<?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 Zarau Bello baidu on Medium]]></title>
        <description><![CDATA[Stories by Zarau Bello baidu on Medium]]></description>
        <link>https://medium.com/@zeerabaidu?source=rss-d747098a9242------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*gPvzT87mld3iNFCN.jpg</url>
            <title>Stories by Zarau Bello baidu on Medium</title>
            <link>https://medium.com/@zeerabaidu?source=rss-d747098a9242------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 14:19:34 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@zeerabaidu/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[A Strategic Roadmap: From Aspiring AI Engineer to Financial Tech Pioneer]]></title>
            <link>https://medium.com/@zeerabaidu/a-strategic-roadmap-from-aspiring-ai-engineer-to-financial-tech-pioneer-5bafd17fd751?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/5bafd17fd751</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Thu, 10 Oct 2024 16:12:47 GMT</pubDate>
            <atom:updated>2024-10-10T16:12:47.150Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*cwIFVBrMj0Rizy1R" /><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><p><strong>Introduction</strong></p><p>The convergence of artificial intelligence (AI) and finance has ushered in a new era of innovation and opportunity. As an aspiring AI engineer with a passion for the financial sector, I have meticulously crafted a five-year roadmap to equip myself with the requisite skills and knowledge to excel in this dynamic field.</p><p><strong>Year 1: Building a Solid Foundation in AI and Data Science</strong></p><p>The initial year will be dedicated to establishing a robust foundation in AI and data science. I will enroll in comprehensive programs such as 3MTT and TDI, acquiring proficiency in machine learning, deep learning, and statistical analysis. To complement my formal education, I will actively engage in online courses on platforms like YouTube, Udemy, Coursera, and DataCamp, focusing on Python programming, data visualization, and natural language processing. Practical experience will be gained through participation in coding challenges and contributions to open-source AI projects.</p><p><strong>Year 2: Deepening Domain Knowledge and Practical Application</strong></p><p>In the second year, I will delve into the intricacies of the financial domain, exploring concepts such as financial markets, risk management, and quantitative finance. To bridge the gap between theory and practice, I will seek internships or part-time roles in financial institutions, applying my AI skills to real-world challenges. Simultaneously, I will continue to enhance my technical expertise by pursuing advanced courses in time series analysis, reinforcement learning, and generative models.</p><p><strong>Year 3 and 4: Academic Pursuit and Professional Development</strong></p><p>During years 3 and 4, I will pursue a master’s degree in a related field, such as computer science, data science, or financial engineering. This will provide me with a deeper understanding of theoretical concepts and research methodologies. To complement my academic pursuits, I will continue to work on advanced projects, possibly involving collaboration with researchers or industry partners, to apply my knowledge to cutting-edge problems in AI for finance.</p><p><strong>Year 5: Ph.D. and Entrepreneurial Ventures</strong></p><p>In the fifth year, I will embark on a Ph.D. program, allowing me to conduct original research and contribute to the academic community. I will focus on a specific research area within AI for finance, exploring new techniques or addressing challenging problems. Additionally, I will explore entrepreneurial opportunities, such as starting my own AI-driven financial technology company or consulting for firms in the industry.</p><p><strong>Conclusion</strong></p><p>This five-year roadmap outlines a strategic plan to develop the skills and knowledge necessary to become a successful AI engineer specializing in finance. By combining formal education, practical experience, and continuous learning, I aim to contribute to the advancement of AI in the financial sector and make a significant impact on the industry.</p><p>#My3MTT and #3MTTWeeklyReflection, <a href="http://twitter.com/3mttNigeria">@3mttNigeria</a>, <a href="http://twitter.com/IHSTowers">@IHSTowers</a>, <a href="http://twitter.com/NITDANigeria">@NITDANigeria</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5bafd17fd751" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[working with categorical data in python]]></title>
            <link>https://medium.com/@zeerabaidu/working-with-categorical-data-in-python-d47ad98becaf?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/d47ad98becaf</guid>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Wed, 31 Jul 2024 20:37:33 GMT</pubDate>
            <atom:updated>2024-07-31T20:37:33.876Z</atom:updated>
