<?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 Bharath Ram K on Medium]]></title>
        <description><![CDATA[Stories by Bharath Ram K on Medium]]></description>
        <link>https://medium.com/@bharath.ram.modder?source=rss-54289e19a064------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*uz16WIE8lv6TpxiN3lLzcA.jpeg</url>
            <title>Stories by Bharath Ram K on Medium</title>
            <link>https://medium.com/@bharath.ram.modder?source=rss-54289e19a064------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 05:15:36 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@bharath.ram.modder/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[PyCon 2023: A Journey into Tech Innovation]]></title>
            <link>https://medium.com/@bharath.ram.modder/pycon-2023-a-journey-into-tech-innovation-c5f554577ed2?source=rss-54289e19a064------2</link>
            <guid isPermaLink="false">https://medium.com/p/c5f554577ed2</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[pycon]]></category>
            <dc:creator><![CDATA[Bharath Ram K]]></dc:creator>
            <pubDate>Sat, 07 Oct 2023 06:31:29 GMT</pubDate>
            <atom:updated>2023-10-07T06:31:59.102Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/932/0*sA0paHJvNnA9K72d" /></figure><p>As I stepped into PyCon 2023, I couldn’t help but feel a rush of excitement. This marked my first foray into the world of tech-related events, and I was eager to explore new ideas, meet like-minded individuals, and soak up the latest innovations from tech companies. Let me take you on a journey through my experience at this remarkable event.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*IPRM1ri0108UQ4-CDEjQCg.png" /><figcaption>I was really this much excited for the event</figcaption></figure><p><strong><em>Day 1: Pre-Event Workshops (29th Sept 2023)</em></strong></p><p>My colleague and I arrived in Hyderabad a day before the main event, on the 29th of September. The venue, JNTU College of Technology, hosted a series of workshops covering <strong><em>Natural Language Processing (NLP), Deep Learning, and various other topics, including 3D modelling-related</em></strong> discussions.</p><p>The event kicked off with a buzz, as attendees poured in, eager to dive into the world of Python and its applications. We received our ID badges, welcome kits, and the coveted food coupons that would keep us fuelled throughout the event.</p><p>Our first session was a deep dive into “<strong><em>Fine-tuning LLM models</em></strong>” by <a href="https://github.com/abhijeet3922">Abhijeeth Kumar</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zTkAXdkJZXKF6gy-QOtBYQ.png" /></figure><p>During this session, we gained insights into the state of fine-tuning, different techniques for inferencing, memory requirements, and the fascinating world of <strong><em>Prompt Tuning (PEFT) and LORA</em></strong>. Code and examples from the session were generously shared on GitHub, enhancing the learning experience.</p><p>After the session, My colleague and I decided to unwind by exploring a nearby mall, indulging in some retail therapy, enjoying a meal at the food court, and catching up on life. By the time we left the mall, the day had come to a satisfying close.</p><p><strong><em>Day 2: Main Event Commences (30th Sept)</em></strong></p><p>The second day marked the official commencement of PyCon’s main events. The familiar rituals of collecting ID badges, goodies, and food coupons set the tone for an exciting day ahead.</p><p>One session I had been eagerly awaiting was the keynote address by <a href="https://www.linkedin.com/in/jessica0greene/">Jessica Greene</a>. Her talk emphasised the importance of community contributions and active participation in community events — a theme that resonated deeply with the tech enthusiasts in attendance.</p><p>Next on the agenda was <a href="https://github.com/snknitin/exploring-pandas-and-numpy">Nitin Kishore Sai Samala</a>’s session titled “<strong><em>Revealing Uncommon Tricks: 25 Obscure Pandas &amp; NumPy Hacks Learned Over 5 Years of Being a Data Scientist</em></strong>”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*7CVWZi-lI_6vbfKXn7BWNw.png" /></figure><p>This enlightening session offered invaluable insights into optimizing data manipulation and analysis workflows, enhancing data visualisation, and efficiently handling large datasets like</p><ul><li>Merging data frame with fuzzy match</li><li>Efficient groupby aggregations</li><li>Smart memory management by chunck processing</li><li>Inplace application with pandas</li><li>Automate EDA</li></ul><p>Later in the day, we attended a session on “<strong><em>Data Pipelines in Production — Bad vs. Best Practices</em></strong>” by <a href="https://www.linkedin.com/in/bhavanicodes/">Bhavani Ravi</a>. Bhavani shared essential insights into pipeline design, workflow considerations, and the importance of addressing bottlenecks within data pipelines.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*ZKkN2eoUC-MWU-jzwfOq8A.png" /></figure><p><em>Some key point made by Bhavani Ravi</em></p><ol><li>POC != Production</li><li>Design your Pipeline</li><li>What is the workflow of the project?