DEV Community: Michael Muthama

Neural Networks, its Inspiration and Various Components.

Michael Muthama — Fri, 09 Oct 2026 11:08:04 +0000

Introduction
In this week, I got introduced to Deep learning, and one key takeaway was on neural networks. I discovered that a neural network is a way of getting a computer to learn patterns from data the whole concept is loosely inspired by how the human brain works. For instancem in your brain, neurons receive signals, decide whether those signals are strong enough to matter, and pass a signal onward if they are. A neural network copies this idea in a much simpler mathematical form where a "neuron" takes in some numbers, combines them, and passes the result forward, layer by layer, until the network produces a final answer.

In this article we will walk through how one of these networks actually gets built and trained in code, using real screenshots from a working example.

1. Building the Architecture

This is the network's structure, built with tf.keras.Sequential. "Sequential" just means the data flows through the layers one after another, in a straight line, and with no branching.

The*input layer* does not do any calculating. It just tells the network the shape of the data coming in, based on how many columns X_train_processed has.
The three hidden layers are where the actual learning happens. Each one is Dense, meaning every neuron in that layer is connected to every neuron in the layer before it. They use "relu" as their activation function, which is a simple rule that says if a neuron's combined signal is positive, pass it on as is and if it's negative, pass on zero instead. This small rule is what lets the network learn patterns that are not just straight lines.
The output layer has 3 neurons with a "softmax" activation, which means the network is choosing between 3 possible classes. It turns the network's raw output into probabilities that add up to 1, so the class with the highest probability becomes the prediction.

2. Compiling the Model

Before a network can learn anything, it needs to be compiled. Compiling tells the network how it should learn and how it should judge its own performance.

optimizer = "adam" sets the method the network uses to adjust itself after each mistake. Adam is a popular default because it adapts how big each adjustment is as training goes on, rather than using one fixed step size the whole time.

loss = "sparse_categorical_crossentropy" is the function that measures how wrong a prediction was. This particular loss function is built for problems like this one, where there are several classes to choose from and the correct answer is given as a plain number (like 0, 1, or 2) rather than one-hot encoded.

**metrics = ["accuracy"] **doesn't affect learning at all and it is just a number reported after each epoch so you can see how the network is doing in a way that is easy to read.

3. Deciding When to Stop Training

Neural networks do not automatically know when to stop learning and thus, if left alone, they can keep training long after they have stopped genuinely improving, and start memorizing the training data instead of learning general patterns. This callback protects against that.

EarlyStopping(monitor = "val_loss", patience = 5, restore_best_weights = True)_ watches the loss on the validation data (data the network does not train on, used only to check its progress). If that validation loss does not improve for 5 epochs in a row (patience = 5), training stops early. restore_best_weights = True makes sure that when training stops, the network keeps the version of itself from its best epoch, not just whatever it looked like when it finally stopped.

Training the Model

This is where everything comes together and the network actually starts learning.

model.fit(...) feeds the training data, X_train_processed and y_train_encoded, into the network. validation_data gives it a separate set to check its progress on after each epoch, without training on it. epochs = 200 sets the maximum number of full passes through the training data, though the EarlyStopping callback from the step before may cut this short. batch_size = 32 means the network looks at 32 examples at a time before updating itself, rather than updating after every single example or waiting until it has seen the entire dataset. Finally, callbacks = [callbacks] plugs in the early stopping rule, so it's actually applied during this run.

Conclusion
In conclusion, when these four pieces are put together, they form the full life cycle of a neural network which revolves around the axis of, the architecture deciding what the network looks like, compiling decides how it learns and how it is judged, the callback decides when it should stop, and fit is the step where all of that is actually put into motion. Each piece is simple enough on its own, but the network's real behavior comes from how these choices interact with each other during training.

Mastering Python in Data Analytics:

Michael Muthama — Sun, 10 May 2026 14:16:59 +0000

What is Python?
Python is a ****interpreted programming language used for building and tell computers what to do .

Guido van Rossum is the main creator of python.

Python can be used to build websites, automate tasks, analyze data, and create machine learning applications.

Python is an interpreted, dynamically typed, general-purpose language.

Python allows analysts to automate repetitive work, reduce human errors, and process large amounts of data much faster than manual methods.

For businesses, this means:

.Faster reporting
.More accurate insights
.mproved decision-making betertter understanding of customer behaviour and business performance

Python Libraries Used in Data Analytics:

-A Python library is a collection of pre-written code that helps developers perform specific tasks more easily and efficiently.

-Instead of building everything from scratch, developers and analysts can use libraries to speed up their work.

Some popular Python libraries used in data analytics include:

Pandas

Pandas is a Python library used to clean, organize, and analyze data quickly and efficiently.

