<?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 Asharib Ahmed on Medium]]></title>
        <description><![CDATA[Stories by Asharib Ahmed on Medium]]></description>
        <link>https://medium.com/@codeashing?source=rss-ab938e743e7c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*e-kLyU6C4wHIOOmY4l-8YQ.jpeg</url>
            <title>Stories by Asharib Ahmed on Medium</title>
            <link>https://medium.com/@codeashing?source=rss-ab938e743e7c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 01:43:12 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@codeashing/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[Why Go Developers Avoid panic() — And When It’s Actually Okay to Use]]></title>
            <link>https://codeashing.medium.com/why-go-developers-avoid-panic-and-when-its-actually-okay-to-use-c75105dc0958?source=rss-ab938e743e7c------2</link>
            <guid isPermaLink="false">https://medium.com/p/c75105dc0958</guid>
            <category><![CDATA[panic]]></category>
            <category><![CDATA[go]]></category>
            <category><![CDATA[golang]]></category>
            <dc:creator><![CDATA[Asharib Ahmed]]></dc:creator>
            <pubDate>Mon, 05 May 2025 04:25:38 GMT</pubDate>
            <atom:updated>2025-05-05T04:25:38.302Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why Go Developers Avoid panic() — And When It’s Actually Okay to Use</h3><p>If you’re coming to Go from another language, you might be surprised to find that Go developers don’t really throw exceptions. In fact, they mostly avoid Go’s built-in panic() function unless absolutely necessary.</p><p>But that doesn’t mean panic is bad. It just has a specific purpose, and misusing it can lead to rigid, brittle, and unpredictable code.</p><p>Let’s break down why panic() is usually discouraged, when it <em>is</em> acceptable, and how Go’s unique approach to error handling makes your code more explicit, testable, and robust.</p><h3>☠️ What Exactly Is panic()?</h3><p>In Go, calling panic() stops the normal flow of execution. The program begins <strong>unwinding the call stack</strong>, running any defer functions along the way, until it either:</p><ul><li>recovers (via recover()), or</li><li>crashes the program entirely.</li></ul><p>It’s Go’s version of an unrecoverable error — a loud, immediate halt.</p><h3>🧘 Why Returning Errors Is Preferred</h3><p>Go was designed around simplicity and <strong>explicit error handling</strong>. The idiomatic way to handle issues is by returning an error as a second value:</p><pre>result, err := doSomething()<br>if err != nil {<br>    // Handle it<br>}</pre><p>Here’s why that’s usually better than panicking:</p><h3>1. Gives the Caller Control</h3><p>When a function returns an error, the caller has <em>options</em>. They can:</p><ul><li>Retry the operation</li><li>Wrap the error with more context</li><li>Log it</li><li>Ignore it (rarely, but sometimes valid)</li></ul><p>With panic(), all control is lost — unless you catch it with recover(), which is clunky and rarely worth it.</p><h3>2. Easier to Test</h3><p>It’s straightforward to test a function that returns an error:</p><pre>if err := myFunc(); err != nil {<br>    t.Errorf(&quot;unexpected error: %v&quot;, err)<br>}</pre><p>Testing for panics requires a deferred recover() block, which adds boilerplate and confusion to your test code.</p><h3>3. Better for Libraries and APIs</h3><p>If you’re writing a package for others, panicking is a poor user experience. Your panic could crash their entire application — something they didn’t sign up for. Returning an error lets <strong>them</strong> decide what to do.</p><h3>4. More Informative Errors</h3><p>Go makes it easy to wrap errors with context:</p><pre>return fmt.Errorf(&quot;loading config: %w&quot;, err)</pre><p>This provides a breadcrumb trail of what went wrong and where. A panic just dumps a stack trace, which may or may not help.</p><h3>5. It’s the Go Way</h3><p>Finally, Go’s standard library overwhelmingly favors returning errors over panicking. This consistency is part of what makes Go easy to read, maintain, and understand.</p><h3>⚠️ So… When Is panic() Appropriate?</h3><p>Despite all the caution, panic() does have its place — just use it <strong>sparingly</strong> and <strong>intentionally</strong>.</p><p>Here are the legitimate use cases:</p><h3>✅ 1. Truly Unrecoverable Errors</h3><p>If your application reaches a state it should never be in, panicking can be appropriate.</p><pre>func mustGetEnv(key string) string {<br>    val, ok := os.LookupEnv(key)<br>    if !ok {<br>        panic(&quot;missing required env var: &quot; + key)<br>    }<br>    return val<br>}</pre><p>You’re signaling: “This is a developer mistake, not a runtime issue.”</p><h3>✅ 2. During Initialization</h3><p>It’s okay to panic when your program can’t even start correctly.</p><pre>var cfg = loadConfig() // panics if config file is malformed</pre><p>Failing fast during startup is better than limping along in a broken state.</p><h3>✅ 3. For Internal Tooling or Scripts</h3><p>If you’re writing a short CLI tool or internal script, and you’re just validating a few assumptions, panicking can save time — though error handling is still better if you expect others to reuse your code.</p><h3>✅ 4. In Tests</h3><p>In unit tests, panic isn’t harmful — it just fails the test. You can use t.Fatal or even call panic() yourself for early exits in setup failures.</p><h3>🔚 Final Thoughts</h3><p>panic() isn’t evil — it’s just <strong>blunt</strong>. Like a fire alarm, you don’t want to pull it unless things have truly gone off the rails.