<?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 Raman Rounak on Medium]]></title>
        <description><![CDATA[Stories by Raman Rounak on Medium]]></description>
        <link>https://medium.com/@raman-rounak?source=rss-7f88b6e6033c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*aWof-slzuCIO_SylDrzaAw.jpeg</url>
            <title>Stories by Raman Rounak on Medium</title>
            <link>https://medium.com/@raman-rounak?source=rss-7f88b6e6033c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 19:47:32 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@raman-rounak/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[From Pixels to Predictions: Unraveling Convolutional Neural Networks and the Magic of Transfer…]]></title>
            <link>https://pub.towardsai.net/from-pixels-to-predictions-unraveling-convolutional-neural-networks-and-the-magic-of-transfer-41f5a30db47d?source=rss-7f88b6e6033c------2</link>
            <guid isPermaLink="false">https://medium.com/p/41f5a30db47d</guid>
            <category><![CDATA[transfer-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[convolutional-network]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Raman Rounak]]></dc:creator>
            <pubDate>Wed, 14 Jun 2023 05:29:21 GMT</pubDate>
            <atom:updated>2023-06-16T14:08:16.270Z</atom:updated>
            <content:encoded><![CDATA[<h3>From Pixels to Predictions: Unraveling Convolutional Neural Networks and the Magic of Transfer Learning</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GARXlnitfEQS2zIy" /><figcaption><strong>Photo by </strong><a href="https://unsplash.com/@dizzyd718?utm_source=medium&amp;utm_medium=referral"><strong>Drew Dizzy Graham</strong></a><strong> on </strong><a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral"><strong>Unsplash</strong></a></figcaption></figure><h3>Introduction</h3><p>Hey there! Welcome to the AI wonderland, where mind-blowing technology is rocking the accuracy and brainpower charts. In this article, we’re diving headfirst into the captivating realm of Computer Vision — a fancy name for machines that can see, understand their surroundings and make spot-on predictions. And guess what? Our heroes on this journey are none other than <strong>Convolutional Neural Networks (CNNs)</strong> and the mind-boggling power of transfer learning. Buckle up, folks, as we embark on a wild ride to decode these concepts and even learn how to code our very own image detection algorithm using top-notch CNNs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qc4K8Tb_oCmfFqT1cAJadg.png" /><figcaption><strong>Image generated by AI</strong></figcaption></figure><h3>Pixels and the Brain</h3><p>Let’s kick things off by drawing a fun parallel between the human brain and CNNs. You know how our brain processes visuals, right? When we see an object or image, the optical nerve sends signals to different layers in the <strong>temporal cortex</strong>. Each layer chips in by extracting essential features like color and shape. Well, guess what? CNNs work similarly, mimicking the activity of neurons with some nifty math tricks. These algorithms feast on image pixels, just like our brain feasts on visual stimuli.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4u8WzNnFMrip7vVJkcvR5Q.png" /><figcaption><strong>Image depicting analogy between Neural Network and human brain</strong></figcaption></figure><h3>Unleashing the Power of CNNs</h3><p>While Artificial Neural Networks (ANNs) are jack-of-all-trades, CNNs are the rock stars of image detection. Images are made up of pixels, which can be grayscale or have three channels (red, green, blue — RGB). Each pixel value ranges from 0 to 255, representing different shades of color. In the world of CNNs, we feed these pixel values into the network as tensors or matrices. To get the data all cozy and ready for action, we use a <strong>min-max scaler</strong> that scales the RGB values between 0 and 1. This step sets the stage for further processing wizardry.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/521/1*acc3RIUMx09i6T_Vd7VMOQ.png" /><figcaption><strong>MinMax Scaler</strong></figcaption></figure><h3>Filters and Feature Extraction</h3><p>Now let’s meet the real heroes of CNNs: the <strong>filters or kernels</strong>. These bad boys have specialized tasks, like sniffing out horizontal or vertical edges in an image. They go all Matrix on the input image, convoluting and extracting precious information. Picture these filters as matrices that slide across the original image, giving us an output of a smaller size. After convolution, we slap on another min-max scaler to normalize the output. For example, a vertical edge detector would produce an output tensor with funky variations in RGB values, waving its hands to say, “Hey, there’s an edge here!” Just like drones use similar filters to dodge obstacles and fly like pros.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/820/1*Cd8tXLnNwr_FANMS4PTuOQ.png" /><figcaption><strong>Vertical Edge Filter</strong></figcaption></figure><h3>Stacking Layers for Deeper Insights</h3><p>One layer of a CNN isn’t enough to gulp down all the information from an image. Nope, we stack multiple layers on top of each other, each one nibbling on different types of tasty info. But you might be thinking, “What about losing important stuff in the final output?” Great question! That’s where <strong>padding</strong> comes to the rescue. The padding adds an extra layer around the input image, saving us from information loss and keeping the output matrix size intact. There are fancy padding flavors like all-zero padding or nearest neighbor padding. Regardless of the type, the convolution process remains the same. Filters cruise through the padded image, gobbling up essential details.