<?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 Harshita Vyas on Medium]]></title>
        <description><![CDATA[Stories by Harshita Vyas on Medium]]></description>
        <link>https://medium.com/@harshita-vyas?source=rss-3c713b11537------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*bjWV4--dm1nyoImZeXD3gg.jpeg</url>
            <title>Stories by Harshita Vyas on Medium</title>
            <link>https://medium.com/@harshita-vyas?source=rss-3c713b11537------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 02:37:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@harshita-vyas/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[UI UX Case Study: QUENCHY- The water intake tracker mobile application]]></title>
            <link>https://harshita-vyas.medium.com/ui-ux-case-study-quenchy-the-water-intake-tracker-mobile-application-b7c33d9f50a?source=rss-3c713b11537------2</link>
            <guid isPermaLink="false">https://medium.com/p/b7c33d9f50a</guid>
            <category><![CDATA[ux-research]]></category>
            <category><![CDATA[prototyping]]></category>
            <category><![CDATA[drinking-water-apps]]></category>
            <category><![CDATA[ux-ui-design-thinking]]></category>
            <category><![CDATA[ui-ux-case-study]]></category>
            <dc:creator><![CDATA[Harshita Vyas]]></dc:creator>
            <pubDate>Sat, 10 Apr 2021 17:46:42 GMT</pubDate>
            <atom:updated>2021-04-12T13:58:41.265Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*2Kk-m158qOEANghS" /><figcaption>Photo by <a href="https://unsplash.com/@dtravisphd?utm_source=medium&amp;utm_medium=referral">David Travis</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p><strong><em>Role: </em></strong><em>Product Designer</em></p><p><strong><em>Overview: </em></strong><em>A mobile app that calculates and reminds users to drink their daily dose of water</em></p><p><strong><em>Number of Screens: </em></strong><em>27(Android)</em></p><p><strong><em>Audience:</em></strong><em> Tech Savvy</em></p><h3><strong>The Problem</strong></h3><blockquote><strong>Thousands Have Lived Without Love, Not One Without Water. — W. H. Auden</strong></blockquote><p>Water is the most astounding substance. Made up of just three simple atoms, two hydrogens fused to one oxygen, yet forming the basis of all life forms on Earth. Besides breathing, water is a critical requirement of our body. It calls for many important functions like regulating body temperature, flushing out waste from the body, and functioning of the brain.</p><p><em>-But regrettably, for most people, their daily water intake is very low as compared to what the body needs every day.</em></p><p>They forget to drink even half a liter of water unless they have exerted some physical effort, and I am a prime example of this. Some also find difficulties in tracking their daily amount of water intake.</p><h3><strong>The Solution</strong></h3><p>This is where ‘Quenchy’ steps in to make a difference. I decided to provide a mobile app that would help people achieve their required daily water intake and beat dehydration. It would help inculcate a good lifestyle with a healthy body and a proactive mind. The app has intuitive data visualizations that acknowledge both aesthetics and human motives.</p><h3>U<strong>ser and Audience</strong></h3><p>This app has been specially designed for the youth as they usually forget to drink water due to their busy schedule causing dehydration which results in headaches, dizziness, and sometimes even anxiety.</p><h3><strong>The Process</strong></h3><blockquote>STEP 1:<strong> Ideating</strong></blockquote><p>During my study, I discovered that the majority of people who use these water-drinking apps struggle to do so daily without being bored. They find the interface to be dull and monotonous. To overcome this, I decided to include the following in the application:</p><ul><li><em>Character: </em>to<em> </em>increase interaction as users will relate to it</li><li><em>Quick Tips:</em> to help people reach their water intake goal faster and achieve a good lifestyle</li><li><em>The Leaderboard:</em> to keep users motivated and disciplined</li></ul><p><strong>Character: </strong>After much sketching and brainstorming, I decided on a cactus plant as the main character and named it ‘Quenchy’. Many cactus plants wither due to dehydration as a result of the widely held misconception that they need little to no water to flourish.