#MyPortfolio ☯︎

profile-image

,
I'm Nischal Pradhan
Data Scientist and Machine Learning Engineer
* Student of AI and Data Science @Loyalist_College_in_Toronto

About Me

INFORMATION ABOUT ME


Hi, this is me, Nischal Pradhan. I am from Kathmandu, Nepal. The land of Himalayas. I completed my Bachelors' in Information Technology (Computing) year: 2019-2022, from Islington College, Kathmandu, Nepal. I found myself very much interested in coding in Python Programming language, as it one of the high level programming language and is world wide popular and very much powerful. This is also a reason, why I chose my future goal to be an AI Developer and ML Engineer. Hence, at the moment I am in Toronto, Canada, studying AI and Data Science, Post-Graduate year: 2024-2025 course which primarily focuses on AI developments, Data Science and Machine Learning.
The next five year plan....?


I think we have a duty to maintain the light of consciousness to make sure it continues into the future. – Elon Musk
pp-nischalpradhan

My Projects

Learning ReactJS w/ APOD API

React Application
Movie Recommendation System

ML & AI (Collaborative Filtering)
Yatayat(Bus) Management System

Python Django `FYP`
Personal Portfolio

Html - Css - Js
Image Slider

Vanilla javascript

Other Projects Lists:


  • Python
    Tic Tac Toe, URL-Shortener(Flask API)

  • Java
    Calculator, Currency-Converter, SpringBoot (Internship - Angular Developer)

  • Linux Shell Application GUI
    Interactive UNIX environment to execute i/0

  • C# Desktop Application
    Recreation Center, C# Web Application

  • More... 

Blog

Learn AI/ML Tools and Technologies


There are many tools and technologies available to help you master AI and ML skills. These tools continuously evolve, bringing new features and improvements. For data preprocessing, visualization, and model building, some essential libraries include:

• Pandas – for handling datasets
• NumPy – for numerical computations
• Matplotlib & Seaborn – for data visualization
• Scikit-Learn – for traditional ML algorithms
• TensorFlow & PyTorch – for deep learning models

If you're working on AI applications, you should also explore cloud platforms like Google Colab, AWS, and Azure for scalable model training and deployment.

Read Blog  

Understanding Machine Learning Algorithms


Machine learning models rely on mathematical and statistical techniques to learn patterns from data. Some of the key algorithms that you should master include:

• Supervised Learning: Linear Regression, Decision Trees, Random Forest, Support Vector Machines (SVM)
• Unsupervised Learning: K-Means Clustering, DBSCAN, PCA
• Deep Learning: Neural Networks, CNNs (for images), RNNs (for sequences)

Start by implementing simple models and gradually move to advanced architectures like transformers and reinforcement learning models.

Read Blog  

Researching || Problem Solving || Experimentation


AI is a rapidly evolving field, and continuous learning is necessary. Researching new techniques, reading papers, and experimenting with different approaches will enhance your skills. Platforms like ArXiv, Google Scholar, and Kaggle are great sources of knowledge.

Becoming proficient in AI requires curiosity, patience, and practice. Learn different machine learning paradigms, participate in AI hackathons, and explore real-world applications like NLP, Computer Vision, and Generative AI.

Mastering AI and Deep Learning


Mastering AI is a long-term journey. Focus on:

• Mathematical foundations: Linear Algebra, Probability, and Optimization
• Understanding architectures: CNNs for vision, LSTMs/Transformers for text
• Cloud computing: Deploying AI models on AWS, GCP, or Azure
• Ethics in AI: Ensuring fairness, transparency, and accountability in AI models

Learn, experiment, debug, and innovate—this is the cycle of AI mastery. Anyone can become an AI expert with dedication and a problem-solving mindset. 🚀

Hyperparameter Tuning & Model Optimization


A well-performing AI model is not just about choosing the right algorithm but also about fine-tuning it. Some important techniques include:

• Grid Search & Random Search – Finding the best hyperparameters
• Cross-Validation – Ensuring the model generalizes well
• Regularization (L1/L2, Dropout) – Preventing overfitting
• Early Stopping & Learning Rate Scheduling – Optimizing training time

AI is about experimentation. Every tweak in parameters can lead to significant improvements, so always analyze the model performance before finalizing it.

Exploring Deep Learning and Neural Networks


Deep Learning is the backbone of modern AI applications like self-driving cars, voice assistants, and generative AI. To master deep learning, start with:

Neural networks are inspired by the human brain. The key components include:

• Neurons & Activation Functions – ReLU, Sigmoid, Softmax
• Forward & Backpropagation – Learning through weight updates
• Loss Functions – MSE, Cross-Entropy, and Hinge Loss
• Optimizers – SGD, Adam, RMSprop

Before diving into complex models, build and train a simple feedforward neural network on datasets like • MNIST or CIFAR-10 using TensorFlow or PyTorch.


Get in  Touch