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*6ARDrzn0-juG5GVgxNpGKQ.png" /><figcaption><strong>Convolutional Neural Networks stacked on each other</strong></figcaption></figure><h3>Pooling the Most Useful Information</h3><p>On our CNN journey, we stumbled upon another game-changer called <strong>MAX POOLING</strong>. This fancy layer cherry-picks the most relevant and distinctive features from the output matrix. Imagine having an input image with three cats. The perfect kernel would produce an output that highlights those mesmerizing eyes or other unique cat features. MAX POOLING then snatches up these key features, giving the boot to any irrelevant noise. It’s like capturing the essence of the image while kicking out the nonsense.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6u3QK9M9ZJws58P4SAr2AQ.png" /><figcaption><strong>MaxPooling Layer</strong></figcaption></figure><h3>Transitioning to ANNs</h3><p>After all the convolution and pooling fun, we reach the final tensor layer, ready to party with an Artificial Neural Network (ANN). But hold your horses, because ANNs prefer their inputs in 1D format. No worries, though! We have a trick up our sleeves called <strong>flatten</strong>. It takes the final matrix and magically turns it into a 1D array. This array is then handed over to an ANN layer, like a Sequential layer with dense neurons. In this architecture, every neuron in one layer is connected with every neuron in the next layer. This connection extravaganza enables the network to classify the image based on the fantastic features it has processed. The intricate details of ANNs deserve their shindig, so we’ll save that for another day.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*CgbjEQWaloYjhoKIp1uItQ.png" /><figcaption><strong>Flattening Layer</strong></figcaption></figure><h3>Enhancing Accuracy with Transfer Learning</h3><p>Now, as impressive as CNN models are, sometimes their accuracy, hovering around 80%, might not cut it for certain applications. We’re talking about the real champs in competitions like <strong>ImageNet</strong>, where accuracy rates shoot past 85%. But fear not! Here comes transfer learning to save the day. ImageNet hosts a competition where participants develop algorithms that can accurately classify thousands of images with over 90% accuracy. How do they do it? By leveraging pre-trained CNN models that have devoured massive datasets. These models share their knowledge, allowing us to achieve jaw-dropping accuracy even with limited training data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*WmWlaoLZxdXGhd_W_hbEZw.jpeg" /><figcaption><strong>Evolution of state-of-the-art algorithms</strong></figcaption></figure><h3>Modifying Pre-trained Models</h3><p>Models like <strong>VGG 16 </strong>follow a similar process, but their output layer spits out thousands of output shapes due to their pre-training on such large datasets. But sometimes, we need a sleek object detection model with fewer outputs than what the original model offers. No problemo! In such cases, we give the pre-trained model a little makeover. We take off the top layer and tweak it to match our project’s specific needs. This magic trick is called<strong> transfer learning</strong>, where we whisk away the knowledge gained from pre-trained models and pour it into our dataset. The result? Mind-blowing results that’ll leave you grinning from ear to ear.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*HzxRI1qHXjiVXla-_NiMBA.png" /><figcaption><strong>VGG 16 Model</strong></figcaption></figure><h3>Conclusion</h3><p>Convolutional Neural Networks and transfer learning are the rockstars of Computer Vision. By grasping how CNNs process images, extract features, and classify objects, we tap into the mind-blowing potential of AI. And with the mighty power of transfer learning, we can harness pre-trained models to achieve high accuracy even with limited data. So whether you’re a curious AI enthusiast or a seasoned developer, get ready to rock the world of CNNs and unlock the jaw-dropping realm of intelligent image analysis. Let the AI adventure begin!</p><p>The code for the above image detection and Object Classification using CNNs and Transfer Learning are mentioned below:</p><ul><li><a href="https://www.tensorflow.org/tutorials/images/cnn">Convolutional Neural Network (CNN) | TensorFlow Core</a></li><li><a href="https://www.tensorflow.org/tutorials/images/transfer_learning">Transfer learning and fine-tuning | TensorFlow Core</a></li></ul><p>Embark on the quest for AI’s hidden lore, 🚀</p><p>Follow me for captivating knowledge galore. 📚💡</p><p>Unlocking secrets, one article at a time, 🔓</p><p>Join the journey and expand your mind. Ciao! ✨🔍</p><p><a href="https://medium.com/@raman-rounak">Raman Rounak - Medium</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=41f5a30db47d" width="1" height="1" alt=""><hr><p><a href="https://pub.towardsai.net/from-pixels-to-predictions-unraveling-convolutional-neural-networks-and-the-magic-of-transfer-41f5a30db47d">From Pixels to Predictions: Unraveling Convolutional Neural Networks and the Magic of Transfer…</a> was originally published in <a href="https://pub.towardsai.net/">Towards AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Supercharge Your AI: Cracking the Code to Prompt Engineering Magic!]]></title>
