<?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 Whoopie Wanja on Medium]]></title>
        <description><![CDATA[Stories by Whoopie Wanja on Medium]]></description>
        <link>https://medium.com/@woupyie99?source=rss-16c660d5d104------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*iZn74CqUggRWJXtX.</url>
            <title>Stories by Whoopie Wanja on Medium</title>
            <link>https://medium.com/@woupyie99?source=rss-16c660d5d104------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 13:30:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@woupyie99/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[Retail Dataset from Kaggle]]></title>
            <link>https://medium.com/@woupyie99/retail-dataset-from-kaggle-950d1f1a27df?source=rss-16c660d5d104------2</link>
            <guid isPermaLink="false">https://medium.com/p/950d1f1a27df</guid>
            <dc:creator><![CDATA[Whoopie Wanja]]></dc:creator>
            <pubDate>Sat, 29 Jun 2024 20:54:41 GMT</pubDate>
            <atom:updated>2024-06-29T20:54:41.483Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>In this blog, we are exploring the <a href="https://www.kaggle.com/datasets/kyanyoga/sample-sales-data?resource=download">Retail Dataset on Kaggle</a>. The dataset was originally used for Pentaho DI Kettle but represents a rich enough sample to practice data analysis on. This task is submitted as part of the <a href="https://hng.tech/internship">HNG internship</a>. It is an interesting challenge for techies to take up in data analysis, backend, frontend, design and several other tracks. The people that make it out are top notch talent and those looking to hire should head over to the <a href="https://hng.tech/hire">website</a>.</p><p>Below, we are importing the libraries needed, reading the file, and having a quick view of the columns and datatypes. You can do the last part with df.head(). The dataset has 25 columns with 2823 rows.</p><pre>import pandas as pd<br>import matplotlib.pyplot as plt</pre><pre>df = pd.read_csv(&quot;sales_data.csv&quot;, encoding=&#39;latin-1&#39;)</pre><pre>df.shape<br>(2823, 25)</pre><p>Here are the different data types present.</p><pre>df.dtypes</pre><pre>ORDERNUMBER           int64<br>QUANTITYORDERED       int64<br>PRICEEACH           float64<br>ORDERLINENUMBER       int64<br>SALES               float64<br>ORDERDATE            object<br>STATUS               object<br>QTR_ID                int64<br>MONTH_ID              int64<br>YEAR_ID               int64<br>PRODUCTLINE          object<br>MSRP                  int64<br>PRODUCTCODE          object<br>CUSTOMERNAME         object<br>PHONE                object<br>ADDRESSLINE1         object<br>ADDRESSLINE2         object<br>CITY                 object<br>STATE                object<br>POSTALCODE           object<br>COUNTRY              object<br>TERRITORY            object<br>CONTACTLASTNAME      object<br>CONTACTFIRSTNAME     object<br>DEALSIZE             object<br>dtype: object</pre><p>Next, we look for empty cells in each column. The empty cells do not seem to be that impactful although they&#39;d be useful in further analysis. In TERRITORY, the 1074 were for North America which was misinterpreted as Not A Number. The STATE column should be dropped if doing an analysis of all the orders. It&#39;d only be useful for analysis of sales in the US.</p><pre># Find empty cells<br>empty_cells = df.isnull().sum()<br><br># Number of empty cells in each column<br>print(empty_cells)</pre><pre>ORDERNUMBER            0<br>QUANTITYORDERED        0<br>PRICEEACH              0<br>ORDERLINENUMBER        0<br>SALES                  0<br>ORDERDATE              0<br>STATUS                 0<br>QTR_ID                 0<br>MONTH_ID               0<br>YEAR_ID                0<br>PRODUCTLINE            0<br>MSRP                   0<br>PRODUCTCODE            0<br>CUSTOMERNAME           0<br>PHONE                  0<br>ADDRESSLINE1           0<br>ADDRESSLINE2        2521<br>CITY                   0<br>STATE               1486<br>POSTALCODE            76<br>COUNTRY                0<br>TERRITORY           1074<br>CONTACTLASTNAME        0<br>CONTACTFIRSTNAME       0<br>DEALSIZE               0<br>dtype: int64</pre><p>This gives all the info together.