</li><li>what are the repeating tasks and unique tasks in each pipeline?</li><li>what is the input?</li><li>what are the possible outputs?</li><li>Think about the scalability like what if the input is 10x / 100x / 1000x greater than the expected</li><li>List out and Handle bottlenecks</li><li>Components and system involved</li><li>Limitations</li><li>Know the tools for each topic in the eco system like<em> Data ingestion, Data streaming, Data analytics, ML training, Inference.</em></li><li>Dev responsibility on each project and work to be taken care on each of the <em>Pipeline design, Workflow design, Debugging</em></li></ol><p>This knowledge was complemented by engaging activities organized by sponsor companies, where we won some delightful goodies and interacted with data scientists and ML engineers.<br>One particular interaction that stood out was with Corteva Agriscience. Their innovative solutions for crop management, based on real-time tracking and data analysis involving time series, vision, and bioinformatics, left a lasting impression.</p><p>The day concluded with a sumptuous lunch provided at the event venue.</p><p><strong><em>Day 3: Final Day (1st Oct 2023)</em></strong></p><p>The third and final day of PyCon 2023 began with the keynote speech by <a href="https://www.linkedin.com/in/cheukting-ho/">Cheuk Ting Hu</a>, highlighting the benefits of contributing to open-source projects — a message that resonated with many in the tech community.</p><p>Following the keynote, we attended a session titled “Embracing Observability in Modern Python Applications: Harnessing the Power of OpenTelemetry” by Angelien John. The session covered the limitations of traditional monitoring, the concept of observability, and the power of OpenTelemetry in simplifying tracing and building observable systems.</p><p>With that, our PyCon journey came to a close.</p><p><strong><em>Conclusion</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/620/1*9NImonjSjsuoGJE1RNKvFA.jpeg" /></figure><p>Reflecting on the overall experience, I must say that PyCon 2023 was a valuable and informative event. It provided a platform to meet new people, gain fresh insights, and explore the latest trends in the tech industry. While I initially expected a stronger focus on Machine Learning (ML) and Deep Learning (DL) concepts, with more hands-on experiences, I found that the event catered effectively to newcomers to the field.<br>In the end, PyCon 2023 was a celebration of technology, innovation, and the vibrant Python community. I left with a sense of fulfilment and a deep appreciation for the ever-evolving world of tech.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c5f554577ed2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Describing data numerically]]></title>
            <link>https://medium.com/@bharath.ram.modder/describing-data-numerically-13517da3fb2b?source=rss-54289e19a064------2</link>
            <guid isPermaLink="false">https://medium.com/p/13517da3fb2b</guid>
            <category><![CDATA[probability]]></category>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Bharath Ram K]]></dc:creator>
            <pubDate>Sun, 19 Mar 2023 17:26:39 GMT</pubDate>
            <atom:updated>2023-03-19T17:26:39.924Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*-zReAeJoJ_WSzx2P.png" /></figure><p><em>Hello all,<br>This blog is for all the data science aspirants who want to have a better understanding of statistics or for those who got lost in the field stats and probs on where to start and what to study next in stats and probs while starting to studying for machine learning, then this blog is for you guys. I am going to write a series of blogs on Statistics and Probability.<br>Feel free to follow my page for more information on Stats and probs.</em></p><p>If you haven’t read my previous blog on <strong><em>Describing data using graphs and tables -II </em></strong>click<strong><em> </em></strong><a href="https://medium.com/@bharath.ram.modder/describing-data-using-graphs-and-tables-ii-816fcd28512c"><strong><em>here</em></strong></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/256/0*S0P8SrjstVnXadt2.jpg" /><figcaption>Measure of central tendency</figcaption></figure><blockquote><strong>Lesson 3.2 Describing data numerically\</strong></blockquote><blockquote><strong>3.2.1 Measure of Center</strong></blockquote><blockquote><strong>Objective</strong><br>1. Calculate the mean for the given data set.<br>2. Find the median and describe why the median s sometimes preferable to the mean.<br>3. Find the mode of a data set.<br>4. Describe how skewness and symmetry affect these measures of center.</blockquote><h4>Mean</h4><p>To find the mean of the values in a data set, simply add up all numbers in the population and divide by total numbers of population.</p><p><strong><em>Mean for Population N</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/227/1*Vkpze7AV5bf1_uglh3i55Q.png" /><figcaption>where N is the population size</figcaption></figure><p><strong><em>Mean for Sample n</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/90/1*jwBajmzi0hEvdwYzOzScwQ.png" /><figcaption>Where n is the sample size</figcaption></figure><blockquote><strong>Note : Mean is sensitive to extreme values.