            <link>https://pub.towardsai.net/supercharge-your-ai-cracking-the-code-to-prompt-engineering-magic-e1df80c1091f?source=rss-7f88b6e6033c------2</link>
            <guid isPermaLink="false">https://medium.com/p/e1df80c1091f</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[gpt-3]]></category>
            <category><![CDATA[prompt-engineering]]></category>
            <category><![CDATA[chatgpt]]></category>
            <dc:creator><![CDATA[Raman Rounak]]></dc:creator>
            <pubDate>Sun, 11 Jun 2023 14:27:46 GMT</pubDate>
            <atom:updated>2023-06-12T12:43:39.314Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*vvP3PELSuZWlwGiz" /><figcaption>Photo by <a href="https://unsplash.com/@ilgmyzin?utm_source=medium&amp;utm_medium=referral">ilgmyzin</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>What the heck is Prompt Engineering all about?</h3><p>Well, readers, prompt engineering is like the secret sauce for making awesome AI systems. It’s all about crafting prompts that make AI models behave like superstars and do specific tasks with pinpoint accuracy. So, imagine you’re the boss, and these prompts are your instructions to the AI model, telling it exactly what you want. The goal is to keep those prompts short, sweet, and crystal clear, so the model knows what the heck it’s supposed to do and delivers the results you’re looking for.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*H-1-84Toq_Fvrggz" /><figcaption>Photo by <a href="https://unsplash.com/@emilianovittoriosi?utm_source=medium&amp;utm_medium=referral">Emiliano Vittoriosi</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Why is prompt engineering such a big deal?</h3><p>In the world of AI, prompt engineering has become the talk of the town. It’s like the Avengers assembling to save the day! See, with GPT-3, a mind-blowing language model from OpenAI, on the rise, prompt engineering has become even more critical. The quality of those prompts can make or break the model’s performance. Think of it as giving your AI buddy a super clear mission brief. Get it wrong, and you’ll end up with wonky results. But get it right, and you’ll be cheering like a fan at a football game!</p><p>So, hence readers, prompt engineering is the secret ingredient that makes AI systems tick. Clear and concise prompts lead to precise and impressive results. Imagine having a killer script for your AI superstar to follow. That’s what prompt engineering does! It’s like the AI community’s secret weapon, propelling AI models to rockstar status and leaving us all in absolute awe.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Zrn_25wmKqZyD-Dt" /><figcaption>Photo by <a href="https://unsplash.com/@hiteshchoudhary?utm_source=medium&amp;utm_medium=referral">Hitesh Choudhary</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Why does precision matter in prompt engineering?</h3><p>Think of it this way: precise prompts are like giving your AI buddy a detailed map with clear directions. It helps the AI model understand exactly what’s expected of it. When the model gets the gist, it can deliver accurate results like a pro, minimizing errors and boosting overall performance. It’s like having a GPS for your AI system!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qYoT9bbfZ9qsWAGj" /><figcaption>Photo by <a href="https://unsplash.com/@iammottakin?utm_source=medium&amp;utm_medium=referral">Mojahid Mottakin</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>How does prompt engineering boost efficiency?</h3><p>Picture this: well-crafted prompts are like turbo boosters for training an AI model. When the prompts are clear and concise, the model can quickly grasp the desired behavior, cutting down the time and resources needed for training. Prompt engineering is akin to equipping your AI companion with a cheat sheet that propels it to master its domain swiftly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/632/1*SmeRwlT15Q-UiFQdrbJyEA.png" /></figure><h3>What’s the deal with flexibility in prompt engineering?</h3><p>Flexibility is the name of the game! Effective, prompt engineering allows AI models to adapt to different scenarios and shine in various contexts. By crafting prompts that can handle different situations, the model becomes a versatile superstar.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*SU56Dq6VjKDFgwKK" /><figcaption>Photo by <a href="https://unsplash.com/@ikukevk?utm_source=medium&amp;utm_medium=referral">Kevin Ku</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>What are the best practices for prompt engineering?</h3><p>The golden rules! First off, <strong>understanding the task</strong> is key. You need to know what you want your AI model to do before crafting prompts that hit the bullseye. Keep it simple! Avoid confusing the model with fancy jargon or unnecessary details. And don’t forget the power of examples! Show your AI buddy what good behavior looks like, and it’ll catch on like lightning. Lastly, prompt engineering is a journey. Test, tweak, and refine those prompts based on the model’s performance. It’s like fine-tuning your AI friend for maximum awesomeness!