</p><p>In almost every case, <strong>returning an error is better</strong>:</p><ul><li>It’s more flexible</li><li>Easier to test</li><li>Safer for library consumers</li><li>More aligned with Go’s philosophy</li></ul><p>Stick with explicit error handling, and reserve panic() for when you really mean, “This should <em>never</em> happen.”</p><h3>TL;DR</h3><p>Use Case Recommendation Invalid user input Return error File not found Return error Internal bug Use panic() Missing config Panic at startup Unexpected nil Panic Library package Never panic</p><p>Want to see code examples comparing panic vs error returns? Drop a comment and I’ll follow up with a companion piece. 🚀</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c75105dc0958" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Use AI in Go with LangChainGo (Very Easy!)]]></title>
            <link>https://codeashing.medium.com/how-to-use-ai-in-go-with-langchaingo-very-easy-1263c05c851c?source=rss-ab938e743e7c------2</link>
            <guid isPermaLink="false">https://medium.com/p/1263c05c851c</guid>
            <category><![CDATA[go]]></category>
            <category><![CDATA[genai]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Asharib Ahmed]]></dc:creator>
            <pubDate>Sun, 04 May 2025 05:50:55 GMT</pubDate>
            <atom:updated>2025-05-04T05:50:55.773Z</atom:updated>
            <content:encoded><![CDATA[<p>Have you ever wanted to use AI in your Go (Golang) programs? It may sound hard, but with <a href="https://github.com/tmc/langchaingo">LangChainGo</a>, it’s actually very easy. In this post, I will show you a simple example to get started.</p><h3>What is LangChainGo?</h3><p>LangChainGo is a Go library that helps you use language models (like Google AI, OpenAI, etc.) in your Go programs. You can give it a prompt (a question or a message), and it will give you an answer using AI.</p><h3>What We’ll Build</h3><p>We will write a small Go program that asks AI to explain something in simple terms. In this case, we ask:</p><blockquote><em>“Explain the concept of quantum entanglement in simple terms”</em></blockquote><p>The AI will give us an answer we can print.</p><h3>The Code</h3><p>Here is the full code:</p><pre>package main</pre><pre>import (<br>	&quot;context&quot;<br>	&quot;fmt&quot;<br>	&quot;log&quot;<br>	&quot;os&quot;</pre><pre>	&quot;github.com/tmc/langchaingo/llms&quot;<br>	&quot;github.com/tmc/langchaingo/llms/googleai&quot;<br>)</pre><pre>func main() {<br>	ctx := context.Background()<br>	apiKey := os.Getenv(&quot;API_KEY&quot;)<br>	llm, err := googleai.New(ctx, googleai.WithAPIKey(apiKey))<br>	if err != nil {<br>		log.Fatal(err)<br>	}<br>	prompt := &quot;Explain the concept of quantum entanglement in simple terms&quot;<br>	answer, err := llms.GenerateFromSinglePrompt(ctx, llm, prompt)<br>	if err != nil {<br>		log.Fatal(err)<br>	}</pre><pre>	fmt.Println(answer)<br>}</pre><h3>What This Code Does</h3><ul><li>First, it sets up a context.</li><li>It gets your Google AI API key from the environment (API_KEY).</li><li>Then it creates a connection to the AI using the LangChainGo library.</li><li>It sends a prompt to the AI.</li><li>The AI gives an answer.</li><li>We print the answer.</li></ul><h3>Example Output</h3><p>When I run the code, I get something like:</p><blockquote><em>“Quantum entanglement is when two tiny particles are connected in a special way, so if you do something to one, the other changes instantly, no matter how far apart they are.”</em></blockquote><p>Amazing, right? With just a few lines of code, you can use the power of AI!</p><h3>Final Thoughts</h3><p>LangChainGo makes it very easy to use AI in Go. You don’t need to be an expert in machine learning or AI. If you know some Go, you can get started right away.</p><p>Let me know if you try it out or if you have any questions!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1263c05c851c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building Server-Side Applications with Bun - JavaScript Runtime]]></title>
            <link>https://codeashing.medium.com/building-server-side-applications-with-bun-javascript-runtime-6393144bddf1?source=rss-ab938e743e7c------2</link>
            <guid isPermaLink="false">https://medium.com/p/6393144bddf1</guid>
            <category><![CDATA[bűn]]></category>
            <category><![CDATA[javascript-frameworks]]></category>
            <category><![CDATA[nodejs]]></category>
            <category><![CDATA[server-side-rendering]]></category>
            <category><![CDATA[backend]]></category>
            <dc:creator><![CDATA[Asharib Ahmed]]></dc:creator>
            <pubDate>Sun, 11 Sep 2022 12:47:45 GMT</pubDate>