</strong></blockquote><p>Let us see an example</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/705/1*BwXwEn3K5umpwOh6HpjkqQ.png" /><figcaption>Try it yourself using the above discussed formulae</figcaption></figure><p><strong><em>Solution</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/687/1*f_V0OTzOHVc8zBFhIqQ9HA.png" /></figure><h4>Median</h4><p>The Median represents the middle of a data set. when data are put in ascending order half lie below median and half lie above.</p><ul><li>If the data size is odd the Median = (N + 1)/2 th position value</li><li>If the data size is even then Median = Either side of the (N + 1)/2 th position.</li></ul><h4>Mode</h4><p>The Mode of the data set is the data value that occurs with more frequency.</p><p>There are three chances of outcome while using mode</p><ul><li>Node mode : If all the values are unique</li><li>Multiple mode : If more the one value has the same frequency.</li><li>Uni mode : If there is only one mode.</li></ul><h4>Skewness and measure of center</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/708/1*IQQb8-MgcO1_P_grTdFP3w.png" /><figcaption>a. Left skewed b. Symmetric c. Right skewed</figcaption></figure><p><em>How skewness affect the Mean and Median</em></p><ul><li>For Left skewed distribution, the median is larger than the mean.</li><li>For Right skewed distribution, the mean is larger than the mean.</li><li>For Symmetric unimodal distribution, the mean, the median, the mode are fairly close to one another.</li></ul><blockquote><strong>3.2.2 Measure of variability</strong></blockquote><blockquote><strong>Objective</strong><br>1.Understand and calculate the range of data set.<br>2.Calculate the variance and the standaed deviation for the population.<br>3.Compare the variance and standard deviation for the sample.<br>4.Use the Empirical Rule to find the approximate percentages for a bell shaped distribution.<br>5.Apply Chebyshev’s Rule to find minimum percentages.</blockquote><h4>The Range</h4><ul><li>The range of data set is the difference between the largest value and smallest value in the data set.</li><li><strong><em>Range = largest — smallest or maximum — minimum</em></strong></li><li>The range is quite sensitive to extreme values.</li><li>We would prefer our measure of variability to quantify spread with respect to actually use all of the available data values.</li></ul><h4>Variance and standard deviation of population.</h4><p><strong>Deviation <br></strong>A deviation for a given data value is the difference between the data value and the mean of the data set.</p><p><strong>Population variance <br>The population variance </strong>σ² is the mean of the squared deviation in the population.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/232/1*XmXWYFqxMBuS81c_TAenUA.png" /></figure><p><strong>The Population Standard deviation σ</strong></p><ul><li>The standard deviation is simply the square root of the variance.</li><li>By taking square root we return the units of measure back to its original data unit.</li><li>The population standard deviation σ is the positive square root of the population variance.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/249/1*yAaZz_zGIsXDc3AwKR-ELA.png" /></figure><p><strong>Sample variance</strong></p><p>The sample variance s² is approximately the mean of th squared deviation in the sample and is found by</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/208/1*bGpa4CUOGUWJEa7RrW0H8Q.png" /></figure><p><strong>Sample Standard deviation</strong></p><ul><li>Most commonly used measure of spread</li><li>The sample standard deviation s is the positive square root of the sample variance s².</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/248/1*uQ8pfcaYV_i9vPza3nnQKQ.png" /></figure><h4>Empirical rule</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/698/0*dbocfAMUFO0Y2wbC.jpg" /><figcaption>Bell shaped distribution of data</figcaption></figure><p>If the data is bell shaped</p><ul><li>About 68% of the data values will fall with 1 standard deviation of the mean.</li><li>About 95% of the data values will fall within 2 standard deviation of the mean.</li><li>About 99.7% of the data values will falls within 3 standard ediation of the mean.</li></ul><h4>Chebyshev’s rule</h4><p>The proportion of values from a data that will fall with in k standard deviation of the mean will be at last</p><p><strong><em>(1- (1/k²)) 100%</em></strong></p><ul><li>k &gt; 1 chebyshev ’s rule may be applied to either samples or population. For example:</li><li>When k =2 at least 3/4 data or 75% data will fall with in 2 standard deviation.</li><li>When k =3 at least 8/9 data or 88% data will fall with in 3 standard deviation of mean</li></ul><p>That’s all about this blog. Hope you guys got an idea about what are the measure of center and measure of variability for numerical data and . Let’s discuss more about <strong><em>Grouped data, Relative position and outliers </em></strong>in the next blog.</p><p><strong><em>Feel free to Comment, Like and Share this blog .</em></strong></p><p><strong>Thank you !!<br>Happy learning :)</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=13517da3fb2b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Describing data using graphs and tables — II]]></title>
