Development > Data Science
21h 51m
£14.99 Free
Python Data Science & AI: Build Real-World Apps with LLMs
What you will learn:
- Master Python programming and essential data science libraries.
- Learn advanced data manipulation, cleaning, and visualization techniques.
- Gain a comprehensive understanding of statistical concepts for data analysis.
- Develop expertise in supervised and unsupervised machine learning algorithms.
- Build and deploy real-world machine learning applications with Streamlit.
- Master the use of local LLMs and build AI applications using Ollama.
- Create a portfolio of practical projects to showcase your skills.
- Understand and apply advanced feature engineering methods.
- Gain experience with various model evaluation and comparison methods.
- Develop deployable AI applications that leverage local LLMs.
Description
Become a proficient data scientist and AI developer with our comprehensive course, "Python Data Science & AI: Real-World Applications." This beginner-friendly program progresses to an advanced level, equipping you to create impactful AI solutions. You'll master Python fundamentals and essential libraries (NumPy, Pandas) before tackling data manipulation, analysis, and visualization.
We delve into core statistical concepts crucial for machine learning, covering both supervised (regression, classification) and unsupervised learning (clustering, dimensionality reduction). The real-world application focus is key. You'll build nine complete projects simulating industry challenges:
- Automate business workflows using Pandas.
- Analyze large datasets obtained from Google Apps.
- Design a movie recommendation engine using Non-negative Matrix Factorization.
- Construct and evaluate predictive models with cutting-edge techniques.
- Develop and deploy a credit risk prediction application using XGBoost and Streamlit.
- Create LLM-powered AI apps utilizing Ollama, LangChain, and Streamlit (cloud-free).
- Implement local Python libraries for seamless AI interactions.
- Build an AI Code Assistant to boost your coding efficiency.
- Develop a RAG (Retrieval Augmented Generation)-based AI research application.
This course uniquely emphasizes local LLMs and AI automation. We explore innovative frameworks like Ollama, allowing you to work with models via Web UI, LM Studio, and more, without dependence on costly cloud APIs. Each module integrates theory, hands-on coding, and practical exercises, resulting in a portfolio of working applications to showcase your new skills. Prepare for your future career in data science and AI!
By the course's end, you'll be capable of:
- Confidently coding in Python.
- Effectively analyzing, visualizing, and modeling data.
- Implementing machine learning algorithms from the ground up.
- Constructing impactful, real-world projects.
- Developing and deploying AI applications leveraging local LLMs and tools such as Ollama and Streamlit.
Ideal for aspiring data scientists, developers, and AI enthusiasts ready to create high-impact solutions. Transition into tech, enhance your current skills, or break into AI development – this is the only course you need.
Curriculum
Introduction
This introductory section lays the groundwork for your journey into data science and AI. You'll begin by understanding the course overview and then dive into essential Python data types and structures, including a detailed explanation of Python's core data types like strings, lists, tuples, sets, and dictionaries. The section concludes with a comprehensive review of control flow mechanisms, including if-else statements, for and while loops, and best practices for efficient coding. You'll learn about fundamental concepts of Python functions, error handling, files, modules, and object-oriented programming (classes, inheritance, polymorphism, and decorators).
Python Refresher
This section serves as a refresher and deep dive into Python programming, covering a comprehensive range of topics. The lectures will thoroughly explain string manipulation techniques, lists, tuples, sets, dictionaries, control flows (if, for, while loops), functions (including advanced concepts like decorators, lambda functions, and memoization), error handling, file and module management, and an in-depth exploration of object-oriented programming concepts in Python, including classes, inheritance, encapsulation, polymorphism, and the use of decorators. This provides a solid foundation for the more advanced material in later sections.
Python NumPy Library
This section introduces the NumPy library, a fundamental tool for numerical computation in Python. The lectures delve into core aspects such as array creation, manipulation, calculations, aggregations, reshaping, transposing, and comparisons. You'll also learn about effective techniques for image processing using NumPy. This section provides a crucial foundation for handling numerical data effectively in data science and machine learning projects.
Python Pandas Library
This section covers the Pandas library, a powerful tool for data manipulation and analysis. It begins by showing how to install Jupyter Lab and Pandas. Learn how to connect to a PostgreSQL database, fetch data using `fetchmany` and `fetchall`, execute queries using Pandas, use various Pandas methods and functions for data analysis and visualization, and handle data scraping from websites. The section concludes with a demonstration of how to save scraped data to a database and covers essential Pandas methods commonly encountered in interviews and case studies.
Project 1: Pandas and Business Automation
This project provides hands-on experience in automating business workflows using Pandas. You'll apply the skills and knowledge learned in previous sections to solve a practical problem, gaining valuable experience in data manipulation, automation, and problem-solving.
Accessing, Manipulating & Filtering DataFrames
This section focuses on advanced techniques for working with Pandas DataFrames. Learn how to efficiently manipulate, access, aggregate, summarize, create new columns, drop unnecessary ones, and perform various data manipulation tasks. Key techniques for understanding and describing your data are covered in detail, including effective filtering methods.