            <atom:updated>2022-09-11T12:47:45.293Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n2WUFqriUsX-ecWo9zLMbA.jpeg" /></figure><p>Since Node.js came into existence on May 27, 2009, it got exponential growth. From the Tech giants like Google, Amazon, Microsoft, and Meta to the all-new local startups consider Node.js for building the backend of their product if they are working in a JS domain. Some folks switch to Nest.js but it&#39;s also a progressive Node.js framework for building efficient, reliable, and scalable server-side applications.</p><blockquote>I know now you are wondering Why Node.js is popular among developers❓</blockquote><h3>👀 Reasons for Node.js Popularity¹</h3><p>There are seven popular reasons why Node.js are popular among developers:</p><ol><li>It is easy to learn Node.js.</li><li>The scalability offered.</li><li>The Mobile-friendly, Cross-platform, and Dev-Friendly nature of Node.js.</li><li>Node.js is light and fast.</li><li>Many hosting providers are available.</li><li>Highly extensible.</li><li>Its caching ability.</li></ol><p>and the list does not end here, you will get shocked that there are <strong>Over 1.3 million</strong> packages available in the main npm registry which are supported by Node.js. There is a library for almost anything you can imagine. — Why would you use something else?</p><h3>🚀 <strong>Here come the Bun²</strong></h3><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fsiteproxy.ruqli.workers.dev%3A443%2Fhttps%2Fgiphy.com%2Fembed%2FcEYFeE4wJ6jdDVBiiIM%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fsiteproxy.ruqli.workers.dev%3A443%2Fhttps%2Fmedia.giphy.com%2Fmedia%2FcEYFeE4wJ6jdDVBiiIM%2Fgiphy-downsized-large.gif&amp;image=https%3A%2F%2Fsiteproxy.ruqli.workers.dev%3A443%2Fhttps%2Fi.giphy.com%2Fmedia%2FcEYFeE4wJ6jdDVBiiIM%2Fgiphy-downsized-large.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="244" frameborder="0" scrolling="no"><a href="https://medium.com/media/9590c217d16546e124bd780751508eee/href">https://medium.com/media/9590c217d16546e124bd780751508eee/href</a></iframe><p>The bun is a newly released JavaScript runtime that, unlike Node.js and Deno uses the JavaScriptCore engine and not the V8 engine. This is one of the many reasons Bun claims to perform much faster than NodeJS and Deno. Additionally, Bun is written in <a href="https://ziglang.org/">Zig</a>, a low-level programming language used to optimize every part of Bun.</p><p>The bun is available in public beta since early July 2022. But I’ve been part of the closed beta already for quite a few months. And yes, even in its beta, it makes JS development much faster and much more fun compared to Node.js. But that is not just because of its execution speed. The bun is also an NPM client. It is compatible with almost every npm package. You probably know how annoying it is to wait npm install to finish. The solution is simple, with Bun you will not experience this. In the projects I tested it on, it was often 2x faster than npm i and 1.6x faster than yarn.</p><p><strong><em>Pretty amazing right</em></strong>❓</p><blockquote>emmm yeah but Asharib can you stop comparing Node.js with Bun and teach me how to build Server-Side Applications with Bun.</blockquote><h3>🔧 Setup Bun on Linux</h3><p>Open your terminal and type this command to install Bun in your operating system</p><pre>curl <a href="https://bun.sh/install">https://bun.sh/install</a> | bash</pre><p>once your installation is done it will say that you need to add these lines in your bashrc file manually.</p><pre>export BUN_INSTALL=&quot;$HOME/.bun&quot; <br>export PATH=&quot;$BUN_INSTALL/bin:$PATH&quot;</pre><p>to add this with nanotext editor type this command</p><pre>nano  ~/.bashrc</pre><p>it will open the text editor, now go to the bottom and paste the exports command there then press control + o and then press Enter</p><p>To confirm your installation type bun -vand it will show you the currently installed version.</p><h3>💣 Get started with Server-Side Applications</h3><p>To get started with a minimal project and tries to guess sensible defaults, create a new folder and open your terminal in that folder, and type:</p><pre>Bun init</pre><p>it will ask you two things, one is the package name and the second is your entry point. you can just simply pressenterto keep the settings as default.</p><p>now your folder contains five files:</p><ol><li>index.ts</li><li>package.json</li><li>readme.md</li><li>tsconfig.json</li><li>.gitignore</li></ol><p>Now you need to install <strong>Hono</strong></p><blockquote><strong>Hono — <em>[炎] means flame🔥 in Japanese</em> — is a small, simple, and ultrafast web framework for Cloudflare Workers, Deno, Bun, and others. Fast, but not only fast.</strong></blockquote><p>Install Hono by this command:</p><pre>bun install hono</pre><p>now its time to add some code snippets to your index.ts</p><pre>import { Hono } from &#39;hono&#39;</pre><pre>import { cors } from &#39;hono/cors&#39;</pre><pre>import { prettyJSON } from &#39;hono/pretty-json&#39;<br></pre><pre>const app = new Hono();</pre><pre>app.use(&#39;*&#39;, prettyJSON())</pre><pre>app.use(cors())</pre><pre>const port = parseInt(process.env.PORT) || 3000;</pre><pre>console.log(`Running at http://localhost:${port}`);</pre><pre><em>export</em> default {</pre><pre>port: 3000,</pre><pre>fetch: app.fetch,</pre><pre>}</pre><p>This code does nothing but import Hono and allow CORS and run it on a port of 3000.</p><p>Now set your first root endpoint which will return JSON objectcontaining a <strong>Hello World</strong> message.</p><pre>app.get(&quot;/&quot;, (req) =&gt; {</pre><pre>return req.json({</pre><pre>&#39;status&#39;: &#39;success&#39;,</pre><pre>&#39;data&#39;: null,</pre><pre>&#39;message&#39;: &#39;Hello World!