Level -0 Report - Nidarshana P S

7 / 9 / 2025

---

TASK 2: API

An API or Application Programming Interface is a set of rules that allows different application software to interact and communicate with each other acting as a mediator of software to receive and deliver the response.

TASK 3: Working With GITHUB

First I familiarized myself with Github workflows, issues, commit changes and pulling request. Steps 1.First I forked the repository and cloned it. 2.Then created a branch and made necessary changes in the code. 3.Pushed the changes to github and pulled request.

TASK 4: GET FAMILIAR WITH LINE COMMAND IN UBUNTU

Familiarized anmd learnt the linux commands. 1.Created a folder,within which I created 2600 subfolders and two text files. 2.Concatenated the text files and displayed the result in command prompt terminal.

TASK 5:BUILD YOUR OWN LINEAR REGRESSION MODEL-FROM SCRATCH

Linear Regression Linear Regression is a machine learning method that predicts a continuous value (like house price, height, marks, etc.) using a straight line. The formula is: y = wx + b where: predicted value ,input (feature), slope (how much changes when changes), intercept.

Error (Cost Function) We need to check how wrong our predictions are. Use Mean Squared Error (MSE) → average of squared differences between actual and predicted values.

MSE = \frac{1}{m} \sum (y - \hat{y})^2

Smaller MSE → better line.

Gradient Descent (How We Improve the Line) Gradient Descent is like walking downhill on an error curve: Start with a random line (random ). Measure error. Adjust slope and intercept step by step in the direction that reduces error. Repeat until the error is minimum → best line is found.

Why Feature Scaling? If features have very different scales (e.g., income in lakhs vs. area in sq. feet), gradient descent moves unevenly and takes longer.

By scaling (normalizing or standardizing), all features contribute fairly - faster convergence.

Linear Regression : Draws the best line to predict future values. Error Function: Tells how far the line is from actual data. Gradient Descent: Adjusts the line step by step until it’s the best fit. Feature Scaling:Makes gradient descent work faster and smoother.

So, linear regression finds the best line, gradient descent is the method used to find it, and feature scaling helps the method work efficiently.

TASK 6: THE MATRIX PUZZLE-DECODE THE NUMPY & REVEAL THE IMAGE

First I familiarized myself with numpy codes and commands,later tried analizing the task and went ahead to decode the puzzle. MATRIX PUZZLE

TASK 7: CREATE A PORTFOLIO WEBPAGE

Understood the HTML syntax and proceeded to make my a portfolio webpage with my basic information. https://github.com/NIDARSHANA21/MY-GIT-ACTIONS/blob/main/portfolio

TASK 8: WRITING RESOURCE ARTICLE USING MARKDOWN

MARKDOWN is an eaSy and Lightweight markup language that's used for editing and designing the documents using it's beautiful hastag heading features and other font designs, giving it a professional look. Given below is my resource article.

https://hub.uvcemarvel.in/article/5c7be1f7-d5df-4969-937f-bbe2493d345c

TASK 9: TINKERCAD

TINERCAD is an useful and beginner friendly browser based software that helps in building circuits,run code giving a virtual visualization of circuit output when performed manually. Connections: SERVO MOTOR---------------ARDUINO UNO

PIN 11---------------------SIGNAL

GND---------------------GND

5V---------------------POWER

ULTRASONIC SENSOR---------------ARDUINO UNO

PIN 10---------------TRIG

PIN 9---------------ECHO

GND-----------------GND

5V----------------POWER

CODE BLOCK

VIDEO OF SIMULATION OF RADAR SYSTEM (The rotation of servo motor when the object is moved within the rage of the system)

TASK 10: SPEED CONTROL OF DC MOTOR

Uderstood the control techniques of DC Motor in general and also ny using PMW Signal,working of L298N driver and the components present to help drive the DC motor accordingly.Using ARDUINO UNO AND L298N Driver i made the connections and made the motor run as shown below.

TASK 12 : SOLDERING PREREQUISITES

Learnt about soldering iron and soldering flux followed by learning the tools and the procedure to carry out soldering. Picture below displays my soldering work with components-LED and Resistor.

TASK 13 : 555 ASTABLE MULTIVIBRATOR

First i understood the 555 IC, its working and circuit connection. then calculated the the ratio of resistances to obtain 60% duty cycle. The components: R1=20K ohm, R2=10K ohm, C1 and C2=0.01 micro farad Connections and OUTPUT:

TASK 14: KARNAUGH MAPS AND DERIVING THE LOGIC CIRCUITS

Familiarized the working of burglar alarm and then derived the k-map

The logic behind the working of burglar is that the alarm rings only when the key is closed and the door is open,Based on that the k map is derived. The result is shown below

TASK 15:

ACTIVE PARTICIPATION IN KAGADA 20 IN POSTER TRACK PRESENTATION AND IMPETUS 2024.