            <content:encoded><![CDATA[<p>Categorical data types are common in data analysis, often representing qualitative variables like categories, labels, or groups. In Python, handling categorical data efficiently is crucial for accurate and meaningful data analysis. This article will explore how to work with categorical data types using Python libraries such as `pandas`, `numpy`, and `scikit-learn`, with practical code examples.</p><p>Understanding Categorical Data</p><p>Categorical data refers to variables that can take on a limited, fixed number of possible values. These categories can be:</p><p>1. norminal: Categories without a natural order (e.g., colors, names).<br>2. Ordinal: Categories with a natural order (e.g., educational levels, ratings).</p><p>Working with Categorical Data in Python</p><p>1. Using `pandas`</p><p>`pandas` is a powerful library for data manipulation and analysis. It provides the `Categorical` data type specifically designed for handling categorical data.</p><p>Creating Categorical Data</p><p>To create a categorical variable in `pandas`, use the `Categorical` constructor or convert existing columns.</p><p>import pandas as pd</p><p>Example data<br>data = {‘Color’: [‘Red’, ‘Blue’, ‘Green’, ‘Red’, ‘Blue’]}<br>df = pd.DataFrame(data)</p><p>Convert the ‘Color’ column to a categorical data type<br>df[‘Color’] = pd.Categorical(df[‘Color’])</p><p>print(df[‘Color’])<br>print(df[‘Color’].cat.categories)</p><p>Adding Order to Categories</p><p>If you have ordinal data, you can specify the order of categories.</p><p>Define an ordered categorical data type<br>categories = [‘Low’, ‘Medium’, ‘High’]<br>df[‘Priority’] = pd.Categorical([‘High’, ‘Medium’, ‘Low’, ‘Medium’], categories=categories, ordered=True)</p><p>print(df[‘Priority’])<br>print(df[‘Priority’].cat.categories)<br>print(df[‘Priority’].cat.ordered)</p><p>Encoding Categorical Data</p><p>Categorical data often needs to be encoded into numerical values for machine learning models.</p><p>Encoding categorical data<br>df[‘Color_Code’] = df[‘Color’].cat.codes<br>print(df)</p><p>2. Using `numpy`</p><p>`numpy` does not directly handle categorical data, but you can use it for encoding purposes.</p><p>import numpy as np</p><p>Example data<br>colors = np.array([‘Red’, ‘Blue’, ‘Green’, ‘Red’, ‘Blue’])<br>unique_colors, color_codes = np.unique(colors, return_inverse=True)</p><p>print(‘Unique Colors:’, unique_colors)<br>print(‘Color Codes:’, color_codes)</p><p>Summary</p><p>Handling categorical data effectively is essential for accurate data analysis and machine learning. `pandas` provides robust tools for managing categorical variables, including creating, ordering, and encoding categories. `numpy` can be used for basic encoding tasks, while `scikit-learn` offers advanced preprocessing options like one-hot and ordinal encoding. Mastering these techniques will improve your ability to preprocess and analyze categorical data in Python.</p><p>#MY3MTT</p><p>#MY3MTTWeeklyReflection</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d47ad98becaf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[5 Common regression evaluation metrics]]></title>
            <link>https://medium.com/@zeerabaidu/5-common-regression-evaluation-metrics-41c63117fbd5?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/41c63117fbd5</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Sun, 21 Apr 2024 07:31:10 GMT</pubDate>
            <atom:updated>2024-04-21T07:31:10.027Z</atom:updated>
            <content:encoded><![CDATA[<p>5 Common regression evaluation metrics</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*c3ynt7NQzp0aQUKk" /><figcaption>Photo by h heyerlein on Unsplash</figcaption></figure><p>Model evaluation is one of the critical steps in building a good machine learning model. in this article we will briefly talk about some common metrics used to measure model effectiveness</p><p>1. Mean Absolute Error (MAE):<br> MAE measures the average magnitude of errors in a set of predictions.</p><pre>from sklearn.metrics import mean_absolute_error<br>   <br>   y_true = [3, -0.5, 2, 7]<br>   y_pred = [2.5, 0.0, 2, 8]<br>   mae = mean_absolute_error(y_true, y_pred)<br>   print(&quot;MAE:&quot;, mae)<br></pre><p>2. Mean Squared Error (MSE):<br> MSE measures the average squared difference between the actual and predicted values.</p><pre>  from sklearn.metrics import mean_squared_error<br>   <br>   mse = mean_squared_error(y_true, y_pred)<br>   print(&quot;MSE:&quot;, mse)<br></pre><p>3. Root Mean Squared Error (RMSE):<br> RMSE is the square root of the MSE, which gives a similar interpretation but in the same units as the target variable.</p><pre>import numpy as np<br>from sklearn.metrics import mean_squared_error<br>   <br>   mse = mean_squared_error(y_true, y_pred)<br>   rmse = np.sqrt(mse)<br>   print(&quot;RMSE:&quot;, rmse)<br></pre><p>4. R-squared (R^2) Score:<br> R-squared measures the proportion of the variance in the dependent variable that is predictable from the independent variable(s).</p><pre>from sklearn.metrics import r2_score<br>   <br>   r2 = r2_score(y_true, y_pred)<br>   print(&quot;R-squared:&quot;, r2)<br></pre><p>5. Mean Absolute Percentage Error (MAPE):<br> MAPE expresses forecast accuracy as a percentage, making it easier to interpret.</p><pre>def mean_absolute_percentage_error(y_true, y_pred): <br>       return np.mean(np.abs((y_true - y_pred) / y_true)) * 100<br>   <br>   mape = mean_absolute_percentage_error(np.array(y_true), np.array(y_pred))<br>   print(&quot;MAPE:&quot;, mape)<br></pre><p>These metrics provide different perspectives on the performance of regression models and are commonly used to evaluate their effectiveness.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=41c63117fbd5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[how to visualize data with matplotlib]]></title>
