Neural network concepts and architectures I have learned and explored
Core Deep Learning Concepts:
A system that learns patterns from data by passing information through connected layers.
Neural Networks
A computer model inspired by the human brain used to learn from data.
Artificial Neural Networks (ANN)
Neurons process input data and weights decide how important each input is.
Neurons and Weights
Bias helps the model adjust predictions when inputs are zero.
Bias
Used to decide whether a neuron should activate or not.
Activation Functions
The process of passing input data through the network to get an output.
Forward Propagation
The process of adjusting weights by learning from prediction errors.
Backward Propagation
One full pass of training data through the model.
Epochs
The number of samples processed at one time.
Batch Size
When the model learns training data too well.
Overfitting Awareness
When the model fails to learn enough patterns.
Underfitting Awareness
Measures how wrong the model’s predictions are.
Loss Function
A method used to reduce errors by updating model weights step by step.
Gradient Descent
Types of Neural Networks:
The simplest neural network that makes basic decisions.
Perceptron
A perceptron with only one layer of weights.
Single Layer Perceptron
A neural network with multiple hidden layers to learn complex patterns.
Multi Layer Perceptron (MLP)
Used mainly for image data to detect features like edges and shapes.
Convolutional Neural Network (CNN)
Used for sequential data where past information matters.
Recurrent Neural Network (RNN)
A model where two networks compete to generate realistic data.
Generative Adversarial Network (GAN)
Activation Functions:
Outputs either 0 or 1 based on a threshold.
Step Function
Converts values into a range between 0 and 1.
Sigmoid Function
Outputs values between -1 and 1 and is centered around zero.
Tanh Function
Outputs zero for negative values and keeps positive values unchanged.
ReLU (Rectified Linear Unit)
Converts output values into probabilities for multiple classes.
Softmax Function
Machine Learning
Libraries Frameworks and Concepts I Work With
Python Libraries:
Data manipulation and analysis library for handling structured datasets using DataFrames.
NumPy
Data manipulation and analysis library for handling structured datasets using DataFrames.
Pandas
Low level plotting library used to create static, animated, and interactive visualizations.
Matplotlib
Statistical data visualization library built on Matplotlib for creating informative charts easily.
Seaborn
Machine learning library providing tools for model training, evaluation, and preprocessing.
Scikit Learn
Regression Techniques:
Predicts a value by finding a simple straight line relationship between inputs and output.
Linear Regression
Predicts values when the relationship is curved instead of straight.
Polynomial Regression
Improves predictions by keeping the model simple and avoiding over learning.
Ridge Regression
Performs feature selection and regularization by shrinking less important coefficients to zero.
Lasso Regression
Predicts values by learning decision rules from data using a tree like structure.
Decision Tree Regression
Combines multiple decision trees to improve prediction accuracy and reduce overfitting.
Random Forest Regression
Classification Techniques:
Decides between two options such as yes or no or true or false.
Logistic Regression
Classifies something by looking at the most similar nearby examples.
K Nearest Neighbors
Separates data into groups by drawing the best possible boundary.
Support Vector Machine
Classifies data using simple step by step decisions.
Decision Tree Classifier
Uses multiple decision trees together to improve classification accuracy.
Random Forest Classifier
Clustering Techniques:
Automatically groups similar data points together.
K Means Clustering
Creates groups within groups to show data relationships clearly.
Hierarchical Clustering
Model Training and Evaluation:
Tests the model multiple times to ensure it works well on new data.
Cross Validation
Balances simplicity and accuracy so the model performs well.
Bias Variance Tradeoff
Shows when a model learns too much or too little from data.
Overfitting and Underfitting
Shows how often the model makes correct predictions.
Accuracy Score
Feature Engineering and Data Preparation:
Brings all numbers to the same range so models work properly.
Feature Scaling
Keeps only the most useful information for better results.
Feature Selection
Converts text values like names or categories into numbers.
Encoding Categorical Data
Fills or fixes missing data to avoid errors.
Handling Missing Values
Finds unusual data points that may affect results.
Outlier Detection
Python Application and Framework Libraries
Libraries Frameworks and Concepts I Work With
App and UI Frameworks:
Data manipulation and analysis library for handling structured datasets using DataFrames.
Streamlit
Data manipulation and analysis library for handling structured datasets using DataFrames.
Tkinter
Automation and Utility Libraries:
Used to send and receive data from APIs and websites.
Requests
Used to interact with the operating system like files folders and environment variables.