&#39;,</pre><pre>&#39;code&#39;: 200,}, 200);</pre><pre>});</pre><h4>🍒 Adding a cherry on top</h4><p>you can add a global error handler that will return specific JSON when any unhandled error occurs</p><pre>app.onError((err, c) =&gt; {</pre><pre>console.error(`${err} on line ${err.stack}`);</pre><pre>return c.json({</pre><pre>&#39;status&#39;: &#39;error&#39;,</pre><pre>&#39;data&#39;: null,</pre><pre>&#39;message&#39;: `${err} on line ${err.stack}`,</pre><pre>&#39;code&#39;: 500,</pre><pre>}, 500)</pre><pre>})</pre><p>Also, add function will handle a response when the client hit an endpoint that does not exist</p><pre>app.notFound((c) =&gt; c.json({</pre><pre>&#39;status&#39;: &#39;error&#39;,</pre><pre>&#39;data&#39;: null,</pre><pre>&#39;message&#39;: `Not Found`,</pre><pre>&#39;code&#39;: 404,</pre><pre>}, 404))</pre><p>now your index.tswill look like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LdAXZFBn-ANpan4GGfzUJQ.png" /></figure><h4>🐎 Final Move</h4><p>run this command and your first basic BUN-based server-side application is up and running:</p><pre>bun run index.ts</pre><p>Your application is running on port 3000, you can check by sending Get a request by Postman or you can simply type the URL on your browser.</p><blockquote>Yuurryyyyyy!!!😎😎 we did it. We have succecfully developed our basic server-side application….</blockquote><blockquote>🤠 Its party timeeeeeeee!!!</blockquote><p>👏 Take some time to clap for you because You are Amazing and yeah Stay tuned for other tutorials.</p><blockquote>Thanks for reading this! If you enjoyed this article make sure to leave a 👏, and if you need any other information or want to request a topic I should cover, feel free to drop a comment below!</blockquote><blockquote>You can also <a href="https://twitter.com/asharib90">follow me on Twitter.</a></blockquote><h4>🖇 References</h4><p>[1]: Geoffrey Mungai. Why is Node.js wildly popular among developers?<em> </em><a href="https://www.section.io/engineering-education/why-node-js-is-popular/"><em>LINK</em></a></p><p>[2]: Yannis. (July 27, 2022). Node.js is Dead. — Long Live Bun<em> </em><a href="https://medium.com/@YNNSme/node-js-is-dead-long-live-bun-3d888eab63d7"><em>LINK</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6393144bddf1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to train AI Beast]]></title>
            <link>https://codeashing.medium.com/how-to-train-ai-beast-888d6169dda?source=rss-ab938e743e7c------2</link>
            <guid isPermaLink="false">https://medium.com/p/888d6169dda</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[unsupervised-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[supervised-learning]]></category>
            <dc:creator><![CDATA[Asharib Ahmed]]></dc:creator>
            <pubDate>Tue, 08 Mar 2022 21:01:35 GMT</pubDate>
            <atom:updated>2022-03-08T21:01:35.461Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/736/1*qMgkdZfySnKqdqDBJ-4NiQ.jpeg" /></figure><p>People from other fields than AI keep asking me, how you let your AI model know to recognize only those things what you showed him in training? How its actually done?</p><p>So I though why not write a short article on it🤷🏻‍♂️… and here I’m😎…!</p><p>Well preparing any profound learning model is the primary piece of each artificial intelligence pipeline. For preparing artificial intelligence models we have two significant methods that are utilized most often which are Supervised learning and Unsupervised learning. These 2 also have few sub categories like regression and clustering but don’t worry we will discuss it as well.</p><h4>👨🏻‍🏫 <strong>Supervised learning</strong></h4><p>Supervised learning is an AI approach that is characterized by its utilization of named datasets. These datasets are intended to prepare or “regulate” calculations into ordering information or anticipating results precisely. Utilizing marked information sources and results, the model can gauge its exactness and learn over the long run.</p><p>Supervised learning can be isolated into two kinds of issues while information mining: Classification and Regression.</p><blockquote>Classification problems 🧩</blockquote><p>Characterization issues utilize a calculation to precisely dole out test information into explicit classifications, for example, isolating apples from oranges. Or then again, in reality, Supervised learning calculations can be utilized to arrange spam in a different organizer from your inbox. Straight classifiers, support vector machines, choice trees and arbitrary woodland are generally normal kinds of arrangement calculations.</p><blockquote>Regression🪜</blockquote><p>Regression is one more sort of supervised learning technique that utilizes a calculation to get the connection among reliant and autonomous factors. Regression models are useful for anticipating mathematical qualities in view of various informative elements, like deals income projections for a given business. Some well known relapse calculations are straight relapse, strategic relapse and polynomial relapse.</p><h4>💫<strong>Unsupervised learning</strong></h4><p>To deconstruct and categories unlabeled informational indices, unsupervised learning employs AI computations. These calculations uncover hidden examples in data without the need for human intervention (henceforth, they are “unaided”).</p><p>For three main tasks, unsupervised learning models are used: Clustering, Association, and dimensionality reduction.</p><blockquote>Clustering🎳</blockquote><p>Clustering is a data-mining technique for aggregating unlabeled data based on similarities or differences. For example, Kmean clustering computations distribute comparable informative items into gathers, with the K value addressing the size and granularity of the gathering. This method is beneficial for market segmentation, image pressure, and so on.</p><blockquote>Association🧬</blockquote><p>Association is another type of unsupervised learning approach that use a variety of criteria to discover relationships between variables in a dataset. As suggested by “Clients Who Purchased This Thing Also Purchased” ideas, these tactics are frequently used for market bin evaluation and proposition motors.</p><blockquote>Dimensionality reduction🪚</blockquote><p>When the number of components (or aspects) in a dataset is extremely large, dimensionality reduction is a learning strategy used. It reduces the number of information contributions to a manageable level while maintaining the integrity of the data. This approach is frequently used at the preprocessing information stage, such as when autoencoders remove disturbance from visual data to improve picture quality.</p><blockquote>YyyYoOoOoo thats It. Thank you💖</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=888d6169dda" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Levels of Natural Language Processing ]]></title>