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*mKvG5wqGAQd08tVu" /><figcaption>Photo by <a href="https://unsplash.com/@dawson2406?utm_source=medium&amp;utm_medium=referral">Stephen Dawson</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Conclusion</h3><p>Prompt engineering is a critical aspect of building effective AI systems. By crafting well-designed prompts, AI models can perform specific tasks accurately, reducing errors, and improving overall performance. As AI continues to advance, prompt engineering will become even more critical, and organizations that invest in this area will have a competitive advantage in the AI space.</p><p>A fun fact is that the research for this article was also done by using different techniques of Prompt Engineering like Role Prompt, Instruction Prompt, Few shots Prompt, etc. It is obvious that the creativity and the structure of any content will always be original but the time and resources to look for in order to produce those content can surely be replaced with generative AIs.</p><p>To learn more about Prompt Engineering, follow the below link: This will help you to learn prompt engineering from basic to advanced levels with suitable examples in between to illustrate its proper use.</p><p><a href="https://learnprompting.org/docs/intro">Learn Prompting: Your Guide to Communicating with AI</a></p><p>This will help you to learn prompt engineering from basic to advanced levels, with suitable examples in between to illustrate its proper use.</p><p>Embark on the quest for AI’s hidden lore, 🚀</p><p>Follow me for captivating knowledge galore. 📚💡</p><p>Unlocking secrets, one article at a time, 🔓</p><p>Join the journey and expand your mind. Ciao! ✨🔍</p><p><a href="https://medium.com/@raman-rounak">Raman Rounak - Medium</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e1df80c1091f" width="1" height="1" alt=""><hr><p><a href="https://pub.towardsai.net/supercharge-your-ai-cracking-the-code-to-prompt-engineering-magic-e1df80c1091f">Supercharge Your AI: Cracking the Code to Prompt Engineering Magic!</a> was originally published in <a href="https://pub.towardsai.net/">Towards AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Computer Vision and Its Application in Facial Recognition and Object Classification.]]></title>
            <link>https://pub.towardsai.net/computer-vision-and-its-application-in-facial-recognition-and-object-classification-e27cc9af3d0f?source=rss-7f88b6e6033c------2</link>
            <guid isPermaLink="false">https://medium.com/p/e27cc9af3d0f</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[neural-networks]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Raman Rounak]]></dc:creator>
            <pubDate>Thu, 08 Jun 2023 07:15:53 GMT</pubDate>
            <atom:updated>2023-06-10T13:30:36.654Z</atom:updated>
            <content:encoded><![CDATA[<h4>Author: Rounak Raman</h4><h3><strong>Introduction</strong>:</h3><p>Vision is the most crucial sense in the human body, allowing us to see, assess, and analyze the environment around us. The majority of information in the world is obtained through the sense of sight. As humans, we have the ability to distinguish and recognize different patterns, including distinct facial features. With the advent of the computer age, scientists and tech companies have been striving to implement this visual capability in computer machines, leading to the birth of the field of computer vision. Thus it is rightly said by a famous English writer:</p><blockquote>“Where words are restrained the eyes often talk a great deal”</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*mISTL5FGQu8TFNTs" /><figcaption><strong>Human Vision</strong></figcaption></figure><h3>Problem Statement:</h3><blockquote>“Just like to hear is not same as to listen, to take pictures is not same as to see”.</blockquote><p>The mere act of capturing images and analyzing them picture by picture does not encompass the fundamental concept of computer vision. Computer vision aims to develop algorithms and techniques that enable computers to understand and interpret visual data in a manner similar to human vision.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BYrqpj8CVIsQhX0W" /><figcaption><strong>Pixel Analysis of Images</strong></figcaption></figure><h3>Early Developments:</h3><h4><strong>Viola-Jones Algorithm</strong>:</h4><p>The Viola-Jones algorithm, introduced in 2001, revolutionized face detection by combining Haar-like features and the AdaBoost learning algorithm. It provided a significant improvement in speed and accuracy for face detection tasks. There always has been a trade-off between accuracy and speed in terms of face detection. If we feed more features to be analyzed in terms of object recognition, the speed gets affected, and if we increase the speed by which the images are processed, the resolution gets affected.