            <link>https://medium.com/@zeerabaidu/how-to-visualize-quantity-distribution-in-matplotlib-1289f8f1d960?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/1289f8f1d960</guid>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Tue, 27 Feb 2024 11:31:35 GMT</pubDate>
            <atom:updated>2024-02-27T11:32:54.763Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*jVwoFME95GsVBmp9" /><figcaption>Photo by <a href="https://unsplash.com/@lukechesser?utm_source=medium&amp;utm_medium=referral">Luke Chesser</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><blockquote>“Data visualization is a great way to simplify data and show it in a form that is understandable, insightful, and actionable. Data visualization is being increasingly seen as the vital final step of any successful data-driven analytics plan.” — Caroline Lee</blockquote><p>In the fast-paced world of data analysis, the ability to extract meaningful insights from raw data is invaluable. Data visualization serves as a potent tool in this endeavor, allowing analysts to communicate complex information effectively. Among the arsenal of data visualization libraries, Matplotlib stands out as a cornerstone, offering a diverse array of chart types and customization options. In this article, we delve into the depths of Matplotlib, exploring its versatility through various types of charts.</p><h3>Introduction to Matplotlib</h3><p>Matplotlib is a widely-used Python library for creating static, interactive, and animated visualizations. Renowned for its flexibility and ease of use, Matplotlib empowers users to generate a wide range of charts, from simple line plots to intricate heatmaps. Before we embark on our journey, let’s ensure Matplotlib is installed:</p><pre>pip install matplotlib</pre><p>With Matplotlib installed, let’s dive into the world of data visualization.</p><h3>Line Plot</h3><p>A line plot is one of the simplest yet most powerful visualization tools, ideal for showcasing trends and patterns over time or across categories. Here’s a basic example:</p><pre>import matplotlib.pyplot as plt</pre><pre># Sample data<br>x = [1, 2, 3, 4, 5]<br>y = [10, 15, 7, 10, 12]</pre><pre># Line plot<br>plt.plot(x, y)<br>plt.title(&#39;Line Plot&#39;)<br>plt.xlabel(&#39;X-axis&#39;)<br>plt.ylabel(&#39;Y-axis&#39;)<br>plt.show()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/563/1*BhOAgOeC9U86papzVdMvxg.png" /></figure><h3>Bar Plot</h3><p>Bar plots are effective for comparing categorical data or displaying the distribution of a variable. Let’s illustrate this with a bar plot showcasing sales data for different products:</p><pre># Sample data<br>products = [&#39;Product A&#39;, &#39;Product B&#39;, &#39;Product C&#39;, &#39;Product D&#39;]<br>sales = [100, 150, 120, 200]</pre><pre># Bar plot<br>plt.bar(products, sales)<br>plt.title(&#39;Bar Plot&#39;)<br>plt.xlabel(&#39;Products&#39;)<br>plt.ylabel(&#39;Sales&#39;)<br>plt.show()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/571/1*_VqdtM_8_vKdVXtaGbKCUA.png" /></figure><h3>Histogram</h3><p>Histograms are indispensable for visualizing the distribution of a continuous variable. Here’s how to create a histogram illustrating the distribution of exam scores:</p><pre># Sample data<br>scores = [85, 90, 70, 95, 80, 75, 85, 90, 95, 100]</pre><pre># Histogram<br>plt.hist(scores, bins=10)<br>plt.title(&#39;Histogram&#39;)<br>plt.xlabel(&#39;Scores&#39;)<br>plt.ylabel(&#39;Frequency&#39;)<br>plt.show()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*nC-NZZ3VYJyoiVS_tdh_vg.png" /></figure><h3>Scatter Plot</h3><p>Scatter plots are excellent for visualizing the relationship between two continuous variables. Let’s generate a scatter plot to explore the correlation between study hours and exam scores:</p><pre># Sample data<br>hours = [4, 5, 6, 7, 8, 9]<br>scores = [70, 75, 80, 85, 90, 95]</pre><pre># Scatter plot<br>plt.scatter(hours, scores)<br>plt.title(&#39;Scatter Plot&#39;)<br>plt.xlabel(&#39;Study Hours&#39;)<br>plt.ylabel(&#39;Exam Scores&#39;)<br>plt.show()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/563/1*bFaQFpTGA8ev2L__TNKbAA.png" /></figure><h3>Conclusion</h3><p>Matplotlib empowers data analysts with a rich toolkit for creating insightful visualizations. From line plots to histograms, scatter plots to bar plots, the versatility of Matplotlib knows no bounds. By leveraging its vast array of chart types and customization options, analysts can unlock the hidden potential of their data, unraveling patterns and trends with ease. So, embrace the power of Matplotlib, and let your data speak volumes through captivating visualizations.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1289f8f1d960" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data preparation with pandas]]></title>