            <link>https://medium.com/@bharath.ram.modder/describing-data-using-graphs-and-tables-ii-816fcd28512c?source=rss-54289e19a064------2</link>
            <guid isPermaLink="false">https://medium.com/p/816fcd28512c</guid>
            <category><![CDATA[probability]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Bharath Ram K]]></dc:creator>
            <pubDate>Sun, 26 Feb 2023 07:06:31 GMT</pubDate>
            <atom:updated>2023-02-26T07:06:31.620Z</atom:updated>
            <content:encoded><![CDATA[<h3>Describing data using graphs and tables — II</h3><p><em>Hello all,<br>This blog is for all the data science aspirants who want to have a better understanding of statistics or for those who got lost in the field stats and probs on where to start and what to study next in stats and probs while starting to studying for machine learning, then this blog is for you guys. I am going to write a series of blogs on Statistics and Probability.<br>Feel free to follow my page for more information on Stats and probs.</em></p><p>This is the 4th blog on the statistics series for data science. If you haven’t read my previous blog click <a href="https://medium.com/@bharath.ram.modder/describing-data-using-graphs-and-tables-161b39e6358d"><strong>here</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3LtTgXEH9vIOUnPy.png" /><figcaption>Image from <a href="https://www.powerslide.io/en/blog/data-visualisation-what-you-need-to-know">https://www.powerslide.io/en/blog/data-visualisation-what-you-need-to-know</a></figcaption></figure><blockquote><strong>Lesson 3.1: Describing Categorical data using graphs and tables</strong></blockquote><blockquote><strong>Objective</strong><br>1) Construction and interpret a frequency distribution and relative frequency distribution for discrete and continuos.<br>2) Use histogram and frequency polyons to summarize quantitaive data.<br>3)Construct and interpret stem and leaf display and dot plot.<br>4) Recognize distribution shape, system and skewness.</blockquote><h3>Frequency Distribution and Relative Frequency distribution for discrete data</h3><p>It is exactly as discussed for categorical data columns but instead here we have qualitative variables.</p><p><strong><em>Let us consider a example</em></strong></p><p><em>The National Center for Missing and Exploited Children keeps an online searchable data base of missing children nationwide. Table below contains a listing of the 50 children who have gone missing from California and who would have been between 1 and 9 years of age as of March 4, 2007. Suppose we are interested in analyzing the ages of these missing children. Use the data to construct afrequency distribution and a relative frequency distribution of the variable age.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/839/1*UvWhiqJMGOqrqoiVJAZouw.png" /><figcaption>Image taken from the Larose and Daniel, Discovering the fundamentals of statistics</figcaption></figure><p>The solutio to the above question is</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/782/1*Rjf9EUC3GZJDnrC0K03_1w.png" /><figcaption>Image taken from the Larose and Daniel, Discovering the fundamentals of statistics</figcaption></figure><p>We can use the following terms to construct a frequency distribution and histogram</p><ul><li><strong><em>Lower class limit</em> </strong>of a class equals the smallest value within that class.</li><li><strong><em>Upper class limit</em></strong> of a class equals the largest value within that class.</li><li><strong><em>Class width </em></strong>equals the difference between the lower limit of 2 successive class.</li><li><strong><em>The Class boundary</em></strong> of the 2 successive classes found by taking the sum of upper class limit of a class and the lower limit of the class to its right and dividing the sum by two</li></ul><h3>Histogram and Frequency polygons</h3><ul><li>A Histogram is constructed using rectangles for each class of data.</li><li>The height of rectangles represent the frequency or relative frequency of the class.</li><li>Width represent the class width of the corresponding class</li></ul><blockquote><strong>Note:</strong> The rectangles in Histogram is a contiguous (touching) where as in Bar graph it does not.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/769/1*zv6yZ4OVv1sPCeqFik5mKQ.png" /></figure><h4>Shifting the histogram to the left</h4><p>If we subtract 10 points on all the elements then the class will reduce by 10. Each and every element will e reduced by 10 then the histogram will be shift by 10.t</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/504/1*DZ_Ox6AFo-FvBzPU9K_PGw.png" /><figcaption>The same histogram used above( a) is shifted by 10 to left</figcaption></figure><h4>Frequency Polygon</h4><p>A Frequency polygon is constructed as follows,</p><ul><li>For each class a point at the class midpoint, at a height equal to the frequency for that class.