Data Visualization
This section covers fundamental data visualization techniques in Python, focusing on effectively communicating insights from data. Learn about different plot types such as histograms and other relevant charts to analyze financial data and how to determine the most appropriate plot for various datasets.
Statistics for Data Science
This section builds a strong foundation in statistics essential for data science and machine learning. You'll learn about data visualization methods (bar charts, line charts, pie charts, histograms, box plots), measures of central tendency and spread, summary statistics, outliers, correlations, heatmaps, covariance, and hypothesis testing (t-tests, chi-square tests). A key component of this section is an in-depth exploration of linear and multiple linear regression using the statsmodels library, covering topics such as OLS models, residual analysis, coefficients, p-values, and R-squared.
Project 2: Google App Data Analysis
This project uses real-world Google App Store data. You will perform exploratory data analysis (EDA), data cleaning, preprocessing, data visualization, statistical analysis, hypothesis testing, and data storytelling. The assignment provides further practice in applying statistical knowledge to a real dataset.
Introduction to ML & Supervised Learning
This section provides an introduction to machine learning using scikit-learn and a focus on the supervised learning workflow, which involves training, testing, and evaluating machine learning models. This foundation will support your understanding of the various algorithms covered later in the curriculum.
Regression Supervised ML Algorithm
This section covers regression algorithms, a fundamental component of supervised machine learning. You'll learn about linear regression, regression performance metrics, cross-validation techniques, and regularized regression methods.
Binary Classification & Multiclass ML Supervised Classifiers
This section dives into binary and multiclass classification algorithms, also crucial for supervised machine learning. It covers k-Nearest Neighbors, Random Forest, Logistic Regression, and hyperparameter tuning using GridSearchCV and RandomizedSearchCV.
Feature Engineering for ML Supervised Learning Algorithms
This section teaches you how to engineer features, a vital step in improving the performance of machine learning models. Learn to preprocess data, handle missing values, and create dummy variables for categorical data, all while building machine learning pipelines.
How to Evaluate Multiple Models
This section demonstrates how to evaluate and compare the performance of multiple machine learning models using various metrics. Understand how to select the best model for a given task.
Advanced Topics regarding ML Supervised Learning Algorithms
This section dives into more advanced aspects of supervised learning, including a detailed exploration of linear classifiers, decision boundaries, loss functions, logistic regression regularization, probabilities, and multi-class logistic regression.
Clustering ML Unsupervised Learning Algorithms
This section introduces unsupervised learning, focusing on clustering algorithms. Learn about k-means clustering, hierarchical clustering, and techniques to evaluate and optimize clustering results.
t-SNE for 2-dimensional maps
This section focuses on using t-SNE for dimensionality reduction and visualization, providing tools for creating 2-dimensional maps of high-dimensional data.
PCA ML Unsupervised Learning Algorithm
This section covers Principal Component Analysis (PCA), another important technique for dimensionality reduction. Learn how to perform PCA, visualize the transformation, and calculate cumulative explained variance.
Project 3: Building a Recommender System using NMF
This project challenges you to create a recommender system, putting your knowledge of unsupervised learning and matrix factorization into practice.
Project 4: Predictive Modeling
This project focuses on building predictive models using machine learning. You'll work with a real-world dataset, applying data exploration, preprocessing, model selection, training, hyperparameter tuning, and model deployment to create a robust solution. This section also has an assignment for further practice.
Project 6: Credit Risk Prediction with XGBoost and Streamlit App
This project provides a complete walkthrough on creating a credit risk prediction application using XGBoost and the Streamlit framework, allowing you to deploy your machine learning model as a user-friendly web application. Detailed instructions will guide you through each step.
Ollama and Local LLMs
This section introduces you to the exciting world of Large Language Models (LLMs) and the Ollama framework. Learn how to install and run Ollama, explore available models, customize settings, interact with text and vision models, and use the Ollama REST API.
Different Apps to interact with Ollama: Msty, Open Web UI, LM Studio & Streamlit
This section shows how to integrate Ollama models with various user interfaces for seamless interaction, including Msty, Open Web UI, LM Studio, and Streamlit, opening up many possibilities for development and interaction.
Ollama Python Library
This section delves into using Ollama's Python library and REST API to integrate LLMs into your Python projects, showing how to work with Ollama in a coding environment.
Project 7: Building a Real-World Affordable AI App: RAG App
This project guides you through building a Retrieval Augmented Generation (RAG) application using Ollama, showcasing how to integrate LLMs into information retrieval systems. You'll learn about the fundamentals of RAG, data ingestion, vector embeddings, and Qdrant integration.
Project 8: Building a Real-World Affordable AI App: AI Code Assistant App
Build an AI code assistant application using Ollama. This project details building memory management, chat capabilities, and the overall coding process, resulting in a functional AI tool to assist you with code-related tasks.
Project 9: Ollama Multi-Models AI Data Science Assistant
Develop a powerful AI data science assistant using Ollama's multi-model capabilities. Learn how to integrate various models for enhanced functionality and apply your accumulated skills to create a comprehensive AI solution.
Thanks
A concluding section with bonus material and closing remarks.
Deal Source: real.discount