It helps analysts:

.Remove duplicates
.Handle missing values
.Filter and sort information
.Organize data into tables
.Work with CSV and Excel files
.Using Pandas increases speed and accuracy compared to manual data handling.

Requests .Requests is a Python library used to communicate with websites and APIs over the internet.

.It helps analysts retrieve data from external systems and online services for analysis.Python Libraries Used in Data Analytics

.A Python library is a collection of pre-written code that helps developers perform specific tasks more easily and efficiently.

.Instead of building everything from scratch, developers and analysts can use libraries to speed up their work.

Some popular Python libraries used in data analytics include:

Pandas
Pandas is a Python library used to clean, organize, and analyze data quickly and efficiently.

It helps analysts:

Remove duplicates
Handle missing values
Filter and sort information
Organize data into tables
Work with CSV and Excel files
Using Pandas increases speed and accuracy compared to manual data handling.

Requests
Requests is a Python library used to communicate with websites and APIs over the internet.

It helps analysts retrieve data from external systems and online services for analysis.

Python is considered one of the best programming languages for beginners because it is easy to learn and has a large and active support community.

Beginners can quickly find:

Tutorials
Documentation
Videos
Forums
Community support

Conclusion
Python has become one of the most important tools in modern data analytics because of its simplicity, flexibility, and powerful libraries.

With tools such as Pandas, Requests, Matplotlib, and Seaborn, analysts can clean, analyze, and visualize large amounts of data efficiently.

More importantly, Python helps businesses transform raw information into meaningful insights that support smarter and faster decision-making.

Understanding SQL: DDL,DML,Filtering and DCL.

Michael Muthama — Mon, 13 Apr 2026 06:55:12 +0000

SQL is an abbreviation for Standard Query Language. It is a programming language that is used to manage and work with data in databases.

Sql is a very important tool as it is used in all types of industries be it Banks, Hospitals, Schools, Mpesa. All these have databases that work behind the scenes
DBMS stands for a Database Management System which is a software used to create, manage and interact with database e.g Postgresql which i use in my class projects. We also have others like Mysql, IBM DB2,Mongo db e.t.c, There are two types of DBMS

Think of it this way; DDL builds the house, DML fills it with furniture.

In my current project, the main commands I used were CREATE, INSERT, UPDATE, and DELETE and I will take you through the 'how' shortly.

I used CREATE to create a schema called nairobi_academy, then CREATE TABLE to build three tables: students, subjects, and exam_results;

I also used ALTER TABLE to modify tables after creation, that is; adding a phone_number column, renaming credits to credit_hours, and dropping phone_number when it was no longer needed. This taught me that SQL is flexible and you can always adjust a table's structure even after it has been created.

Next, I used INSERT command, where I added 10 students, 10 subjects, and 10 exam results into the database -see example below;

I then used UPDATE command and corrected data that had changed. For instance; updating Esther's city from Nakuru to Nairobi, and fixing incorrect marks from 49 to 59.

In addition, I used DELETE command to remove a cancelled exam result from the table.

NB: The most important lesson with UPDATE and DELETE is that, always use a WHERE clause, else every single row in the table gets affected and will be deleted.

Another key lesson has been how to use SQL filtering functions e.g., WHERE, SEARCH, IN, NOT IN, COUNT & CASE WHEN.

The WHERE clause filters rows based on conditions while CASE WHEN works like an if-else statement. It creates new labels based on conditions without changing the original.

The biggest challenge this week was managing errors when running scripts multiple times. Some errors I encountered today was "relation already exists" and it kept appearing till it got frustrating. I learned that using IF NOT EXISTS and DROP SCHEMA CASCADE solves this cleanly.

I also learned the hard way that column names matter. For instance; renaming credits to credit_hours after inserting data caused errors because the INSERT still referenced the old name. Order really matters in SQL! Do not give up when creating and querying data in SQL, remain calm, research on how to debug and you've got it!

“Power BI Essentials: Tools, Functions, and Real-World Applications”

Michael Muthama — Sun, 29 Mar 2026 18:57:11 +0000

Understanding Power BI and Its Functions

In today’s data-driven world, organizations rely heavily on tools that can transform raw data into meaningful insights. One of the most powerful tools for this purpose is Microsoft Power BI. It is a business intelligence (BI) platform that enables users to connect, analyze, visualize, and share data efficiently.

What is Power BI?

Microsoft Power BI is a suite of tools developed by Microsoft that helps users create interactive dashboards and reports. It is widely used by businesses, analysts, and students to turn complex datasets into clear visual insights.