            <link>https://codeashing.medium.com/levels-of-natural-language-processing-7a6b1fc00a7a?source=rss-ab938e743e7c------2</link>
            <guid isPermaLink="false">https://medium.com/p/7a6b1fc00a7a</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[naturallanguageprocessing]]></category>
            <dc:creator><![CDATA[Asharib Ahmed]]></dc:creator>
            <pubDate>Wed, 02 Mar 2022 21:37:55 GMT</pubDate>
            <atom:updated>2022-03-02T21:37:55.974Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uNdWwyi-5WKoEc9GwDjBDw.jpeg" /></figure><p>The most informative technique for introducing what really occurs inside a Natural Language Processing system is through the ‘levels of language’ approach. This is likewise alluded to as the synchronic model of language and is recognized from the prior successive model, which guesses that the degrees of human language handling follow each other in a stringently consecutive way. Psycho linguistic examination proposes that language handling is considerably more powerful, as the levels can interface in an assortment of orders. Reflection uncovers that we often use data we gain based on what is normally considered a more significant level of handling to aid a lower level of examination. For instance, the practical information that the archive you are perusing is about science will be utilized when a specific word that has a few potential faculties is experienced, and the word will be deciphered as having the science sense. Of need, the accompanying depiction of levels will be introduced consecutively. The central issue here is that importance is conveyed by every single degree of language and that since people have been displayed to utilize all degrees of language to acquire understanding, the more skilled a NLP framework is, the more degrees of language it will use.</p><p>so the processing levels we will discuss are Phonology, Morphology, Syntax, Semantics, Pragmatics and Lexical!👨🏻‍🍳</p><p><em>ooppssssss sorry I used chef emoji and now i’m thinking this 👨🏻‍🏫 is more suitable but whatever….</em></p><blockquote>Phonology 🔊</blockquote><p>This level arrangements with the translation of discourse sounds inside and across words. There are, indeed, three sorts of rules utilized in phonological investigation:</p><p><em>Phonetic rules:</em> It is used for sound within words.</p><p><em>Phonemic rules</em> : It is used for variations of pronunciation when words are spoken together.</p><p><em>Prosodic rules</em> :<strong> </strong>It is used to check for fluctuation in stress and intonation across a sentence.</p><p>In a NLP framework that acknowledges spoken input, the sound waves are dissected and encoded into a digitized signal for understanding by different standards or by correlation with the specific language model being used.</p><blockquote>Morphology<strong> </strong>🌳</blockquote><p>Morphology is the primary phase of investigation whenever input has been gotten. It takes a gander at the manners by which words separate into their parts and how that influences their syntactic status. Morphology is fundamentally valuable for distinguishing the grammatical features in a sentence and words that associate together. The accompanying statement from Forsberg gives a little foundation on the area of morphology.</p><p>Morphology is an efficient depiction of words in a characteristic language. It portrays a bunch of relations between words’ surface structures and lexical structures. A word’s surface structure is its graphical or spoken structure, and the lexical structure is an investigation of the word into its lemma (otherwise called its word reference structure) and its syntactic depiction. This errand is all the more definitively called inflectional morphology. Having the option to distinguish the grammatical form is fundamental to recognizing the syntactic setting a word has a place with. In English, normal action words have a ground structure with a restricted arrangement of changes, in any case, unpredictable action words don’t adhere to these alteration guidelines, and extraordinarily increment the intricacy of a language. The data accumulated at the morphological stage readies the information for the linguistic stage which looks all the more straightforwardly at the objective language’s syntactic construction.</p><p><em>Ohh GOD!!! Why NLP is so complicated </em>🏴‍☠️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/301/1*LoKBGFq-bzxjiHlw6Zi0CA.jpeg" /></figure><p><em>but do bear with me…</em></p><blockquote>Syntax 🧮</blockquote><p>Punctuation includes applying the standards of the objective language’s syntax, its errand is to decide the job of each word in a sentence and arrange this information into a construction that is all the more handily controlled for additional investigation. Semantics are the assessment of the significance of words and sentences.