TASK 16: DATA SHEET REPORT WRITING

Report on MQ-135 Gas Sensor

Introduction Gas sensors play an important role in monitoring harmful gases. The MQ-135 gas sensor is a widely used metal oxide semiconductor (MOS) sensor designed for detecting air quality. It can sense gases like ammonia (NH₃), nitrogen oxides (NOₓ), alcohol, benzene, smoke, and carbon dioxide (CO₂).

Working Principle The MQ-135 sensor works on the principle of adsorption and change in resistance. The sensor contains a tin dioxide (SnO₂) sensitive layer. In clean air, oxygen molecules adsorb onto the SnO₂ surface, capturing free electrons and increasing resistance. When target gases (like NH₃, CO₂, or benzene) are present, they react with the adsorbed oxygen, releasing electrons back into the conduction band, which decreases the resistance. The change in resistance is measured and converted into a corresponding gas concentration. Features Operating voltage: 5V Detection Range: 10 – 1000 ppm (parts per million) Preheat Time: 24–48 hours for stable readings Analog & Digital outputs

Calibration of MQ-135 Calibration is necessary because the sensor does not give gas concentration directly. Instead, it measures a ratio of resistance in gas (Rₛ) to resistance in clean air (R₀).

{frac{R_s}{R_0} vs. gas concentration (ppm)}

R₀ = resistance in clean air . Rₛ = resistance in presence of target gas.

Approximate Calibration Points (from datasheet): Ammonia (NH₃): Sensitive in the range 10 – 300 ppm. Benzene (C₆H₆): Detectable in range 10 – 1000 ppm. CO₂: Detectable in 350 – 10000 ppm. Alcohol & Smoke: Detectable in range 10 – 300 ppm. Each gas has a different curve slope → hence calibration must be gas-specific

Applications Air quality monitoring systems Indoor air purifiers Smart IoT-based pollution monitoring Industrial safety system.

Freundlich Adsorption Isotherm for MQ-135 Since the MQ-135 works based on adsorption of gases on the sensitive SnO₂ layer, its behavior can be studied using the Freundlich adsorption isotherm: \log x/m = \log K + \frac{1}{n} \log P

When plotted, a graph of log(x/m) vs log(P) gives a straight line. Sample Freundlich Graph (for MQ-135) X-axis → log P (Concentration of gas in ppm) Y-axis → log(x/m) (Adsorbed amount).

TASK 17: INTRODUCTION TO VR

Virtual reality acts as a computer based simulation that allows the users to innteract and respond with a 3D vitual world with realistic scenario.It gives an experince that immerses the users in computer generated environment and this is percieved by a specialized device or hardware such asVR HEADSET and CONTROLLERS which displays 3D images and videos that tracks the user's movements with the help of sensors. This virtual reality is mainly applied in gaming and others fields like education healthcare and training. FEATURES: INTERACTIVE : sers can interact, manipulate tye virtual environment by using motion controllers. IMMERSIVE : users once are into the VR world sense their surrounding as real and natural environment an try to play,move around and perform their tasks. REALISM : aims for realistic simulation, sounds,visual effects. COMPLETE 360 DEGREE VIEW* : allows users to view 360 degree field and help them perform actions in any direction as they wish.

DIFFERENCE BETWEEN VIRTUAL REALITY AND AUGMENTED REALITY : Both virtual and augmentd reality are immersive technology but they differ in howthey interact with the user's environment. augmented reality overlay digital information onto the real world while vitual reality creates a completely simulated environment. virtual requires a headset and powerful computing hardware while augmented is accessed through smartphones and other devices. Keytrends include AI Integration for upgraded version's experience,WebAR for broader access,growth of Metaverse.

Technology stacks are:

Evolution of VR with 5G for faster and improved performances Improved hardware and software for blending the features of VR and AR resulting in technology where the world makes use of XR(Extended reality) This developed technology will paly a key roles in the various feild related to it. Companies that belong to this Domain include : SIMBOTT, iBoson Innovation, Archimaze

TASK 18 :SAD SERVERS - LIKE " LEETCODE" FOR LINUX

A troubleshooting scenario was given- Command Line Murders. Before solving this I went through the linux commands and learnt how to use them in solving in similar scenarios. I made use of two clues to solve the case.

DOMAIN SPECIFIC TASK.

TASK 20: NOTEBOOK NINJA- GETTING STARTED WITH JUPYTER

Jupyter notebook is a tool using for beautifying the notebook and also help us perform coding. This makes the notebook appear neat,well organized and readable to the viewers.The link provided below includes my jupyter notebook task completed step by step.

My Jupyter Notebook

TASK 21: WATCH AND REFLECT- INTRO TO MACHINE LEARNING

INTRO TO MACHINE LEARNING