Power BI consists of several components:

Power BI Desktop – for creating reports
Power BI Service – for sharing and collaboration (cloud-based)
Power BI Mobile – for accessing reports on mobile devices

Key Features of Power BI

1. Data Connectivity

Power BI can connect to a wide range of data sources, including:

Excel files
Databases (SQL Server, Oracle)
Cloud services (Google Analytics, Azure)
Web data sources

This flexibility allows users to gather data from multiple places into one unified system.

2. Data Transformation (Power Query)

Power BI includes a powerful data cleaning tool called Power Query. It helps you:

Remove duplicates
Handle missing values
Filter and sort data
Merge and reshape datasets

This ensures your data is accurate and ready for analysis.

3. Data Modeling

Data modeling allows you to define relationships between tables. Power BI uses:

Relationships (one-to-many, many-to-one)
Calculated columns
Measures

This step is essential for building meaningful reports.

Important Functions in Power BI (DAX)

Power BI uses a formula language called DAX (Data Analysis Expressions) to perform calculations. Below are some commonly used functions:

1. SUM Function

Adds up values in a column:

Total Sales = SUM(Sales[Amount])

2. AVERAGE Function

Calculates the average of a column:

Average Sales = AVERAGE(Sales[Amount])

3. COUNT Function

Counts the number of rows:

Total Orders = COUNT(Sales[OrderID])

4. IF Function

Performs logical tests:

Profit Status = IF(Sales[Profit] > 0, "Profit", "Loss")

5. CALCULATE Function

One of the most powerful DAX functions, used to modify filters:

Total Sales in 2025 = CALCULATE(SUM(Sales[Amount]), Year = 2025)

6. FILTER Function

Filters data based on conditions:

High Sales = FILTER(Sales, Sales[Amount] > 1000)

7. RELATED Function

Fetches related data from another table:

Customer Name = RELATED(Customers[Name])

Data Visualization in Power BI

One of the standout features of Microsoft Power BI is its ability to create stunning visualizations such as:

Bar charts
Line graphs
Pie charts
Maps
Dashboards

These visuals make it easier to identify trends, patterns, and insights.

Advantages of Power BI

User-friendly interface – easy for beginners
Real-time data updates
Integration with other Microsoft tools like Excel
Powerful data analysis capabilities
Interactive dashboards

Conclusion

Microsoft Power BI is a powerful and versatile tool for data analysis and visualization. By understanding its core features and functions—especially DAX—you can transform raw data into valuable insights that support better decision-making.

Whether you are a student, business professional, or data analyst, learning Power BI is a valuable skill that can significantly enhance your ability to work with data.

“From Beginner to Expert: How to Effectively Use Microsoft Excel”

Michael Muthama — Sun, 29 Mar 2026 17:40:55 +0000

Introduction
Microsoft Excel is one of the most powerful and widely used spreadsheet applications in the world. It is part of the Microsoft Office suite and is used for data entry, analysis, visualization, and automation. From students to financial analysts, Excel plays a crucial role in managing and interpreting data efficiently.

What is Microsoft Excel?
Microsoft Excel is a spreadsheet program that organizes data into rows and columns. Each intersection of a row and column is called a cell, where data can be stored, manipulated, and analyzed.

Excel allows users to:

Perform calculations
Analyze large datasets
Automate repetitive tasks

Types of Excel Formulas

Arithmetic Formulas Used for basic mathematical operations such as addition, subtraction, multiplication, and division.

Examples:

=A1 + B1 (Addition)
=A1 - B1 (Subtraction)
=A1 * B1 (Multiplication)
=A1 / B1 (Division)

Statistical Formulas Used to analyze data.

Common functions:

=AVERAGE(A1:A10) – Calculates the average
=COUNT(A1:A10) – Counts numeric values
=MAX(A1:A10) – Finds the highest value
=MIN(A1:A10) – Finds the lowest value

Logical Formulas Used to make decisions based on conditions.

Example:

=IF(A1>50, "Pass", "Fail")
This checks if the value in A1 is greater than 50.

Text Formulas Used to manipulate text.

Examples:

=CONCAT(A1, " ", B1) – Combines text
=LEFT(A1, 4) – Extracts first 4 characters
=RIGHT(A1, 3) – Extracts last 3 characters

Date and Time Formulas Used to work with dates and times.

Examples:

=TODAY() – Returns current date
=NOW() – Returns current date and time
=DATEDIF(A1, B1, "Y") – Calculates years between dates

Lookup and Reference Formulas Used to search for data.

Examples:

=VLOOKUP(A2, A1:B10, 2, FALSE)
=HLOOKUP(A2, A1:D5, 2, FALSE)
=INDEX(A1:B10, 2, 1)
=MATCH("John", A1:A10, 0)

Benefits of Using Excel Formulas
Accuracy: Reduces human error in calculations
Efficiency: Saves time on repetitive tasks
Automation: Automatically updates results when data changes
Data Analysis: Helps uncover trends and insights