            <link>https://medium.com/@zeerabaidu/data-preparation-with-pandas-c83ee672cbb1?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/c83ee672cbb1</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-cleaning]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Tue, 27 Feb 2024 10:43:03 GMT</pubDate>
            <atom:updated>2024-02-27T10:43:03.063Z</atom:updated>
            <content:encoded><![CDATA[<p>how to achieve an error free dataset</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*mD5Y5QnGSl9hB5Ij" /><figcaption>Photo by <a href="https://unsplash.com/@campaign_creators?utm_source=medium&amp;utm_medium=referral">Campaign Creators</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>in most of cases, raw data is filled with error like inconsistent formatting,missing values and duplicates which can have a great impact on the end result. in this article we will talk about how to detect missing values as well as how to deal with it.</p><h4>import library and constructing the dataframe</h4><pre>import pandas as pd<br><br>#constructing and storing into example2<br>example2 = pd.DataFrame([[1,      &quot;&quot;, 7], <br>                         [2,      5,      8], <br>                         [&quot;&quot;, 6,      9]])<br>example2</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/128/1*Y_dSFham0baHBDRtRz2r_g.png" /></figure><h4>finding the overview of data</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/308/1*Wh0kzgmyTgzrGHLU4LFN-g.png" /></figure><p>as we can see, our data has rows and 3 columns with different data types like int64 and objects</p><h4>missing values</h4><p>missing values are common errors found in data especially from a third party source data. pandas has features we can use to detect if we have such issue using <strong><em>isnull</em></strong> and <strong><em>notnull </em></strong>tool. both <strong><em>isnull</em></strong> <strong><em>and notnull </em></strong>return a boolean mask value .</p><pre>df.isnull()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/209/1*C6T23GVb72DoctP95ENDsg.png" /></figure><p>the false value indicate a valid cell which means according to pandas an empty string is equal to an object</p><h4>dropping of missing value cells</h4><p>pandas <strong>dropna()</strong> can be used to drop or delete missing value cells, you can use the axis argument to drop a specified column. in case where youwant to drop based on a condition pandas dropna has “how” and “thresh” to indicate that</p><pre>example2.dropna()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/316/1*DB6vxkrDFNGY9oBHndc0ZA.png" /></figure><h4>filling missing values</h4><p>pandas provide us with another way to deal with missing values instead of discarding them away we can use the <strong>fillna()</strong> method to replace empty cell with a value we choose. we can as well used parameters inside it like <strong>ffill</strong> to fill with the previous valid value or <strong>bfill</strong> to fill with next valid value.</p><h4>conclusion</h4><p>in this article we covered how we can detect and deal with missing values in our data</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c83ee672cbb1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to get and know your data with pandas]]></title>
            <link>https://medium.com/@zeerabaidu/how-to-get-and-know-your-data-with-pandas-5af2e737cd2a?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/5af2e737cd2a</guid>
            <category><![CDATA[pandas]]></category>
            <category><![CDATA[datanalysis]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Tue, 27 Feb 2024 08:54:33 GMT</pubDate>
            <atom:updated>2024-02-27T08:54:33.919Z</atom:updated>
            <content:encoded><![CDATA[<h4>uploading and exploring your dataset</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*OBbgyEU4s-iHsdFK" /><figcaption>Photo by <a href="https://unsplash.com/@austindistel?utm_source=medium&amp;utm_medium=referral">Austin Distel</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>whatever you are set to do,understanding the situation is important as it will help you in navigating your way which will optimize the outcome. same goes with data,knowing the nitty-gritty of your data that can shape the approach you might take in achieving your goals. in this article you will know how to explore your data for the purpose of understanding it.</p><h4>importing the necessary library</h4><p>before you can used pandas to explore the dataset, you should make sure the library is install in your environment and then import it</p><pre># the pd is an alias generally used for pandas<br>import pandas as pd</pre><p>it doesn&#39;t show anything if the script run successfully.