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*6pim1wxuNQSJ3zpRWr64eQ.png" /><figcaption><strong>Flowchart for Viola-Jones Algorithm</strong></figcaption></figure><h4>AdaBoost Learning Algorithm:</h4><p>This supervised algorithm basically combines the predictive power of different machine learning algorithms like decision trees, KNN, Logistic Regression, etc., on the same dataset to generate output with high accuracy. In the case of the decision trees algorithm, it first assigns equal weights to all data points and then measures the incorrect predictions, it then assigns heavy weights to those data points, and the weights of the correct prediction data points are decreased so that the next time we run the model, those particular data points are given more importance. Hence AdaBoost is a classifier just like gradient descent to improve the accuracy of the model.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*SoVqyRfS3eKUErlyYVR_3g.png" /><figcaption><strong>AdaBoost Learning Algorithm</strong></figcaption></figure><h3>Convolution and Deep Learning Algorithms:</h3><p>Convolutional Neural Networks (CNNs) emerged as a groundbreaking technique for computer vision tasks. CNNs leverage convolutional layers to extract features from input images and enable accurate object classification. Here deep neural networks are used to build a model in order to detect different types of objects. Each layer added to the neural network extract different features and then convolute with other feature to accurately predict the objects or faces.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*n0jApacjEZiSO_c8fWPkZQ.png" /><figcaption><strong>Feature extraction by different hidden layers of neural network</strong></figcaption></figure><h3>Pipeline for the Development of Computer Vision Algorithms:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/673/1*dhZPMNizPKalZ10pbz90NQ.png" /><figcaption><strong>Basic Face Recognition Algorithm</strong></figcaption></figure><h4>1. Data Acquisition:</h4><p>Collecting a diverse dataset of labeled images is crucial for training computer vision algorithms. This includes an unbiased collection of images from either already present datasets or generating new ones.</p><h4>2. Preprocessing:</h4><p>Applying image preprocessing techniques such as resizing, normalization, and noise reduction to enhance the quality of input images. This also includes augmenting the images in order to increase the size of the sample or training dataset.</p><h4>3. Feature Extraction:</h4><p>Extracting relevant features from images using techniques such as edge detection, color histograms, and texture analysis.</p><h4>4. Model Training:</h4><p>Utilizing machine learning algorithms like Viola-Jones or CNNs, to train the model on the labeled dataset.</p><h4>5. Model Evaluation:</h4><p>Assessing the performance of the trained model using metrics like accuracy, precision, and recall.Also, plotting the performance of the model.</p><h3>F.R.I.D.A.Y-Facial Recognition Image Detection and Analysis System</h3><p>This was the project I developed following the above steps I mentioned in this article as well as researching the topic. A brief summary of the same In the <strong>F.R.I.D.A.Y</strong> project, I created an application focused on facial recognition, image detection, and analysis. The main objective was to develop a reliable system capable of real-time face detection and analysis using computer vision techniques. The project involved building a face detection model and implementing an object classification pipeline. To train the face detection model, I collected images using a webcam and used an augmentation library to increase the dataset size by randomizing the images through adjustments in brightness, gamma value, and cropping. This resulted in a larger and more diverse dataset for training. Bounding boxes were used to annotate the faces in the images, providing ground truth information for the model. The face detection model consisted of a classification model to determine the presence of a face and a regression model to draw a bounding box around the face by estimating the coordinates. <strong>Binary Entropy Loss</strong> was used for the classification model, while<strong> Mean Squared Error (MSE)</strong> loss or localization loss was employed for the regression model. The neural network model was built using the Keras API, specifically utilizing the <strong>VGG16 model</strong> pre-trained on a vast image dataset. Additional layers were added to the VGG16 model for classification and regression. The trained model produced five values as output, including a probability value for classification and four coordinates for the bounding box. Overall, the F.R.I.D.A.Y project successfully developed a robust system for real-time facial recognition and image analysis, integrating various computer vision techniques, data augmentation, accurate annotation, and well-defined loss functions to achieve accurate face detection and localization. The code for the same is given below:</p><p><a href="https://github.com/RounakRaman/F.R.I.D.A.Y/blob/main/F.R.I.D.A.Y.py">F.R.I.D.A.Y/F.R.I.D.A.Y.py at main · RounakRaman/F.R.I.D.A.Y</a></p><h3>Recent Developments and Research:</h3><p>Highlight recent advancements in computer vision, such as a. State-of-the-art object detection algorithms like <strong>YOLO</strong> (You Only Look Once) and <strong>SSD</strong> (Single Shot MultiBox Detector).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*fCl0AKb_q1k0lwNEKIU3bA.png" /><figcaption><strong>FaceNet Architecture</strong></figcaption></figure><p>Advancements in facial recognition techniques, including the use of deep learning models like FaceNet and ArcFace.