</p><p><strong>Grammar</strong>📒</p><p>In English, an articulation comprises of a thing expression, an action word state, and at times, a prepositional expression. A thing expression addresses a subject that can be summed up or recognized by a thing. This expression might have articles and descriptive words or potentially an installed action word state as well as the actual thing. An action word express addresses an activity and may incorporate an installed thing phrase alongside the action word. A prepositional expression depicts a thing or action word in the sentence. Most of normal dialects are comprised of various grammatical forms basically: action words, things, descriptors, intensifiers, conjunctions, pronouns and articles.</p><h4>Parsing<strong> </strong>⛓️</h4><p>Parsing is the most common way of changing over a sentence into a tree that addresses the sentence’s syntactic design. The assertion: “The green book is perched on the work area” comprises of the thing expression: “The green book” and the action word state: “is perched on the work area.” The sentence tree would begin at the sentence level and separate it into the thing and action word express. It would then name the articles, the modifiers and the things. Parsing decides if a sentence is substantial according to the language’s punctuation rules.</p><blockquote>Semantics<strong> 💫</strong></blockquote><p>It develops a portrayal of the items and activities that a sentence is depicting and incorporates the subtleties given by modifiers, qualifiers and relational words. This cycle assembles data essential to the commonsense examination to figure out which importance was planned by the client.</p><blockquote>Pragmatics 👨‍👧‍👦</blockquote><p>Pragmatics is “the examination of the genuine importance of an expression in a human language, by disambiguating and contextualizing the expression”. This is achieved by recognizing ambiguities experienced by the framework and settling them utilizing at least one kinds of disambiguation strategies .</p><h4><strong>Ambiguity</strong>🛡️</h4><p>Uncertainty is clarified as “the issue that an expression in a human language can have more than one potential importance.</p><p><strong>Types of Ambiguity</strong>⚔️</p><p>Syntactic Uncertainty is available when more than one parse of a sentence exists. “He lifted the branch with the red leaf.” The action word expression might contain “with the red leaf” as a component of the implanted thing phrase portraying the branch or “with the red leaf” might be deciphered as a prepositional expression depicting the activity rather than the branch, inferring that he utilized the red leaf to lift the branch.</p><ul><li>Semantic Vagueness is existent when more than one potential importance exists for a sentence as in “He lifted the branch with the red leaf.” It might imply that the individual being referred to utilized a red leaf to lift the branch or that he lifted a branch that had a red leaf on it.</li><li>Referential Vagueness is the aftereffect of alluding to something without unequivocally naming it by utilizing words like “it”, ‘he” and “they.” These words require the objective to be turned upward and might be difficult to determine, for example, in the sentence: “The point of interaction sent the fringe gadget information which made it break”, it could mean the fringe gadget, the information, or the point of interaction.</li><li>Nearby Vagueness happens when a piece of a sentence is muddled however is settled when the sentence overall is inspected. The sentence: “this lobby is colder than the room,” embodies nearby equivocalness as the expression: “is colder than” is endless until “the room” is defined.</li></ul><blockquote>Lexical<strong> </strong>📊</blockquote><p>In Lexical, people, as well as NLP frameworks, decipher the significance of individual words. Various kinds of handling give to word-level arrangement — the first of these being a grammatical feature tag to each word. In this interaction, words that can go about as more than one grammatical feature are allocated the most likely grammatical feature label in view of the setting in which they happen. At the lexical level, Semantic portrayals can be supplanted by the words that make them mean. In NLP framework, the idea of the portrayal fluctuates as per the semantic hypothesis deployed</p><p><strong>Finally its comes to end!!</strong>🙏🏻</p><p><em>Yooooo!!! its to much to digest hmmm???</em></p><p>That’s all folks. I hope it helped you a little bit, just a little little bit! <em>Just like bit in data bus ;p</em></p><h4>Thanks ya’ll 🤜🏻❤ 🤛🏻</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7a6b1fc00a7a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Adopt 10x approach in Computer Science Degree…!]]></title>
            <link>https://codeashing.medium.com/adopt-10x-approach-in-computer-science-degree-9767ee7caf76?source=rss-ab938e743e7c------2</link>
            <guid isPermaLink="false">https://medium.com/p/9767ee7caf76</guid>
            <category><![CDATA[10x-thinking]]></category>
            <category><![CDATA[information-technology]]></category>
            <category><![CDATA[future]]></category>
            <category><![CDATA[computer-science]]></category>
            <category><![CDATA[bachelor]]></category>
            <dc:creator><![CDATA[Asharib Ahmed]]></dc:creator>
            <pubDate>Thu, 25 Mar 2021 18:28:23 GMT</pubDate>