</p><h4>upload the data into a pandas dataFrame</h4><p>after that you then upload your data using one of the most powerful pandas feature <strong>read_csv(). read_csv </strong>allows uploading dataset from a <strong>file</strong> or from <strong>url</strong> and it support other file format which can be specified using the <strong>sep</strong> argument of read_csv</p><pre>#reading the dataset file and storing it in a variable called df<br>#(it can be anything)<br><br>df = pd.read_csv(&#39;/home/zeera/practice/customer_shopping_data_TDI.csv&#39;)<br>#using the head() to print the top value of the data<br>df.head()<br></pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/865/1*SOnvYA6HS8qMPkfdVmCEXg.png" /><figcaption>code output</figcaption></figure><h4>knowing the number of records and observation in the dataset</h4><p>now that you have a preview about your data, the next is to know how many columns and rows your data has, pandas has several features to get that and one is using the<strong> shape</strong> attribute , the shape returns the counts of both x and y axis and it can be just one of them if specified.</p><pre>df.shape</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/164/1*LsoAX7NHJQIH0XxRqTzeCw.png" /><figcaption>code output</figcaption></figure><h4>knowing names of dataset columns</h4><p>next is knowing the column name of your dataset, it would be much easier knowing what columns are included in your dataset and pandas has a feature for that,using columns methods will print the list of columns names you had and the data type of each column.</p><pre>df.columns</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/628/1*5_N6vYB3AuYuFUuRud7_Cg.png" /><figcaption>code output</figcaption></figure><h4>printing a single column</h4><p>when you want to explore your data more like printing out a whole column to understand its content better then sub-setting is what you should do, pandas subset allows accessing a specified column or row using double [],using single [] return the same result but as a series. so is upto to you to decide based on what you need.</p><pre># this will print the payment_method column<br>df[[&#39;payment_method&#39;]]</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/212/1*QVpzp_kzQVJtHaYaGqcdaA.png" /><figcaption>code output</figcaption></figure><h4>finding the count of unique value in a column</h4><p>you might also want to know the number of value occurrences in a column, for example; the “category” columns consist of different product values and pandas has value_counts() method for that. it return value along with its total number of occurrence.</p><pre># finding the each category type and its total<br>df.value_counts(&#39;category&#39;)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/394/1*Ll-YIucfOuaB3CtcraIsjA.png" /></figure><h4>description of the dataset</h4><p>lastly, pandas has yet another powerful feature that is used for describing the whole dataset, it give a summary to the data and its content. by default <strong>describe()</strong> function return the statistical summary of data, but you can use the <strong>include=all</strong> argument to include the categorical values as well.</p><pre>df.describe()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*cH0E2ICjtjhAOyJSMtcP0A.png" /></figure><h4>conclusion</h4><p>as you can see pandas is incredible when it comes to manipulating data, we covered different features it offers that can make our work simple and easy.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5af2e737cd2a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Numpy Array Manipulation Methods Every Data Scientist Needs To Know]]></title>
            <link>https://medium.com/@zeerabaidu/numpy-array-manipulation-methods-every-data-scientist-needs-to-know-d5a0d26ba6af?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/d5a0d26ba6af</guid>
            <category><![CDATA[numpy-array]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Sun, 11 Feb 2024 12:19:33 GMT</pubDate>
            <atom:updated>2024-02-11T12:19:33.902Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="numpy libraryimage" src="https://cdn-images-1.medium.com/max/646/1*ohXKD3FJhC0E-8bj_DWTOw.jpeg" /></figure><h3><strong>Introduction</strong></h3><p>In the world of data science and machine learning, Numpy is an important library in Python. It provides a high-performance multidimensional array object, and tools for working with these arrays. This article will talk on some of Numpy array manipulation methods with code examples.</p><h4>reshape()</h4><p>numpy reshape methods allow change of original array shape without changing the data, reshape methods take three argument <strong>reshape(ndarray,new shape,[order ])</strong>the arguments inside bracket are optional</p><pre>import numpy as np<br>a = np.arange(6)<br>np.reshape(a,(2,3)<br><br><br>array([[0, 1],<br>       [2, 3],<br>       [4, 5]])</pre><h4><strong>resize()</strong></h4><p>numpy resize methods work similar like <strong>reshape()</strong> only that <strong>resize()</strong> changes the original data to fill up the new size if the old array is not equal to it. it also doesn’t consider axis which make it unsuitable to fix axis object like image e.t.c. its syntax is <strong>resize(ndarray,(shape of new array))</strong></p><pre>a=np.array([[0,1],[2,3]])<br>np.resize(a,(2,3))<br><br><br>array([[0, 1, 2],<br>       [3, 0, 1]])</pre><h4>ravel()</h4><p>numpy ravel methods turn array of list into 1D flattened array,it remove the input array shape and return a vector. its syntax is <strong>ravel(ndarray,[order])</strong></p><pre>x = np.array([[1, 2, 3], [4, 5, 6]])<br>np.ravel(x)<br><br><br>array([1, 2, 3, 4, 5, 6])<br>concatenate()</pre><h4>concatenate()</h4><p>numpy concatenate method works similar to <strong>ravel()</strong> with added flexibility, concatenate joins different arrays and allows choosing of axis on which arrays would be join and it must be of the same size. its syntax is <strong>concatenate((a,b,c),axis=0)</strong> 0 is the default value if not specified .