</p><p>Real-time video analysis and tracking using optical flow and motion detection techniques.</p><h3>Assumptions and Challenges Faced in Facial Recognition and Object Classification:</h3><p>Discuss the assumptions made and challenges encountered in the development and implementation of computer vision algorithms for facial recognition and object classification tasks. This may include issues related to lighting conditions, occlusion, pose variation, and the need for large labeled datasets.</p><p>In recent years the AI generative image model has been facing issues in generating an accurate image of hands. This is because, unlike of face, the hands include fingers and palms, which can be depicted in various different shapes and combinations, and there is still not enough data for the model to be trained on. This highlights the issue of how sufficient data should be present in order for the model to be accurate.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/310/1*c6YrU-eMWu11xL8JfaUxaQ.jpeg" /><figcaption><strong>Images of hands generated by AI</strong></figcaption></figure><p>This situation is well depicted and documented by <strong>Vox</strong> in one of its videos:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F24yjRbBah3w%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D24yjRbBah3w&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F24yjRbBah3w%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/06a60c3319e5106684343c232c6736dd/href">https://medium.com/media/06a60c3319e5106684343c232c6736dd/href</a></iframe><h3>Future Prospects and Applications:</h3><p>Explore the potential applications of computer vision in various fields, such as surveillance systems, autonomous vehicles, augmented reality, medical imaging, and robotics. Discuss future prospects, including the integration of computer vision with other emerging technologies like artificial intelligence and the Internet of Things. Its use is dependent on the user. E.g., on the one hand, we are seeing the rise of GPT-4, which is bringing a revolution in the access to information and its accuracy, and the recent launch of Vision Pro glasses by Apple Inc. On the other hand, we can see the use of Computer Vision being used to regulate and make a dystopian society in China with little to no privacy by using the Social Credit System. There is a detailed video you can see below:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F5ZhVGsh2KZ8%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D5ZhVGsh2KZ8&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F5ZhVGsh2KZ8%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/568c43cabdfc60173fae1408117de853/href">https://medium.com/media/568c43cabdfc60173fae1408117de853/href</a></iframe><h3>Source Citation and References Used:</h3><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F-4E2-0sxVUM%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D-4E2-0sxVUM&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F-4E2-0sxVUM%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/b9f3b46cc5fa33476812452ff9665b12/href">https://medium.com/media/b9f3b46cc5fa33476812452ff9665b12/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FN_W4EYtsa10%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DN_W4EYtsa10&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FN_W4EYtsa10%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c96292824209dadd8c4eba7bf31046d7/href">https://medium.com/media/c96292824209dadd8c4eba7bf31046d7/href</a></iframe><p><a href="https://iopscience.iop.org/article/10.1088/1742-6596/1755/1/012006/pdf">https://iopscience.iop.org/article/10.1088/1742-6596/1755/1/012006/pdf</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e27cc9af3d0f" width="1" height="1" alt=""><hr><p><a href="https://pub.towardsai.net/computer-vision-and-its-application-in-facial-recognition-and-object-classification-e27cc9af3d0f">Computer Vision and Its Application in Facial Recognition and Object Classification.</a> was originally published in <a href="https://pub.towardsai.net/">Towards AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Can The Economy Grow Forever?]]></title>
            <link>https://medium.com/illumination/can-the-economy-grow-forever-eb124f47270c?source=rss-7f88b6e6033c------2</link>
            <guid isPermaLink="false">https://medium.com/p/eb124f47270c</guid>
            <category><![CDATA[sustainable-development]]></category>
            <category><![CDATA[impact-investing]]></category>
            <category><![CDATA[circular-economy-2]]></category>
            <dc:creator><![CDATA[Raman Rounak]]></dc:creator>
            <pubDate>Sun, 05 Mar 2023 07:18:14 GMT</pubDate>