OS
Used to work with dates and time values.
Datetime
Used to handle system level operations like command-line arguments.
Sys
Used to fetch and analyze Python package metadata directly from PyPI including versions dependencies and release details.
PyPI JSON
Used to generate random values in games and simulations.
Random
Used to manage file and folder paths in a clean and readable way.
Pathlib
A simple natural language processing library used for text processing tasks such as spelling correction sentiment analysis and basic linguistic processing.
TextBlob
Model Persistence and Utilities:
Used to save and load trained machine learning models.
Pickle
Used to efficiently store large models and speed up model loading.
Joblib
Data Handling and File Processing:
Used to read and write CSV files.
CSV
Used to work with JSON data formats commonly used in APIs.
JSON
API:
Used to build lightweight backend APIs for machine learning models.
Flask
Used to create high performance APIs for deploying ML models.
FastAPI
An ASGI server used to run FastAPI applications efficiently for development and production. Commonly used to serve machine learning models and REST APIs built with FastAPI.
uvicorn
Image and QR Utilities:
Used to generate QR codes programmatically.
qrcode
Used for image processing and displaying generated QR codes.
Pillow (PIL)
Automation and Scraping:
Used to download images and videos from public Instagram profiles.
Instaloader
Web Development
Tools technologies and concepts I use to build modern websites
Core Web Languages:
Used to add basic interactivity and dynamic behavior to web pages.
JavaScript
Used to handle server side logic in WordPress websites.
PHP
Used to structure the content and layout of web pages.
HTML
Used to style websites, including colors, layouts, and responsive design.
CSS
CMS and E Commerce Platforms:
Used to build flexible, content managed websites for businesses and portfolios.
WordPress
Used to work with enterprise level content management systems and components.
Adobe Experience Manager (AEM)
Used to create and customize e commerce websites with products and payments.
Shopify
Page Builders and Ecosystem:
Used to build dynamic layouts, forms, and advanced page features.
Elementor Pro
Used to create dynamic content, listings, filters, and custom site logic.
Jetengine|crocoblock
Used to add and manage custom data fields in WordPress.
ACF
Used to add e commerce functionality to WordPress websites.
WooCommerce
Dynamic Content and Custom Structures:
Used to organize content like services, projects, and case studies.
CPT
Used to store dynamic information such as prices, features, and dates.
Custom Fields
Used to display content automatically based on data and conditions.
Dynamic Listings
Used to organize content using categories, tags, and filters.
Taxonomies
UI Components and Site Features:
Used to ensure websites work well on mobile, tablet, and desktop.
Responsive Design
Used to create advanced navigation menus with structured content.
Mega Menus
Used for announcements, lead generation, and call to action prompts.
Popups and Modals
Used to collect inquiries, bookings, and contact details.
Forms and Lead Capture
Used to show or hide content based on rules or user roles.
Conditional Visibility
Backend and Databases:
Used to store and manage website data.
MySQL
Used to connect websites with external services and tools.
REST APIs
Performance and Optimization:
Used to reduce image sizes and improve page load speed.
Image Optimization
Used to load images and content only when needed.
Lazy Loading
Used to improve website speed and performance.
Caching Concepts
Used to improve loading speed, layout stability, and user interaction.
Core Web Vitals Awareness
SEO and Analytics:
Used to optimize content, headings, and meta information.
On Page SEO
Used to improve website structure and crawlability.
Technical SEO
Used to track website traffic and user behavior.
Google Analytics
Used to monitor indexing, search performance, and SEO issues.
Google Search Console
Used to help search engines understand website content better.
Schema and Structured Data
Security and Maintenance:
Used to keep websites updated and running smoothly.
Website Maintenance
Used to maintain compatibility and security.
Regular Updates
Used to prevent data loss and restore websites when needed.
Backup Management
Used to control access and permissions.
User Role Management
Used to reduce common WordPress security risks.
Security Hardening
Hosting Deployment and Infrastructure
Uses OpenLiteSpeed for speed, excellent for WordPress sites.
cPanel
Used to manage hosting settings, files, and databases.
CyberPanel
Used to improve website security and performance.
Cloudflare
Used for hosting and cloud related services.
Google Cloud
Used to connect domains and configure website access.
Domain and DNS Management
Used to secure websites with HTTPS.
SSL Configuration
Development and Collaboration Tools:
Used for version control of code.
Git
Used to store and manage project repositories.
GitHub
Automate, customize, and execute your software development workflows