</p><pre>a = np.array([[1, 2], [3, 4]])<br><br>b = np.array([[5, 6]])<br><br>np.concatenate((a, b), axis=0)<br><br>array([[1, 2],<br>       [3, 4],<br>       [5, 6]])<br><br>np.concatenate((a, b), axis=None)<br><br>array([1, 2, 3, 4, 5, 6])</pre><h3>split()</h3><p>Numpy split method divides array into multiple sub-arrays. split return a number of arrays specified in the indices argument. its syntax is <strong>split(ndarray,no_of_indices or sections,axis=0)</strong></p><pre>x = np.arange(9.0)<br><br>np.split(x, 3)<br>[array([0.,  1.,  2.]), array([3.,  4.,  5.]), array([6.,  7.,  8.])]</pre><h4>transpose()</h4><p>Numpy transpose method change array axis format, it come handy when there’s a need to change data from wide or long type. it doesn’t work with vector array as it return the same input. its syntax is <strong>transpose(ndarray,axis = none)</strong></p><pre>a = np.array([[1, 2], [3, 4]])<br><br>a<br>array([[1, 2],<br>       [3, 4]])<br><br><br>np.transpose(a)<br>array([[1, 3],<br>       [2, 4]])</pre><h4>flip()</h4><p>Numpy flip method return reversed element of array, The shape of array is preserved, but the elements are reordered. its syntax is <strong>flip(ndarray,axis)</strong></p><pre>A = np.arange(8).reshape((2,2,2))<br><br>A<br>array([[[0, 1],<br>        [2, 3]],<br>       [[4, 5],<br>        [6, 7]]])<br><br>np.flip(A, 0)<br>array([[[4, 5],<br>        [6, 7]],<br>       [[0, 1],<br>        [2, 3]]])</pre><h3>Conclusion</h3><p>These are just a few examples of the vast array of methods that Numpy offers for array manipulation. Understanding and using these methods can greatly increase the efficiency of your data processing tasks.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d5a0d26ba6af" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data Ethics Challenges: A Case Study On Algorithm Design]]></title>
            <link>https://medium.com/@zeerabaidu/unethical-data-practice-a-case-study-on-marketers-e3b6f84ec161?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/e3b6f84ec161</guid>
            <category><![CDATA[data-ethics]]></category>
            <category><![CDATA[mental-health]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Sat, 27 Jan 2024 09:15:01 GMT</pubDate>
            <atom:updated>2024-02-04T07:52:15.391Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*2bbqC2MCjfRxJvSs" /><figcaption>Photo by NASA on Unsplash</figcaption></figure><p>From health care,technology,transportation and food data are used in various stages to make decision, And those decisions can be good or bad depending on the type of data used or the processes followed. In this article we will discuss Algorithm Design bias and how they affect Data Ethics Principles.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*0WKcnD6NuUjAtqJH" /><figcaption>Photo by Claudio Schwarz on Unsplash</figcaption></figure><p><strong>Twitter image crop AI</strong></p><p>Twitter introduced an AI-powered image cropping algorithm in 2018 to automatically crop preview images to display the most relevant part. soon after its released user began to notice unusual output as it tend to prioritise white skin individuals over black. another experiment also on dogs founds thesame bias. it also focus on women over men</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e3b6f84ec161" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introduction to Machine Learning]]></title>
            <link>https://medium.com/@zeerabaidu/introduction-to-machine-learning-c8289518f674?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/c8289518f674</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Sun, 02 Apr 2023 07:29:41 GMT</pubDate>
            <atom:updated>2023-04-02T07:29:41.908Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*XiOXxZDYlLWToFBc" /><figcaption>Photo by <a href="https://unsplash.com/@alpridephoto?utm_source=medium&amp;utm_medium=referral">Андрей Сизов</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Machine learning is one of the most exciting and rapidly growing fields in technology today. It has the potential to revolutionize our world by helping us make sense of vast amounts of data, automate mundane tasks, and even create new products and services. In this article, we’ll take a look at the history of machine learning, its current state, and where it’s headed in the future.</p><h3>History of Machine Learning</h3><p>The history of machine learning can be traced back to the early days of computing, when scientists and engineers first began exploring the idea of creating machines that could learn from data. In the 1950s and 1960s, researchers like Arthur Samuel and Frank Rosenblatt developed some of the first machine learning algorithms, which were used to teach computers to play games like checkers and recognize patterns in images.