</p><p><em>Although these plants are drought-resistant, it is good to keep in mind that they can easily die if they are not provided with sufficient water.</em></p><p>Just like cactus’ need for water is neglected, we tend to forget our daily water requirement. Hence, users can relate themselves with Quenchy and consume their daily water requirements. Every time the user drinks water, Quenchy’s expressions transition from being sad and thirsty to being elated. Quenchy feels ecstatic when you complete your daily requirement.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fbV8_ZSXY6Plii8UeZgaFQ.png" /></figure><p><strong>Quick Tips: </strong>This feature assists users by providing them with suggestions for incorporating drinking water into their busy schedules, which will help them maintain a healthier lifestyle.</p><p><strong>The Leaderboard: </strong>A healthy competition can help people challenge themselves to push their limits and achieve milestones they could never think of. Adding friends and comparing each other’s progress would not only increase social engagement but also encourage the users to work hard and achieve their daily intake target. Users can also compare and monitor their progress through a small, fun map game where if they complete their target, they move one step ahead otherwise, three steps back. Additionally, daily personal progress can also be tracked.</p><blockquote>STEP 2:<strong> Wireframing</strong></blockquote><p>It aids in determining a project’s goals and primary emphasis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wd1zFXu_71r0VCiBV4ERMA.png" /><figcaption>Low fidelity wireframes</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IhWx98NgDCh19KUxnlpOhg.png" /><figcaption>High fidelity wireframes:</figcaption></figure><blockquote>STEP 3:<strong> Prototyping</strong></blockquote><p><a href="https://www.figma.com/proto/4I4m6dVq3N1Vv3CINXe5dL/Buddy?node-id=70%3A120&amp;viewport=305%2C382%2C0.0625&amp;scaling=scale-down&amp;page-id=0%3A1">Prototype link</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IdYWvJZS_arbwjtc91uCIw.png" /></figure><blockquote>STEP 4:<strong> Usability Testing:</strong></blockquote><p>To evaluate this app, I decided to conduct an online test with 5 participants. I shared with them a questionnaire along with the prototype link.<em> </em>This allowed me to gain a deeper understanding through combining both qualitative and quantitative information.</p><p><strong>Participants were asked the following questions:</strong></p><p>Q1) Imagine you just drank water, were you able to add this to the app?</p><p>Q2) Were you able to find how much water you already drank today and how much you still need to drink?</p><p>Q3) Do you think that comparing your daily water intake with your friends would help you in accomplishing your daily target?</p><p>Q4) Imagine you forget to drink water, so do you want this app to remind you through alert messages?</p><p><em>For all the above questions, I got an affirmative response from all of the participants.</em></p><p><strong>Below is a bar graph for some rating-related questions.</strong></p><p>Q1) How helpful was the ‘Quick Tips’ feature?</p><p>Q2) Rate your experience in navigating from one page to another.</p><p>Q3) How satisfied are you with the overall design of the app?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_p_npjXIL1VkassMHhTHDg.png" /></figure><h3><strong>Outcomes and Results</strong></h3><p>Although users enjoyed the graphics and visuals, they still struggled to keep with drinking water after regular intervals. So, after some discussions with the users, I came up with some developments in the app.</p><ol><li><strong>Reminders:</strong> A notification would pop up on the screen of the user from time to time. This will remind them to drink water.</li><li><strong>Widgets:</strong> Widgets can be added to your phone’s home screen to provide a simple way to access details from apps without having to open them. This would help users to quickly add water from their home screen itself and they would be able to see their water intake every time they unlock their phone.