            <atom:updated>2021-03-25T18:28:23.516Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*KB9WxJBbdcREuz19eXXbYA.jpeg" /></figure><blockquote>Key Reason of writing this <strong><em>Article</em></strong></blockquote><p>I’m pursuing my bachelors in Computer Science from last 2 and a half year, what I have seen in my fellows that there are numerous students who end up with lower job positions or having less skills after their degree because they didn’t got the proper counseling regarding their field or market required skills!</p><p>So, I will share 10x approach with y’all how you folks can pursue your degree in a way that will make you ready for the interview in Google, Facebook type companies and what are the key courses that you have to most focus on and some tips and tricks for the better skills development :)</p><blockquote>Discussion 💬</blockquote><p>My end goal is to make you comprehend that CS has whole universe in it, there is nothing impossible to do, you just desire a proper path way. so, we’ll be talking over following things which you can follow to become a master in <strong><em>Computer Science</em> </strong>😎</p><ul><li><em>Most worth full Concepts</em></li><li><em>Projects you have to build</em></li><li><em>How to become better programmer</em></li><li><em>How to get hired in tech giants</em></li><li><em>Soft Skills</em></li></ul><p>Let’s get Started 🤩</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fsiteproxy.ruqli.workers.dev%3A443%2Fhttps%2Fgiphy.com%2Fembed%2F2uIlaHVsql55CLP3as%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fsiteproxy.ruqli.workers.dev%3A443%2Fhttps%2Fgiphy.com%2Fgifs%2Fhuntforthetrumptapes-excited-ready-tweet-2uIlaHVsql55CLP3as&amp;image=https%3A%2F%2Fsiteproxy.ruqli.workers.dev%3A443%2Fhttps%2Fmedia1.giphy.com%2Fmedia%2F2uIlaHVsql55CLP3as%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="244" frameborder="0" scrolling="no"><a href="https://medium.com/media/433b413fd3cf371f3a7bf2123f9483da/href">https://medium.com/media/433b413fd3cf371f3a7bf2123f9483da/href</a></iframe><blockquote><strong>🕯 Beginning of your 4 years journey</strong></blockquote><p>In very first semester mostly universities will teach you the course of Introduction to programming fundamental, Applied physics, programming language like C or C++ and Ms Office. They are just giving you an synopsis regarding these concepts, they never took you to the deeper concepts like the main reason of having a course of Applied physics is to implementation of physics principles in computer science but they teach you just momentum, force and newtons laws. so don’t just rely on your university, do learn about gravity equations, Navier stoke equation by your own which is gonna enable you build simulation. In fact Microsoft, Google have their own simulation system in which they test computer science design principle and their underlying principle which is completely based on applied physics. More over when you play any game you see Air, water, motions and movements they all are develop by inverse and forward cymatics and you can see it by your own if you download the standard assets or character controller of unity and see its script written on C#, so you’ll find the gravity equation there. This is why applied physics is crucial in CS and so you have to comprehend these concepts.</p><blockquote>⚒️ <strong>Project Idea: </strong>Build your own 2D cylindrical game engine on C++ in which you can render object and change their shapes.</blockquote><h3>📐 <strong>Single and multi variable Calculus</strong></h3><p>If you have any interest in Artificial Intelligence so you have to consider this course extremely significant because machine learning is nothing but math and if you go in a bit deeper so you’ll find calculus concept’s like derivatives, integration etc. like we can’t do back propagation in neural network without chain rule. In CS you&#39;ll see calculus principles on and off in many places.</p><h3>💪🏻 <strong>Power of C++</strong></h3><p>C++ is versatile language, you can build really amazing stuff by using C++. Our world wide web is completely written on C++ but still our universities asked you to create simple console base read and write tool which can read an employee name, age etc and write it on a .txt file. but for God sake go for something creative which is based on GUI, not go for console based tool because this habit will stuck with you till the end of your degree. In universities they will just teach you C++ up to functions and classes but you have to discover about GPU programming, Directx programming, Open GL programming, Web GL programming type thing by your own because if you gonna apply for a internship at google after 3rd or 4th semester so they will asked you for you semester projects and if you gonna exhibit them console based application, they will definitely kick you out but on the other hand if you have something like TCP/IP socket server as your semester project then they will consider you as a potential candidate.</p><blockquote>🛠️ <strong>Project Ideas :</strong></blockquote><ul><li><em>Develop game using Simple and Fast Multimedia Library.</em></li><li><em>First person shooter game on unreal engine.</em></li></ul><h3>📏 <strong>Probability and Statistics</strong></h3><p>This course is the essential for Artificial Intelligence developer. Entire <em>Machine Learning </em>works on probability and statistics principles’ like K nearest neighbors, Linear regression, Random Forest etc. No one can survive in AI without having a understanding of this course. You can initiate python library like <strong><em>Pstats </em></strong>as a semester project <strong><em>, </em></strong>which can implement stats function mean, mode and Standard Deviation.