</p><p>Over the years, machine learning has evolved to include a wide range of techniques and applications, from simple linear regression models to complex neural networks. Thanks to advances in computing power, data storage, and algorithmic design, machine learning has become a powerful tool for businesses and organizations across many different industries.</p><h3><strong>Fairness in Machine Learning</strong></h3><p>As machine learning becomes more pervasive, concerns about fairness and bias have become increasingly important. Machine learning algorithms are only as good as the data they are trained on, and if that data contains biases or discriminatory patterns, the algorithms themselves can perpetuate those biases.</p><p>To address these concerns, researchers are developing techniques to ensure that machine learning algorithms are fair and unbiased. These techniques include things like data anonymization, bias mitigation strategies, and fairness metrics that can be used to evaluate the performance of machine learning algorithms.</p><h3><strong>Techniques of Machine Learning</strong></h3><p>There are several techniques used in machine learning, including <strong><em>supervised learning, unsupervised learning, and reinforcement learning.</em></strong></p><p>Supervised learning involves training a machine learning algorithm on a labeled dataset, where the desired output is known. The algorithm learns to predict the output based on the input data.</p><p>Unsupervised learning involves training a machine learning algorithm on an unlabeled dataset, where the desired output is not known. The algorithm learns to identify patterns in the data without any explicit guidance.</p><p>Reinforcement learning involves training a machine learning algorithm to make decisions based on feedback from the environment. The algorithm learns to maximize a reward signal by taking actions that lead to positive outcomes.</p><p>Other techniques used in machine learning include deep learning, which involves training artificial neural networks, and transfer learning, which involves applying knowledge learned from one task to another task.</p><h3><strong>conclusion</strong></h3><p>Machine learning is a powerful tool that has the potential to revolutionize many industries. However, it is important to ensure that machine learning algorithms are designed and used in a way that is fair and ethical. By using diverse and representative data sets, evaluating algorithms for fairness and bias, and involving stakeholders in the development and deployment of machine learning algorithms, we can ensure that machine learning is used in a way that benefits everyone.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c8289518f674" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[exploring the power of numpy array]]></title>
            <link>https://medium.com/@zeerabaidu/exploring-the-power-of-numpy-array-1f901e7d1d59?source=rss-d747098a9242------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f901e7d1d59</guid>
            <category><![CDATA[arrays]]></category>
            <category><![CDATA[numpy]]></category>
            <category><![CDATA[python-programming]]></category>
            <dc:creator><![CDATA[Zarau Bello baidu]]></dc:creator>
            <pubDate>Fri, 24 Mar 2023 06:11:17 GMT</pubDate>
            <atom:updated>2023-03-24T06:13:05.234Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*r8cCrympjP3ND7KCY0aisQ.png" /></figure><h3>introduction</h3><p>NumPy is an open-source library for the Python programming language. It provides support for large, multi-dimensional arrays and matrices, along with a large collection of high-level mathematical functions to operate on these arrays.</p><p>NumPy can be used to perform various mathematical operations on arrays. It adds powerful data structures to Python that ensure efficient computation with arrays and matrices and provides a huge library of high-level mathematical functions to manipulate these arrays and matrices.</p><p>What is a NumPy Array?</p><p>A NumPy array is a multi-dimensional container for data of the same data type. It is similar to a list in Python, but with additional functionality specifically designed for numerical computations. NumPy arrays are designed to handle large data sets efficiently, making them ideal for data analysis, machine learning, and other scientific applications.</p><h3>installing numpy</h3><p>before you can use numpy you have to install and import numpy. you can install it using the following command in your terminal</p><pre>pip install numpy</pre><p>after installation then you import numpy in your project folder with the code editor you preferred.</p><pre>import numpy as np</pre><p>the above command gives you access to numpy and its functions. numpy is shorten to np for better readability of code and its widely used by most numpy users.