Welcome to the AI & Data Science Workshop! This workshop is designed to introduce undergraduate students to the exciting fields of Artificial Intelligence and Data Science. Whether you are a beginner or already have experience with AI, this workshop will provide valuable insights and hands-on practice to help you advance in your learning journey.
- Start Date:
- Batch II: September 9th, 2024(Bhadra 24, 2081 B.S.)
- Batch III: September 15th, 2024(Bhadra 30, 2081 B.S.)
- Duration: 15 days
- Sessions: 2 hours per day
- Total Hours: 30 hours
- Mode: Onsite/Online
- Assignments: Daily assignments will be provided via GitHub Classroom
- Goal: Set the stage, introduce AI & Data Science concepts.
- Topics:
- Overview of AI and Data Science.
- Real-world applications.
- Python programming basics.
- Activity: Icebreaker session; selecting/setting up Python environments.
- Goal: Ensure foundational Python skills.
- Topics:
- Variable, data type, function, and control structures.
- Basic input/output and file handling in Python.
- Activity: Python exercises for beginners and advanced level participants.
- Goal: Introduction to data manipulation using Python.
- Topics:
- Introduction to Pandas and NumPy.
- Reading, writing and analyzing data with Pandas.
- Basic statistical operations.
- Activity: Hands-on with CSV data (loading, manipulating, visualizing).
- Goal: Understanding how to visualize data.
- Topics:
- Introduction to Matplotlib and Seaborn.
- Creating bar charts, line graphs, scatter plots, and histograms.
- Customizing plots and analyzing trends.
- Activity: Create visualizations from a dataset.
- Goal: Introduce basic machine learning concepts.
- Topics:
- Overview of supervised vs unsupervised learning.
- Introduction to Scikit-learn.
- Linear Regression and k-Nearest Neighbors (k-NN).
- Activity: Implement a simple Linear Regression model.
- Goal: Dive deeper into supervised learning algorithms.
- Topics:
- Decision Trees, Random Forests, and Support Vector Machines (SVM).
- Model training, testing, and validation.
- Performance metrics: accuracy, precision, recall, F1-score.
- Activity: Build classifier using real-world dataset.
- Goal: Introduce unsupervised learning methods.
- Topics:
- Clustering: K-Means, Hierarchical Clustering.
- Dimensionality reduction: PCA.
- Activity: Implement K-Means clustering and visualize the results.
- Goal: Provide a foundational understanding of neural networks.
- Topics:
- Introduction to neural networks.
- Basics of deep learning and neural network architecture.
- Introduction to TensorFlow and Keras.
- Activity: Build a basic neural network using TensorFlow/Keras to classify images.
- Goal: Introduce to computer vision via image-based learning models.
- Topics:
- Architecture of Convolutional Neural Networks (CNNs).
- Applications in image classification and recognition.
- Activity: Train a simple CNN on the MNIST dataset (handwritten digits dataset).
- Goal: Introduction to NLP.
- Topics:
- Basics of text processing (tokenization, stemming, lemmatization).
- Sentiment analysis and text classification.
- Introduction to NLP libraries like NLTK and spaCy.
- Activity: Build a sentiment analysis model using a dataset of tweets or reviews.
- Goal: Learn model optimization and evaluation techniques.
- Split data into training and testing sets multiple times to evaluate the model more reliably.
- Understand the true positives, false positives, true negatives, and false negatives in model predictions.
- Learn how tuning helps prevent overfitting (when the model learns too much from training data) and underfitting (when the model doesn’t learn enough).
- Learn metrics like accuracy, precision, recall, and F1 score to assess model performance.
- Activity: Optimize a pre-trained model using hyperparameter tuning.
- Goal: Introduce cloud-based AI solutions and model deployment.
- Introduction to cloud-based AI solutions (AWS, Google Cloud AI, Azure).
- Building app with Streamlit, Flask or FastAPI.
- Online model deployment platforms like Render, Hugging Face Spaces, Gradio Hub etc.
- Activity: Deploy a simple AI model on a cloud platform.
- Goal: Discuss ethical considerations and the impact of AI.
- Topics:
- Bias in AI, ethical AI, and transparency.
- AI governance, fairness, and privacy issues.
- How AI is transforming industries (healthcare, finance, education).
- Activity: Group discussion or debate on ethical issues in AI.
- Goal: Provide support and guidance for participants working on real world projects.
- Topics:
- Review the concepts learned throughout the workshop.
- Group division and role assignment
- Mentorship and troubleshooting for project work.
- Activity: Group project work (participants apply AI techniques to real-world problems).
- Goal: Wrap up the workshop and project presentation.
- Topics:
- Project presentation by groups or individuals.
- Review key takeaways from the workshop.
- Q&A session and feedback.
- Activity: Present final projects(participants demonstrate the models they built and explain their methodology) and receive feedback.
Assignments will be provided daily through GitHub Classroom. Each day, students are expected to complete the tasks related to the day's content. The assignments will range from basic Python coding exercises to more advanced AI model implementations.
- Assignment Submission: All submissions must be made via GitHub Classroom.
- Due Time: Assignments must be submitted by 11:59 PM on the day they are assigned.
All workshop materials, including presentations, datasets, and sample code, will be made available in this repository. Please clone this repository to your local machine to follow along with the workshop and access the resources.
- Python: Download Python
- Jupyter Notebooks: Install Jupyter
- IDE: VS Code or PyCharm
- Google Colab: (https://colab.research.google.com/)
Ensure you have the following Python libraries installed:
pip install numpy pandas matplotlib seaborn scikit-learn tensorflow keras nltk