</li><li>Then each consecutive pair of points with a line segment</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/588/1*THoJpuM3hff3qLP3MRsXJg.png" /></figure><h3>Stem and Leaf display and dot plots</h3><h4>Step and Leaf display</h4><p>Stem and Leaf display contains more information than the histogram and frequency distribution.</p><p><em>Let us see an example</em></p><p><em>Construct a stem-and-leaf display for the exam scores of 20 statistics students, given below: 57 60 61 65 69 73 74 75 75 75 76 77 78 81 82 82 85 91 95 98</em></p><p><strong>Step 1 </strong>: Take the leading digit of the tens position number in this case 5,6,7,8,9</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/255/1*e5luO9CukDIVZyh2n-ykaw.png" /><figcaption>Taking the 10,s position value from the given input data</figcaption></figure><p><strong>Step 2 : </strong>Add the one’s number one by one</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/228/1*AEuSfENBumzSTVMKWWKn3g.png" /><figcaption>Adding the one’s position number to the tree</figcaption></figure><h4>Dot plot</h4><ul><li>A simple but effective graphical display is a dot plot</li><li>In a dot plot each data point is represented by a dot above the number line.</li><li>When the sample is large each dot any represent more than one data point.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PL8brQl5ul779vdrtv4sig.png" /></figure><h4>Distribution Shape, Symmetry</h4><p>The distribution of a variable is a table graph or formula that identifies the variable values and frequency for all element in the dataset.</p><p><strong>Analysing the shape of distribution</strong></p><p>A image has symmetry if there is a line that splits the image i half so that one side is the mirror image of the other. This shape of distribution is called bell shape curve.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/643/1*_f-Yzmz-w0ECVjkYlPCwUA.png" /><figcaption>We can split the image exactly into two half</figcaption></figure><blockquote><strong>Eight Common Methods for Making a Graph Misleading</strong></blockquote><blockquote>1. Graphing /selecting an inappropriate statistic.</blockquote><blockquote>2. Omitting the zero on the relevant scale.</blockquote><blockquote>3. Manipulating the scale.</blockquote><blockquote>4. Using two dimensions (area) to emphasize a one-dimensional difference.</blockquote><blockquote>5. Careless combination of categories in a bar graph.</blockquote><blockquote>6. Inaccuracy in relative lengths of bars in a bar graph.</blockquote><blockquote>7. Biased distortion or embellishment.</blockquote><blockquote>8. Unclear labeling.</blockquote><p>That’s all about this blog. Hope you guys got an idea about what are te graphs and tables for categorical data in stats. Let’s discuss more about <strong><em>Describing data numerically</em></strong> in the next blog.</p><p><strong><em>Feel free to Comment, Like and Share this blog .</em></strong></p><p><strong>Thank you !!<br>Happy learning :)</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=816fcd28512c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Describing data using graphs and tables — I]]></title>
            <link>https://medium.com/@bharath.ram.modder/describing-data-using-graphs-and-tables-161b39e6358d?source=rss-54289e19a064------2</link>
            <guid isPermaLink="false">https://medium.com/p/161b39e6358d</guid>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[probability]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Bharath Ram K]]></dc:creator>
            <pubDate>Sun, 19 Feb 2023 11:16:42 GMT</pubDate>
            <atom:updated>2026-06-30T10:56:53.363Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Hello all,<br>This blog is for all the data science aspirants who want to have a better understanding of statistics or for those who got lost in the field stats and probs on where to start and what to study next in stats and probs while starting to studying for machine learning, then this blog is for you guys. I am going to write a series of blogs on Statistics and Probability.<br>Feel free to follow my page for more information on Stats and probs.</em></p><p>This is the 3rd blog on the statistics series for data science. If you haven’t read my previous blog click <a href="https://medium.com/@bharath.ram.modder/statistics-data-classification-e4f744473dfa">here</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/698/0*mrInU6aPAcakocfO.jpg" /><figcaption>Image taken from <a href="https://chartexpo.com/blog/types-of-charts-and-graphs">https://chartexpo.com/blog/types-of-charts-and-graphs</a></figcaption></figure><blockquote><strong>Lesson 3.1: Describing Categorical data using graphs and tables</strong></blockquote><blockquote><strong>Objective</strong><br>1) Construct and interpret a frequency distribution and a relative frequency distribution for qualitative data.<br>2)Construct and interpret bar graph and pareto charts.<br>3)Construct and interpret pie chart.