            <atom:updated>2023-06-14T04:45:04.162Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>A Historic Outlook</strong></p><p>Let’s take a trip back in time, shall we? Picture this: a bustling marketplace filled with traders from all over the world, haggling over goods and exchanging stories. This wasn’t just any marketplace — this was the Silk Road, the original global economy.</p><p>The Silk Road was a vast network of trade routes that connected China, India, Persia, Arabia, and Europe. Traders travelled great distances to bring exotic goods like silk, spices, and precious metals to eager customers in far-off lands. Along the way, they exchanged ideas, beliefs, and cultural practices, creating a vibrant tapestry of human experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jVUVs6LrHMr6S772K6cUaw.png" /><figcaption>Image by author</figcaption></figure><p><strong>The Impact of Global Economies</strong></p><p>Fast-forward a few centuries, and we see another transformation in the global economy. With the advent of modern transportation and communication technology, the world became more interconnected than ever before. Suddenly, goods could be shipped halfway around the world in a matter of days, and information could be transmitted instantaneously.</p><p>This gave rise to multinational corporations, which could operate in multiple countries and take advantage of the unique resources and markets available in each one. It also led to the growth of free trade agreements, which allowed for the free flow of goods and services across borders. The global economy had truly arrived.</p><p>But with great power comes great responsibility, as they say. The global economy has had a profound impact on the planet, both positive and negative. It has opened new opportunities for businesses and individuals alike, but it has also led to job losses in some regions and environmental degradation on a global scale.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fzfn0XHCfDHA%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dzfn0XHCfDHA&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fzfn0XHCfDHA%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/cc1a7e1624ce8b1a7d336e5d74838c57/href">https://medium.com/media/cc1a7e1624ce8b1a7d336e5d74838c57/href</a></iframe><p><strong>Growth with Sustainability?</strong></p><p>However, as the world economy grows exponentially, the question arises — can it grow forever? While some economists argue that growth is necessary for the economy to prosper, others believe that unlimited growth is unsustainable and may lead to negative consequences.</p><p>As our global economy increases exponentially, so does human intellect; hence, according to numerous economists, the growing economy will pave the way for the discovery of sustainable concepts that cannot be foreseen at this time. Consider Germany, whose economy increased by 16% while carbon emissions decreased by 12%. The global economy may expand by either creating more goods or inventing ideas using efficient technology. To achieve sustainable growth, the economy should be converted to Post-economy growth, in which humanity is not dependent on growth but rather on what is deemed necessary. This, in turn, allows people to have universal access to healthcare and education services while eliminating inequality and uniform wages.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/607/1*Ii1d29xGiqKgGVlRW3Y8lQ.jpeg" /><figcaption>Photo by <a href="https://www.shutterstock.com/image-illustration/eco-friendly-sustainable-growth-concept-3d-1078231196">petrmalinak on Shutterstock</a></figcaption></figure><p><strong>Revolutionizing Global Economy</strong></p><p>The good news is, there are innovative ideas and theories emerging that could change the way we think about global economies. Let’s explore some of these ideas.</p><p>Circular Economy — An Economic System that Prioritizes Sustainability</p><p>The circular economy is an economic system that aims to minimize waste and maximize the use of resources. Instead of the traditional linear economy, where materials are used, consumed, and discarded, a circular economy focuses on reusing, recycling, and repurposing materials as long as possible to extract the most value from them.</p><p>For example, a company could develop a product that can be disassembled and recycled, allowing it to recover valuable components while reducing the demand for virgin resources. This approach can not only reduce waste but also create new economic opportunities in the recycling and waste management industries.</p><p>Impact Investment — A New Approach to Investing</p><p>Impact investment is an investment strategy that seeks both financial returns and social or environmental impact. By investing in companies or initiatives that have a clear social or environmental goal, such as renewable energy or sustainable agriculture, investors can generate positive change while also earning profits.</p><p>The creation of green bonds is one example of impact investment in action. Green bonds are bonds issued to fund environmentally sustainable initiatives. This approach can not only benefit the environment but also create new business opportunities.</p><p>Circular Supply Chain — A Technique to Promote Material Reuse and Recycling</p><p>A circular supply chain is a supply chain management technique that promotes material reuse and recycling. This includes developing products that can be disassembled and recycled easily, as well as building connections with suppliers and customers who prioritize material reuse.</p><p>A circular supply chain can decrease waste, save resources, and open up new business opportunities. A corporation, for example, may develop a product that can be readily disassembled and recycled, allowing it to recover valuable components while reducing the demand for virgin resources. This strategy may also result in the creation of new employment in the recycling and waste management industries.