr3sMPhZUk0-bSoo3Q.png" /><figcaption>table/array</figcaption></figure><p>In PL SQL, you can declare a custom type, such as an array of characters, or modify it to be an array of integers. The “INDEX BY BINARY_INTEGER” clause allows you to index the array starting from 0, 1, 2, and so on.</p><ul><li><strong>Record</strong></li></ul><p>If you’re a data analyst working with Oracle databases, you’ll frequently use this type, as it allows you to declare only the columns needed for retrieval.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AEGpuCpMV5sM1qaZy4yMIg.png" /><figcaption>declaration of a record</figcaption></figure><p>Before using it, let’s explore a useful feature in PL SQL: SELECT INTO. This statement retrieves data from the database and assigns it to variables.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NuAbg8BlAHYNZjFo0hO8MQ.png" /></figure><p>The SELECT INTO statement retrieves data from a single or multiple rows, depending on the WHERE clause. If it returns multiple rows, a TOO_MANY_ROWS exception is raised.</p><h3>Control Structures</h3><ul><li><strong>IF…THEN…ELSE</strong> statement is used for conditional execution</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/592/1*XGQ6xlcZ5RNwTDpIk99uAw.png" /></figure><p>and now let’s look at this example :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/790/1*Wp3Rdk-3_ymZdASfgHrcCg.png" /></figure><p>- <strong>LOOP</strong> statement is used for executing a block of code repeatedly until a condition is met.</p><p>Three types of loops exist in PL SQL :</p><h4>the For Loop</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/709/1*wwRZDveJg4ZvFJucs_h03Q.png" /></figure><p>The FOR loop in PL SQL iterates a specific number of times, using a loop index variable and a defined range of values. It executes the code block once for each value in the range.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/598/1*0ryKBj4F0kLg4nqHnyJy_g.png" /><figcaption>For Loop Example</figcaption></figure><p>Output : <strong>1 2 3 4 5 6 7</strong></p><h4>The While Loop :</h4><p>The WHILE LOOP repeatedly executes a block of code until a specified condition is met.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/598/1*KUOpuUTfYwpJzBrTzK2udA.png" /></figure><h4>Basic Loop</h4><p>The basic loop repeats a block of code until an exit condition is met. Defined with the “LOOP” keyword, it can be exited using the “EXIT” keyword. Since it lacks a specific iteration count, it may loop infinitely if the exit condition is not met.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*avKsfLCML2JWatnztwGGDA.png" /></figure><h3>Exception Handling</h3><p>The EXCEPTION block is used to handle errors and exceptions that occur during execution.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/912/1*EBcNPL0I-_A5oRPktqHUKg.png" /></figure><p>Exception handling in PL/SQL allows errors to be caught and managed when they occur during execution. PL/SQL automatically raises exceptions, which can be handled using an exception handler.<strong>Common exceptions</strong> include <strong>NO_DATA_FOUND</strong>,<strong> TOO_MANY_ROWS…</strong></p><h3>Cursors</h3><p>A cursor in PL/SQL allows you to retrieve and process data one row at a time. It works by defining a SQL statement, opening the cursor to execute it, and then fetching and processing one row at a time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/592/1*sny-b7KM6nlnbZ_Ap725hg.png" /></figure><p>After fetching a row of data into variables, you can process it (e.g., update tables, calculate results, or display information). This continues until there are no more rows to fetch, at which point the cursor is closed.</p><h3>Triggers</h3><p>Triggers are stored procedures that automatically execute in response to specific events.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/799/1*oBKgCw6iLW3o7HA9H9cpFQ.png" /></figure><p>Triggers can be used to enforce business rules, maintain data integrity, and automate tasks , let’s see an example</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*y2rhGirw_PPYEScJZ8xpqg.png" /></figure><p>This SQL trigger is designed to run after an update on the “student” table. It inserts a record into the “log_table” for each updated row, logging the action (“update”), the old value of the “nom” column, and the new value. The trigger appears syntactically correct, assuming “nom” exists in the “student” table and “log_table” is set up for logging.</p><h4>Conclusion</h4><p>PL SQL is essential for Oracle database users, offering powerful tools for automation, data handling, and custom procedures. Mastering it can greatly enhance efficiency and unlock the full potential of Oracle databases.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=087fba903452" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The PageRank algorithm : Google’s Favorite Algorithm]]></title>
            <link>https://medium.com/@anouar_bakouch/the-pagerank-algorithm-googles-favorite-algorithm-c7403446bc2e?source=rss-32e75440c3c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/c7403446bc2e</guid>
            <category><![CDATA[pagerank]]></category>
            <category><![CDATA[google]]></category>
            <category><![CDATA[algorithms]]></category>
            <dc:creator><![CDATA[Anouar Bakouch]]></dc:creator>
            <pubDate>Sat, 21 Oct 2023 12:43:19 GMT</pubDate>
            <atom:updated>2023-10-21T12:43:19.805Z</atom:updated>
            <content:encoded><![CDATA[<h3>The PageRank algorithm : Google’s Favorite Algorithm</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/0*LecELw5hwgFsy910.png" /><figcaption>From Delante : Picture shows The page rank</figcaption></figure><p>The PageRank algorithm is a way of ranking web pages based on their importance. It is used by Google to rank the search results that it displays to users.</p><p>The PageRank algorithm works by counting the number and quality of links to a page. A page with a lot of high-quality links is considered to be more important than a page with a few low-quality links.</p><p>The PageRank algorithm is calculated using the following formula:</p><pre>PR(A) = (1 - d) + d (PR(T1)/C(T1) + ... + PR(Tn)/C(Tn))</pre><p>where:</p><ul><li>PR(A) is the PageRank of page A</li><li>d is a damping factor, usually set to 0.85</li><li>T1 to Tn are the pages that link to page A</li><li>C(Ti) is the number of links on page Ti</li></ul><p>The damping factor is used to prevent the PageRank algorithm from giving too much weight to pages that have a lot of links to them. This is necessary because some pages, such as spam pages, may have a lot of links to them but are not actually very important.</p><p>The PageRank algorithm is calculated iteratively. This means that the algorithm starts with an initial value for the PageRank of each page and then repeatedly calculates the PageRank of each page using the formula above. The algorithm stops when the PageRank values for all of the pages converge.</p><p>The PageRank algorithm is a complex algorithm, but it is very effective at ranking web pages based on their importance. It is one of the main reasons why Google is able to provide such accurate search results.</p><p><strong>Here is an example of how the PageRank algorithm works:</strong></p><p>Suppose there are two web pages, A and B. Page A has a link to page B, but page B does not have any links to page A.</p><p>Initially, the PageRank of both pages is set to 1.0.</p><p>When the PageRank algorithm is calculated, the PageRank of page B is increased because it has a link from page A. The PageRank of page A remains the same because it does not have any links to other pages.</p><p>The PageRank algorithm is then calculated again, and the PageRank of page B is increased again because it has a link from page A. The PageRank of page A remains the same.</p><p>This process continues until the PageRank values for both pages converge. At this point, the PageRank of page B will be higher than the PageRank of page A because page B has a link from page A.</p><p>The PageRank algorithm is a very powerful algorithm for ranking web pages based on their importance.</p><p>It is one of the main reasons why Google is able to provide such accurate search results.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c7403446bc2e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>