</p><blockquote>🧰 <strong>Project Idea :</strong></blockquote><ul><li><em>Create your own Algorithm (with team).</em></li><li><em>Solve real world problem related to </em><strong><em>Fintech Market</em></strong></li></ul><h3>🎯 <strong>Digital Logic and Design (DLD)</strong></h3><p>This course is lovely and bit easier as compare to others. You have tones of amazing projects on which you can work on. You can build keyloggers in hardware, signal jammers , WiFi authors etc. These are the projects which can teach you real world material ground truth like how signal jammers work, how WiFi get disabled, how WiFi access point stable and how WiFi connects with the help of access points. these mini project can help you to get command on hardware programming.</p><h3>🧮 <strong>Linear Algebra</strong></h3><p>This is also a prerequisite of Artificial Intelligence. Egin values and Egin vectors implementation in any programming language is the primary concept that you have to discover. Coordinators transformation is also vital if you wanna go for robotics including rotation matrix, rotation vectors etc.</p><h3>🖱️ <strong>Object Oriented Programming (OOP)</strong></h3><p>Universities mostly prefer <em>JAVA </em>for OOP but you are in good university so they will teach you in C# otherwise. You need to learn multi-threading in this, which use to run single code on a multiple machines. Do try to learn about GPU programming, if you have NVIDIA graphic card then download <em>Compute Unified Device Architecture (CUDA) </em>and learn its programming. Because if you know GPU programming so you can increase any device speed, you can decrease their run-time etc and almost every tech giant required GPU programming because they are using handling large scale traffic every day.</p><blockquote>🔧 <strong>Project Idea : </strong>GUI App management by using (Java/Visual Basic)</blockquote><h3>📊 <strong>Discrete Mathematics</strong></h3><p>In this course you have to learn about inductive and deductive logic because this is the time when you are suppose to showcase yourself to the world by Competitive Programming. CP is the only way by which you can prove yourself to the world. Gennady Korotkevich is amazing example of this, who is the world best competitive programmer, the guy who rejected Microsoft, Google and Facebook type companies’ offer. And this course is a prerequisite for competitive programming so you have to take this seriously.</p><p>Following are the platforms from where you can polish your competitive programming skills.</p><p>🔥 <strong>Platforms :</strong></p><ul><li><em>Hacker Rank</em></li><li><em>Leetcode</em></li></ul><p>following are the competitions that you can be a part of.</p><p>⚔️ <strong>Competitions :</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3eZ0XiKRxoeJ0vN2UVbatQ.jpeg" /></figure><ul><li><em>Google hash code and code jam</em></li><li><em>Facebook Hacker Cup</em></li><li><em>International Collegiate Programming Contest</em></li></ul><h3>💾 <strong>Database Administration</strong></h3><p>So, data structure is the only concept what you will deal with in every IT related job. If you are a developer so I guaranteed you that one day time will be come when you have to deal with Databases. So you have to be prepared for it. Do try to create your own database with few features as a semester project rather than just using pre-build database.</p><h3>🗣️ <strong>Soft Skills</strong></h3><p>We have been discussing about the concepts and project that you to focus on but there is main key on which you have to work on and that key is <em>SOFT SKILLS. </em>Search Results. Soft skills are attributes that enable you to engage in meaningful interactions with others. Since most jobs require teamwork, it’s important to possess soft skills to enhance your employability and achieve your dream job.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*9q3ehYBvLkWZwivPE23Vcg.jpeg" /></figure><blockquote><strong>Yooooo!!! its to much to digest hmmm???</strong></blockquote><p>No worries you have four years for this so just stay concentrated on what you are doing and if you gain good understanding of all these concepts and build likewise projects then congratulation now you are ready to join GOOGLE, FACEBOOK and MICROSOFT.</p><blockquote><strong>Resources</strong></blockquote><ul><li><em>For any Book </em><a href="https://www.pdfdrive.com/"><em>LINK</em></a></li><li><em>For DLD Projects </em><a href="https://maltronics.com/"><em>LINK</em></a></li><li><em>For system Design </em><a href="https://github.com/donnemartin/system-design-primer"><em>LINK</em></a></li></ul><p><strong>Weeyy Haaa!! 🤠</strong> We are done and you are a completely ready to be a part of Google, Facebook and Microsoft!</p><p>Congratulations!! 🥳</p><p>That’s all from my side, I hope anything from above trash would help you. Good luck for your future endeavor. And Yeah one thing I forgot:</p><h3>Please Give Some Claps!! 👏 Please..!! AaaaHAHA0HHAHA! 😭</h3><p>HAHAH! Just kidding :)</p><p>Special thanks to <a href="https://www.linkedin.com/in/saad-haxxan/">SAAD HASSAN</a> and <a href="https://www.linkedin.com/in/mhamzadev/">M HAMZA</a> for helping me on this 😍</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9767ee7caf76" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>