</p><pre>df.info()</pre><pre>&lt;class &#39;pandas.core.frame.DataFrame&#39;&gt;<br>RangeIndex: 2823 entries, 0 to 2822<br>Data columns (total 25 columns):<br> #   Column            Non-Null Count  Dtype  <br>---  ------            --------------  -----  <br> 0   ORDERNUMBER       2823 non-null   int64  <br> 1   QUANTITYORDERED   2823 non-null   int64  <br> 2   PRICEEACH         2823 non-null   float64<br> 3   ORDERLINENUMBER   2823 non-null   int64  <br> 4   SALES             2823 non-null   float64<br> 5   ORDERDATE         2823 non-null   object <br> 6   STATUS            2823 non-null   object <br> 7   QTR_ID            2823 non-null   int64  <br> 8   MONTH_ID          2823 non-null   int64  <br> 9   YEAR_ID           2823 non-null   int64  <br> 10  PRODUCTLINE       2823 non-null   object <br> 11  MSRP              2823 non-null   int64  <br> 12  PRODUCTCODE       2823 non-null   object <br> 13  CUSTOMERNAME      2823 non-null   object <br> 14  PHONE             2823 non-null   object <br> 15  ADDRESSLINE1      2823 non-null   object <br> 16  ADDRESSLINE2      302 non-null    object <br> 17  CITY              2823 non-null   object <br> 18  STATE             1337 non-null   object <br> 19  POSTALCODE        2747 non-null   object <br> 20  COUNTRY           2823 non-null   object <br> 21  TERRITORY         1749 non-null   object <br> 22  CONTACTLASTNAME   2823 non-null   object <br> 23  CONTACTFIRSTNAME  2823 non-null   object <br> 24  DEALSIZE          2823 non-null   object <br>dtypes: float64(2), int64(7), object(16)</pre><p>We will also check basic statistics for the numerical columns. Note, all numerical columns are picked and the formula does not discriminate IDs. Converting IDs to categorical columns to avoid numerical manipulation could be useful.</p><pre>df.describe()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/804/1*w6u9bnDIALQdRmdHQ6kS7A.png" /><figcaption>Numerical columns statistics</figcaption></figure><h3>Observations</h3><p>In the next part, we look for anomalies like duplicate rows, financial columns being in the negative, or quantities ordered being greater than 100.</p><pre># Check for duplicate rows<br>duplicate_rows = df[df.duplicated()]<br><br># Check for outliers in the quantity column<br>outliers = df[df[&#39;QUANTITYORDERED&#39;] &gt; 100]<br><br># Check for negative values in a few price columns<br>negative_prices = df[df[&#39;PRICEEACH&#39;] &lt; 0]<br>negative_prices2 = df[df[&#39;SALES&#39;] &lt; 0]<br><br># Print the number of anomalies found<br>print(&quot;Number of duplicate rows:&quot;, len(duplicate_rows))<br>print(&quot;Number of outliers in quantity:&quot;, len(outliers))<br>print(&quot;Number of negative values in unit price:&quot;, len(negative_prices))<br>print(&quot;Number of negative values in Sales:&quot;, len(negative_prices))</pre><pre>Number of duplicate rows: 0<br>Number of outliers in quantity: 0<br>Number of negative values in unit price: 0<br>Number of negative values in Sales: 0</pre><p>There were no anomalies in the dataset.</p><p>In terms of frequency, the deals are mostly medium with the average sale being 3553.</p><pre>df[&#39;DEALSIZE&#39;].value_counts().plot(kind=&#39;bar&#39;)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/560/1*VbpWijs5ZqF7cLhfF9d5og.png" /><figcaption>Graph showing Deal Size</figcaption></figure><p>Mean sales for each product line can be seen in the graph below. The average difference between Trains (lowest) and Classic Cars (Highest), seems to be a 1000.</p><pre>plt.figure(figsize=(12, 6))<br>df.groupby(&#39;PRODUCTLINE&#39;)[&#39;SALES&#39;].mean().plot(kind=&#39;bar&#39;)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/994/1*L9kzXT8mJfO7M3aw47B3pA.png" /><figcaption>Mean sales for each product line</figcaption></figure><p>There was a sales increase between 2003 and 2004 then a dip in 2005.