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*O1PjnXk9LqRSoE5_D5tEZQ.png" /></figure><h3><strong>Conclusion</strong></h3><p>After the addition of reminders and widgets, the participants were satisfied and confident that they would now be able to keep up with their daily water intake. The work on this project was up and down, but I am confident of my approach, and the next one will be even better. This was my first experience working on a full-fledged UX project and I have just begun exploring this area. As a result, I would be delighted to hear from you.</p><h3>References and Contact Information</h3><ol><li>Information about cactus plants: <a href="https://cactusway.com/how-do-you-know-when-a-cactus-needs-water/#:~:text=Final%20Thoughts-,The%20notion%20that%20cacti%20plants%20require%20little%20to%20no%20water,provide%20them%20with%20sufficient%20water">link</a></li><li>Prototype: <a href="https://www.figma.com/proto/4I4m6dVq3N1Vv3CINXe5dL/Buddy?node-id=70%3A120&amp;viewport=305%2C382%2C0.0625&amp;scaling=scale-down&amp;page-id=0%3A1">link</a></li><li>Email id: vyasharshita0@gmail.com</li><li>Linkedin: <a href="http://www.linkedin.com/in/harshita-vyas-819745181">www.linkedin.com/in/harshita-vyas-819745181</a></li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b7c33d9f50a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Fish weight prediction using Linear Regression and PyTorch]]></title>
            <link>https://harshita-vyas.medium.com/fish-weight-prediction-using-linear-regression-and-pytorch-6547d8ec7906?source=rss-3c713b11537------2</link>
            <guid isPermaLink="false">https://medium.com/p/6547d8ec7906</guid>
            <category><![CDATA[pytorch]]></category>
            <category><![CDATA[deeplearing]]></category>
            <category><![CDATA[jovian-ml]]></category>
            <category><![CDATA[kaggle]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Harshita Vyas]]></dc:creator>
            <pubDate>Fri, 03 Jul 2020 15:09:13 GMT</pubDate>
            <atom:updated>2020-07-03T15:15:35.524Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aPt6gO80mkC0JHU5" /><figcaption>Photo by <a href="https://unsplash.com/@davidclode?utm_source=medium&amp;utm_medium=referral">David Clode</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>So, I recently started working with PyTorch, and the best way to learn something is by implementing it.</p><p>After exploring a bit, I tried linear regression using PyTorch using the fish market <a href="https://www.kaggle.com/aungpyaeap/fish-market">data set</a> from Kaggle (because it is comparatively easy and has less number of columns).</p><h3><strong><em>Objective:</em></strong></h3><p><em>Find the weight of the fish, given its measurements.</em></p><h3><strong>Importing libraries:</strong></h3><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fjovian.ml%2Fembed%3Furl%3Dhttps%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D3&amp;dntp=1&amp;display_name=Jovian&amp;url=https%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D3&amp;image=https%3A%2F%2Fstorage.googleapis.com%2Fjvn%2Fassets%2Fvyasharshita0%2Fprofile_images%2F5d94b86e8b4b4d6cacc5ab9cdda160bb%2Fmedium.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;scroll=auto&amp;schema=jovian" width="800" height="200" frameborder="0" scrolling="no"><a href="https://medium.com/media/86d297445e0a8d55e704b0579bddf875/href">https://medium.com/media/86d297445e0a8d55e704b0579bddf875/href</a></iframe><p>After importing all important libraries, I started the <strong>EDA(exploratory data analysis)</strong>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fjovian.ml%2Fembed%3Furl%3Dhttps%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D21&amp;dntp=1&amp;display_name=Jovian&amp;url=https%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D21&amp;image=https%3A%2F%2Fstorage.googleapis.com%2Fjvn%2Fassets%2Fvyasharshita0%2Fprofile_images%2F5d94b86e8b4b4d6cacc5ab9cdda160bb%2Fmedium.