<br>4)Construct Crosstabulation to describe the relationship between 2 variables.<br>5)Construct a clustered bar graph to describe the relationship between 2 variables.</blockquote><h3><strong>Frequency Distribution and Relative Frequency Distribution</strong></h3><p><strong>Relative Frequency</strong></p><ul><li>Frequency distribution is a method used to summarize values in a dataset.</li><li>The frequency or count of a categorial value refers to the number of observation in each category.</li><li>Frequency distribution for a qualitative variable is a listing of all the values.</li></ul><p>Example for frequency distribution</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Fxblmo9Xbd9Quz9L3FXNog.png" /><figcaption>List of games created by each console in a calendar</figcaption></figure><p>The Frequency of each console</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/502/1*r1HUTM9xIsQ95irfOUQThw.png" /><figcaption>Frequency of each console in that calendar</figcaption></figure><p><strong>Relative Frequency</strong></p><p>The Relative frequency of a qualitative variable is the Frequency divided by the sample size.</p><blockquote>Relative frequency = Frequency / Sample Size</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/785/1*rr0IoXS2CepVa6AMLKT6EA.png" /><figcaption>Relative frequency for the gaming consoles</figcaption></figure><h3>Bar graphs and Pareto graphs</h3><p><strong>Bar graph:</strong></p><p>A Bar graph is used to represent the frequency or relative frequency of a categorical data.</p><p>It is constructed as follows:</p><ul><li>On the horizontal axis provide the label for each category.</li><li>Draw a rectangle (bars) of equal width for each category. The height of each rectangle represents the frequency of each category.</li><li>Ensure the bars are not touching.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QLektjWBCzzeEc_CrsOaEA.png" /><figcaption><strong>a)</strong> Frequency bar chart of console gaming <strong>b)</strong> Relative frequency bar chart of gaming console</figcaption></figure><p><strong>Pareto Chart</strong></p><p>A Pareto chart is a bar graph in which thee rectangles are represented in a decreasing order from left to right.</p><h3>Pie Charts</h3><ul><li>Pie chart are common graphical device for displaying the relative frequency of a categorical variable.</li><li>A pie chart is a circle divided into sections with each section representing a particular category.</li><li>The size of the section is proportional to the relative frequency of category.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/603/1*7sg3JfmMLNksFPnK0BE3iw.png" /><figcaption>Pie chart of the gaming console data</figcaption></figure><h3>Cross tabulation</h3><ul><li>Cross tabulation is a tabular method for simultaneously summarizing the data of two categorical variables.</li><li>we can use cross tabulation to find the patterns in a data set.</li></ul><p>Steps for constructing a cross tabulation</p><p>Step 1: Put the categorical of one variable at the top of each column and categories of another variable at the beginning of each row.</p><p>Step 2: For each row and column combination, enter the number of observations that fall in 2 categories.</p><p>Step 3: The bottom of the table gives the column totals and the right side of the table gives the row totals.</p><p>Let us consider a example, <br><em>The table below contains information about the size (compact, midsize, or large) and the recommended gasoline (regular or premium) for a sample of ten 2011 automobiles<br>a. Construct a crosstabulation of the variables size and gasoline.<br>b. Identify any patterns.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/749/1*pSZoQLVsKy7GKFgNSI26zQ.png" /><figcaption>Automobile details</figcaption></figure><p>Cross tabulation of the above table is given below</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/644/1*xYp_Of1sfuKOnCOqS2agAA.png" /><figcaption><strong>Cross tabulation of the above given automobile details table</strong></figcaption></figure><h3>Clustering Bar graph</h3><ul><li>Clustering bar graph are useful in comparing 2 categorical variables and are often used in conjunction with the cross tabulation</li><li>Each of the bars in clustering bar graph represents a single category of one variable across all category of the other categorical variable.</li></ul><p><strong><em>Let us take an example,</em></strong><em><br>The original survey question read, Which one of the following emotions do you feel the most strongly in response to these terrorist attacks: sadness, fear, anger, disbelief, vulnerability?</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/749/1*6sCVwX3FqHGt_oM0IKyY8A.png" /><figcaption>Frequency table</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/756/1*6B6icBG2PDAqnEDCHaoS_w.png" /><figcaption>Relative frequency table</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/783/1*FjALHK7-QC7iai-MzaBEKQ.png" /><figcaption>Clustering bar graph of surveyed emotions</figcaption></figure><p>That’s all about this blog. Hope you guys got an idea about what are the graphs and tables for categorical data in stats. Let’s discuss more about the Graphs and Tables for numerical data in the next blog.</p><p><strong><em>Feel free to Comment, Like and Share this blog .</em></strong></p><p><strong>Thank you !!<br>Happy learning :)</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=161b39e6358d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Statistics — Data Classification]]></title>