</p><h3>how to initialize an array</h3><p>One way we can initialize NumPy arrays is from Python lists, using nested lists for two- or higher-dimensional data. for example</p><pre># one dimentional array<br>a = np.array([1, 2, 3, 4, 5, 6])<br><br>#two or higher dimentional array<br>a = np.array([[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12]])</pre><p>You can access elements within an array using square brackets. Note that NumPy indexing starts at 0 when accessing items. This means that if you want to access the first element of the array, you access element ‘0’.</p><pre>(a[0])<br><br># output<br>[1 2 3 4]</pre><p>An array is usually a fixed-size container of items of the same type and size. The number of dimensions and items in an array is defined by its shape.</p><p>In NumPy, dimensions are called <strong>axes</strong>. This means that if you have a 2D array that looks like this:</p><pre>[[0., 0., 0.],<br> [1., 1., 1.]]</pre><p>Your array has 2 axes. The first axis has a length of 2 and the second axis has a length of 3.</p><h3>methods of creating an array</h3><p>there is few methods you can use to create array in numpy</p><ul><li>using the np.array() function</li><li>using np.ones() to create all one element array</li><li>using np.zeros() to create all zero element array</li><li>using the np.arange() function to create range array</li><li>using np.linspace() to create an array with values that are spaced linearly in a specified interval</li></ul><pre># np.array<br>a = np.array([1, 2, 3])<br><br><br>#np.ones<br>np.ones(2)<br>#output <br>array([1., 1.])<br><br><br><br>#np.zeros<br>np.zeros(2)<br>#output<br>array([0., 0.])<br><br># np.arange<br>np.arange(4)<br>#output<br>array([0, 1, 2, 3])<br><br># np.linspace<br>np.linspace(0, 10, num=5)<br>#output<br>array([ 0. ,  2.5,  5. ,  7.5, 10. ])</pre><h3>numpy array methods</h3><p>there several methods used to manipulate an array.</p><ul><li>using the np.concatenate to add different arrays together</li><li>using the np.sort to sort array data in ascending order</li><li>using indexing to remove and keep the element you want</li><li>using indexing and slicing to change or alter element</li></ul><pre>a = np.array([1, 2, 3, 4])<br>b = np.array([5, 6, 7, 8])<br><br>np.concatenate((a, b))<br>#output<br>array([1, 2, 3, 4, 5, 6, 7, 8])<br><br><br>arr = np.array([2, 1, 5, 3, 7, 4, 6, 8])<br><br>np.sort(arr)<br>#output<br>array([1, 2, 3, 4, 5, 6, 7, 8])<br><br>arr1 = arr[2:]<br>#output<br>array([3 , 4, 5, 6, 7, 8])<br><br><br>data = np.array([1, 2, 3])<br><br>data[1]<br>2<br><br>data[0:2]<br>array([1, 2])<br><br>data[1:]<br>array([2, 3])<br><br>data[-2:]<br>array([2, 3])</pre><h3>numpy array attribute</h3><p>NumPy arrays have several attributes that can be useful in data analysis. Some of the most common attributes include:</p><p>ndarray.ndim will tell you the number of axes, or dimensions, of the array.</p><p>ndarray.size will tell you the total number of elements of the array. This is the <em>product</em> of the elements of the array’s shape.</p><p>ndarray.shape will display a tuple of integers that indicate the number of elements stored along each dimension of the array. If, for example, you have a 2-D array with 2 rows and 3 columns, the shape of your array is (2, 3).</p><p>dtype: the data type of the array (e.g. int, float, etc.)</p><p>arr.reshape() <strong>gives</strong> <strong>the</strong> <strong>array</strong> a new shape without changing the data.</p><p><strong>Note</strong> that when <strong>using</strong> the reshape method, the array you <strong>create</strong> <strong>must</strong> have the same number of elements as the original array. If you start with an array with 12 elements, <strong>yo</strong>u need to make sure that <strong>the</strong> new array also <strong>contains</strong> a total of 12 elements.</p><pre><br><br>array_example = np.array([[[0, 1, 2, 3],<br><br>                           [4, 5, 6, 7]],<br><br><br>                          [[0, 1, 2, 3],<br><br>                           [4, 5, 6, 7]],<br><br><br>                          [[0 ,1 ,2, 3],<br><br>                           [4, 5, 6, 7]]])<br>array_example.ndim<br>3<br><br>array_example.size<br>24<br><br>array_example.shape<br>(3, 2, 4)<br><br>a = np.arange(6)<br><br>print(a)<br>[0 1 2 3 4 5]<br><br>print(array_example.dtype)<br><br>b = a.reshape(3, 2)<br>print(b)<br>[[0 1]<br> [2 3]<br> [4 5]]</pre><h3>numpy array with mathematics</h3><p>NumPy arrays come with a lot of built-in functions and methods for performing mathematical operations. For example, we can perform element-wise addition, subtraction, multiplication, and division using the +, -, *, and / operators respectively. Here&#39;s an example:</p><pre>a = np.array([1, 2, 3])<br>b = np.array([4, 5, 6])</pre><pre>print(a + b)<br>print(a - b)<br>print(a * b)<br>print(a / b)</pre><p>This will output:</p><pre>array([5, 7, 9])<br>array([-3, -3, -3])<br>array([ 4, 10, 18])<br>array([0.25, 0.4 , 0.5 ])</pre><p>In conclusion, NumPy arrays are a powerful tool for working with large amounts of data. They provide a lot of built-in functions and methods for performing mathematical operations, making them an essential library for scientific computing and data analysis.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f901e7d1d59" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>