</p><pre>df[&#39;YEAR_ID&#39;].value_counts()</pre><pre>YEAR_ID<br>2004    1345<br>2003    1000<br>2005     478In [30]:</pre><pre>df[&#39;YEAR_ID&#39;].value_counts().plot(kind=&#39;bar&#39;)<br>plt.show()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/560/1*kal4Mv1LzSxxfzK8hKSPYw.png" /><figcaption>Value count of transactions through the Years</figcaption></figure><h3>Conclusion</h3><p>Further analysis could help us find out why the cells were empty for STATE and TERRITORY and if the information can be found in other ways. This also helps the team ensure integrity in data entry is adhered to during data collection in other projects.</p><p>An investigation should also be done to evaluate any reasons behind the deep in 2005. Was the data collected throughout the year? Was there a financial crisis? Is it something in the supply chain? These would be some appropriate questions to ask.</p><p>I’d also recommend an analysis that touches on the correlation between the different columns.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=950d1f1a27df" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Writing an Enticing Career Summary]]></title>
            <link>https://medium.com/waape/writing-an-enticing-career-summary-ece2053cdbb9?source=rss-16c660d5d104------2</link>
            <guid isPermaLink="false">https://medium.com/p/ece2053cdbb9</guid>
            <category><![CDATA[about-me]]></category>
            <category><![CDATA[career-summary]]></category>
            <category><![CDATA[bio]]></category>
            <category><![CDATA[freelancing]]></category>
            <category><![CDATA[careers]]></category>
            <dc:creator><![CDATA[Whoopie Wanja]]></dc:creator>
            <pubDate>Tue, 01 Jun 2021 22:34:10 GMT</pubDate>
            <atom:updated>2021-06-01T22:34:10.690Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xkCsp7Zdy-FmVWqV" /><figcaption>Photo by <a href="https://unsplash.com/@91magazine?utm_source=medium&amp;utm_medium=referral">91 Magazine</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>The career summary portion of your profile allows you to sell yourself. Undoubtedly, many people find it difficult to do this efficiently in a few paragraphs. Think of it like an elevator pitch. You only have a few minutes to capture your reader&#39;s attention, so a whole essay won’t be ideal.</p><p>First and foremost, pick a title that clearly describes the scope of your work. You want to be specific here as a broad title will drive away clients contrary to popular belief. You want the client to think you are an <em>expert</em> as a jack of all trades would be deemed a master of none. Also, please do not make it too long. Aim to be simple and specific.</p><p>Secondly, the first line of your overview is what the client sees before clicking to read more. If possible, do not start with ‘I am …’. as that is how most profiles start, and you will not <em>stand out</em>. You could start with a testimonial from a previous client or ‘You…’. The point is to make it catchy or different. A testimonial also increases your credibility.</p><p>Thirdly, highlight what you’re good at but remember it is about the client. Describe your strengths and expertise and how they can help solve the client’s problem. Do not be shy to unveil your great skills and experiences. Use keywords that you think clients will be looking for, as that will help you optimize your profile for the algorithm.</p><p>The last thing you need to do is make sure you end with a call to action. You might be surprised how much a prompt does. A good example would be telling your reader to contact you to see if their project fits. Ensure you proofread what you have written to eliminate any grammatical errors that may convey you as unprofessional.</p><p>Now that you have a skill or two up your sleeves, you can go back to your profile, or if you do not have an account yet, please register <a href="http://waape.io/"><strong>here</strong></a> and start the journey with us!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ece2053cdbb9" width="1" height="1" alt=""><hr><p><a href="https://medium.com/waape/writing-an-enticing-career-summary-ece2053cdbb9">Writing an Enticing Career Summary</a> was originally published in <a href="https://medium.com/waape">Waape</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>