            <link>https://medium.com/@bharath.ram.modder/statistics-data-classification-e4f744473dfa?source=rss-54289e19a064------2</link>
            <guid isPermaLink="false">https://medium.com/p/e4f744473dfa</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[statístics]]></category>
            <category><![CDATA[data-classification]]></category>
            <dc:creator><![CDATA[Bharath Ram K]]></dc:creator>
            <pubDate>Sun, 05 Feb 2023 06:26:17 GMT</pubDate>
            <atom:updated>2023-02-19T11:17:09.694Z</atom:updated>
            <content:encoded><![CDATA[<h3>Statistics — Data Classification</h3><p><em>Hello all,<br>This blog is for all the data science aspirants who want to have a better understanding of statistics or for those who got lost in the field stats and probs on where to start and what to study next in stats and probs while starting to studying for machine learning, then this blog is for you guys. I am going to write a series of blogs on Statistics and Probability.<br>Feel free to follow my page for more information on Stats and probs.</em></p><p>This is the 2nd blog on the statistics series for data science. If you haven’t read my previous blog click <a href="https://medium.com/@bharath.ram.modder/getting-started-with-statistics-for-data-science-1426786bc959">here</a>.</p><blockquote><strong>Lesson 2 : Data Classification</strong></blockquote><blockquote><strong>Objective</strong><br>After the end of this blog you can able to identify and differentiate the below topics<br>1) Qualitative and Quantitative<br>2) Discrete, Continuous or Neither<br>3 ) Normal, Ordinal, Interval or Ratio</blockquote><p><strong>Data Types</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/678/0*wD9_iBULq8diLBJm" /></figure><p>Majorly there are 2 types of data called Qualitative and Quantitative data.</p><ul><li><strong><em>Qualitative data </em></strong>also known as categorical data consists of labels or description of traits.</li><li><strong><em>Quantitative data</em></strong> also known as numeric data consists of counts or measurements.</li></ul><p>Quantitative data divides in to 2 subcategories namely <strong>Continuous data and Discrete data.</strong></p><p><strong><em>Continuous data</em></strong> are quantitative data that can take on an values in a given interval and are usually measurements.<br><strong>Examples </strong><br>1) The amount of time required by all to complete a marathon.<br>2) The height of children.</p><p><strong><em>Discrete data</em></strong> are quantitative data that take only particular values and are usually counts.<br><strong>Examples</strong><br>1) Number of languages an individual speaks.<br>2) The number of parts damaged during transportation.</p><p><strong><em>Levels of Measurements</em></strong></p><p>The level of measurements of variable describes the amount of information that variable contains. The four levels are</p><ol><li><strong>Nominal </strong>— Description</li><li><strong>Ordinal </strong>— Ordering</li><li><strong>Interval </strong>— Difference between level</li><li><strong>Ratio</strong>— True Zero</li><li>Data at <strong>Nominal level</strong> of measurement are quantitative data consisting of labels or names.</li><li>Data at <strong>Ordinal level</strong> of measurement are quantitative data that can be arranged in a meaningful order but calculation such ass addition or division do not make sense.</li><li>Data at <strong>interval level</strong> of measurement are quantitative data that can be arranged in a meaningful order and differences between data are meaningful entries.</li><li>Data at <strong>ratio level</strong> of measurement are quantitative data that can be ordered and differences between data entries are meaningful and zero point indicates the absence of something.</li></ol><p>That’s all about this blog. Hope you guys got an idea about what are the data types in stats. Let’s discuss more about the <a href="https://medium.com/@bharath.ram.modder/describing-data-using-graphs-and-tables-161b39e6358d">Graphs and Tables</a> in my next blog.</p><p><strong><em>Feel free to Comment, Like and Share this blog .</em></strong></p><p><strong>Thank you !!<br>Happy learning :)</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e4f744473dfa" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Getting started with Statistics for Data science]]></title>
            <link>https://medium.com/@bharath.ram.modder/getting-started-with-statistics-for-data-science-1426786bc959?source=rss-54289e19a064------2</link>
            <guid isPermaLink="false">https://medium.com/p/1426786bc959</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[probability]]></category>
            <category><![CDATA[statistics]]></category>