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;scroll=auto&amp;schema=jovian" width="800" height="200" frameborder="0" scrolling="no"><a href="https://medium.com/media/540d595d5d0b7b7219b5bcf2e21b3a79/href">https://medium.com/media/540d595d5d0b7b7219b5bcf2e21b3a79/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fjovian.ml%2Fembed%3Furl%3Dhttps%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D23&amp;dntp=1&amp;display_name=Jovian&amp;url=https%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D23&amp;image=https%3A%2F%2Fstorage.googleapis.com%2Fjvn%2Fassets%2Fvyasharshita0%2Fprofile_images%2F5d94b86e8b4b4d6cacc5ab9cdda160bb%2Fmedium.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;scroll=auto&amp;schema=jovian" width="800" height="200" frameborder="0" scrolling="no"><a href="https://medium.com/media/a3290d21bf2452ef74aea6c70b9bb14c/href">https://medium.com/media/a3290d21bf2452ef74aea6c70b9bb14c/href</a></iframe><p>This <a href="https://pytorch.org/docs/master/generated/torch.nn.Linear.html">Linear</a> PyTorch model applies a linear transformation to input data and uses the <a href="https://pytorch.org/docs/master/generated/torch.nn.L1Loss.html">l1_loss </a>loss function. Its a commonly used loss function for linear regression. It is the numerical distance between the estimated and actual value. To learn more about loss functions in PyTorch, please click <a href="https://medium.com/udacity-pytorch-challengers/a-brief-overview-of-loss-functions-in-pytorch-c0ddb78068f7">here</a>.</p><p>The following is my code snippet.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fjovian.ml%2Fembed%3Furl%3Dhttps%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D47&amp;dntp=1&amp;display_name=Jovian&amp;url=https%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D47&amp;image=https%3A%2F%2Fstorage.googleapis.com%2Fjvn%2Fassets%2Fvyasharshita0%2Fprofile_images%2F5d94b86e8b4b4d6cacc5ab9cdda160bb%2Fmedium.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;scroll=auto&amp;schema=jovian" width="800" height="200" frameborder="0" scrolling="no"><a href="https://medium.com/media/8f08a2e2c68379aa0a26a17d91135d3b/href">https://medium.com/media/8f08a2e2c68379aa0a26a17d91135d3b/href</a></iframe><h3>Training the model:</h3><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fjovian.ml%2Fembed%3Furl%3Dhttps%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D52&amp;dntp=1&amp;display_name=Jovian&amp;url=https%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D52&amp;image=https%3A%2F%2Fstorage.googleapis.com%2Fjvn%2Fassets%2Fvyasharshita0%2Fprofile_images%2F5d94b86e8b4b4d6cacc5ab9cdda160bb%2Fmedium.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;scroll=auto&amp;schema=jovian" width="800" height="200" frameborder="0" scrolling="no"><a href="https://medium.com/media/0595de83eb74d361f5413389f7184182/href">https://medium.com/media/0595de83eb74d361f5413389f7184182/href</a></iframe><p>After evaluating this is the <strong>r2 score</strong>, which I obtained.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fjovian.ml%2Fembed%3Furl%3Dhttps%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D72&amp;dntp=1&amp;display_name=Jovian&amp;url=https%3A%2F%2Fjovian.ml%2Fvyasharshita0%2Ffish-market%2Fv%2F297%26cellId%3D72&amp;image=https%3A%2F%2Fstorage.googleapis.com%2Fjvn%2Fassets%2Fvyasharshita0%2Fprofile_images%2F5d94b86e8b4b4d6cacc5ab9cdda160bb%2Fmedium.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;scroll=auto&amp;schema=jovian" width="800" height="200" frameborder="0" scrolling="no"><a href="https://medium.com/media/3a7bae2ed706d7f97437c28b2a51f35e/href">https://medium.com/media/3a7bae2ed706d7f97437c28b2a51f35e/href</a></iframe><p>After a lot of fluctuations and trials and errors, I got this r2 score.</p><h3>Conclusion:</h3><p>1) Fluctuations reason (according to me ): Neural nets are not able to get proper inferences and trends from small datasets(although I used single architecture neural net).</p><p>2) For this dataset, I think that simple linear regression would be better instead of using the PyTorch linear functions.</p><p>3) The main focus here should be on PyTorch functions for the practice point of view rather than just seeing the accuracy.</p><h3>References:</h3><ol><li>My <a href="https://jovian.ml/vyasharshita0/fish-market">notebook</a> for this data set</li><li>For more <a href="https://jovian.ml/aakashns">details</a></li><li><a href="https://www.kaggle.com/aungpyaeap/fish-market">Dataset</a></li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6547d8ec7906" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>