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/626/1*SndN9vkhyAZV3WB7wFNt8A.jpeg" /><figcaption><a href="https://www.123rf.com/photo_186163293_futuristic-circular-economy-concept-with-glowing-low-polygonal-infinity-loop-windmill-recycle-sign.html?vti=undefined-1-54">https://www.123rf.com/photo_186163293_futuristic-circular-economy-concept-with-glowing-low-polygonal-infinity-loop-windmill-recycle-sign.html?vti=undefined-1-54</a></figcaption></figure><p><strong>Conclusion — A Sustainable and Fair Global Economy</strong></p><p>In conclusion, while the question of whether the economy can grow forever remains a topic of debate, there is always potential for development. By embracing innovative ideas like circular economies, impact investment, and circular supply chains, we can transform global economies and create a more sustainable and fair world. As we continue to evolve and adapt to changing times, let us welcome new ideas and strive towards a better future for all.</p><p>PS: If you reached this part I want to thank you for taking your valuable time and giving read to this article. This topic is connected to each and every one of us as we define the global economy. Every small decision of us creates a Butterfly Effect and the ripples are then felt in the economy. As it is rightly said that it is the perception that guides an economy and with growing time we have changed our outlook towards the economy making it more sustainable and bringing equality among people.</p><p>See you next time. Till then Happy Reading.❤❤</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=eb124f47270c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/illumination/can-the-economy-grow-forever-eb124f47270c">Can The Economy Grow Forever?</a> was originally published in <a href="https://medium.com/illumination">ILLUMINATION</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is Stoicism?]]></title>
            <link>https://raman-rounak.medium.com/what-is-stoicism-7ef53d1f10a5?source=rss-7f88b6e6033c------2</link>
            <guid isPermaLink="false">https://medium.com/p/7ef53d1f10a5</guid>
            <category><![CDATA[attack-on-titan]]></category>
            <category><![CDATA[stoicism]]></category>
            <category><![CDATA[philosophy]]></category>
            <dc:creator><![CDATA[Raman Rounak]]></dc:creator>
            <pubDate>Fri, 14 Oct 2022 21:35:54 GMT</pubDate>
            <atom:updated>2023-06-15T13:11:41.200Z</atom:updated>
            <content:encoded><![CDATA[<p>“<strong>The world is merciless, and it’s also very beautiful</strong>.” This is a quote from our all-time favorite anime ‘<strong>The Attack on Titan</strong>’. What does it tell you? Is it pointing toward our depressing, anxiety-filled reality?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*bW3JhhCbjT3MKHmx" /><figcaption>Photo by Hassan Sherif on Unsplash</figcaption></figure><p>The answer is <strong>NO</strong>. It’s much more than that, as we will learn ahead.</p><p>We all know how unjust, unfair and random our world is, but we always have two choices to deal with it.</p><p>We can either try to reduce this randomness or we can adopt the school of study called <strong>Stoicism</strong>.</p><p>Our world is huge with so many moving forces. Trying to resist these forces of nature ends up giving us anxiety and making us feel bad about our limited power. This is where<strong> </strong>Stoicism comes in. I recently read about Stoicism, which deals with values of greater acceptance. The radical practice of Stoicism makes you emotionally resilient to all the misfortune or mishaps taking place in your life.</p><p>While preparing for my JEE Mains exams, I couldn’t understand where I was lacking. I used to prepare hard for each attempt, I even scored good marks but I was unable to get a good rank. I started to blame the education system, the percentile system, and the Mains scandal, which made me anxious but there was nothing I could do about it. Many students indeed made it to my college by false means and I don’t hate them now.</p><p>I’ve come to accept the reality that our government couldn’t catch all those students. <br>After reading about Stoicism, I’ve realized that some things are not under our control. We should be able to separate the events that can be changed and those that can’t. <br>I also learned that without the understanding of Stoicism, we vent our emotions on the people that we love or on those who don’t have any relation to the situation.</p><p>Shouting at a flight attendant causes a flight to be delayed due to bad weather or thinking about the response you would get from your crush are just useless because neither is in our control.</p><p>The best thing we can do is to write down all our thoughts. As long as we have an outlet to vent out these emotions, we would feel much better.</p><p>This is what Stoicism teaches us, how a leaf moves through the storm which otherwise would get cracked on opposing the forces of nature. It teaches us to detach from the good and bad of nature and it is truly a power that takes away the control of nature from you. It teaches us how to learn from the situation and how to adapt to our space.</p><p>If you’ve made it here, thanks for reading.</p><p>PS: This is my first post so there might be some grammatical errors please ignore it for now.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7ef53d1f10a5" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>