            <dc:creator><![CDATA[Bharath Ram K]]></dc:creator>
            <pubDate>Fri, 27 Jan 2023 19:23:29 GMT</pubDate>
            <atom:updated>2023-02-19T07:08:30.618Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*rO0M8MH4M5sfdESR.jpg" /></figure><p><em>Hello all,<br>This blog is for all the data science aspirants who want to have a better understanding of statistics or for those who got lost in the field stats and probs on where to start and what to study next in stats and probs while starting to studying for machine learning, then this blog is for you guys. I am going to write a series of blogs on Statistics and Probability. <br>Feel free to follow my page for more information on Stats and probs.</em></p><h4><strong><em>What is Statistics and why do we need it?</em></strong></h4><p>Statistics is a branch of applied mathematics that involves the collection, description, analysis and inference of conclusion from data.</p><p>Let’s see the examples of the importance of statistics in daily life using predictions</p><ol><li>In daily life, we make various predictions; for instance, we set the alarm to wake up in the morning when we think that we can not wake up in the early morning. Here, we use the statistical predictions as we find that the number of times a person is unable to wake up in the morning. By analysing the data, we come to a decision that we need to set up an alarm to get up in the early morning.</li><li>During the elections, various news reporters make predictions about which party has maximum chances of winning the elections. This prediction is made on the basis of political campaigns. With the help of campaign statistics, news reporters are able to predict the result of the elections. The statistics are based on the previous and current scenario of the election.</li></ol><p>With the help of statistical data, you can easily collect the relevant information that exists around you.</p><p>Lets dive deep into the basics of Statistics !!</p><blockquote><strong>Lesson 1 : Basics Terms used in Statistics</strong></blockquote><blockquote>Objective:<br>1) Learn basic vocabulary of statistics<br>2) Distinguish between population and sample <br>3) Distinguish among the two types of statistics<br>4) Distinguish among the two types of analysis</blockquote><p><strong>Vocabulary of Statistics</strong></p><p><strong><em>Statistics </em></strong><br>Science of gathering, describing and analysing data (Or)<br> Numerical description of sample data.</p><p><strong><em>Target Population</em></strong><br>Particular group about which the information is required and estimates is required.</p><p><strong><em>Sample</em></strong><br>A subset of population from which data are collected.</p><p><strong><em>Sample population</em></strong><br>A group from which sample is taken.</p><p><strong><em>Sampling frame</em></strong><br>A physical list of all members of sampled population.</p><p><strong><em>Census</em></strong><br>A study in which data is obtained from every member of the population.</p><p><strong><em>Variable</em></strong><br>Value or characteristic that changes among the members of population.</p><p><strong><em>Data<br></em></strong>Data are the counts of measurements or observations gathered about a specific variable in a population in order to study it.</p><p><strong><em>Parameter</em></strong><br>Numerical description of a population characteristics.</p><p><strong><em>Sample Statistics</em></strong><br>A Numerical description of sample characteristics.</p><p><strong>Population Vs Sample</strong><br><strong><em>Population</em></strong></p><ul><li>Whole group</li><li>It characteristics are called Parameters</li><li>Parameters are generally unknown</li><li>Parameters are fixed</li></ul><p><strong><em>Sample</em></strong></p><ul><li>Part of group</li><li>Its characteristics are called statistics</li><li>Statistics are always known</li><li>Statistics changes with sample</li></ul><p><strong>Branches of statistics</strong></p><ul><li><strong><em>Descriptive Statistics<br></em></strong>This branch of statistics as a science we gather, summarize and display the data.</li><li><strong><em>Inferential Statistics</em></strong><br>This branch of statistics as a science involve using descriptive statistics to estimate population parameters.</li></ul><p><strong>Types of Analysis</strong></p><ul><li><strong><em>Exploratory Analysis</em> </strong>uses data to estimate parameters.</li><li><strong><em>Confirmatory Analysis</em></strong> uses statistics to test claims about reality (hypothesis).</li></ul><p>That’s all about the some of the basics of stats which you need to know before diving deep into the other concepts. On the next lesson we will discuss about the <a href="https://medium.com/@bharath.ram.modder/statistics-data-classification-e4f744473dfa"><strong>Data Classifications</strong></a><strong>.</strong></p><p>Kindly use the comment section to share your thoughts and doubts regarding this blog.</p><p><strong>Thank you !!<br> Happy learning :)</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1426786bc959" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>