2. Introduction to
AI/ML
Hands-on experience
with Machine Learning
Custom Gemini Model -
Making an AI Chatbot
Dive into the Google
Colab Environment
ML Template
Coding
AI/ML Integration
with Web Apps
Google AI Studio
Overview
Agenda
4. What are LLMs and
Generative AI?
• A large language model (LLM) is a
language model notable for its ability to
achieve general-purpose language
generation and understanding.
• LLMs take a text prompt, from a user or
another program to generate text.
5. What is Transformer
Architecture?
• A transformer architecture is a type of
neural network architecture that
changes an input sequence into an
output sequence.
• It consists of two parts:
o An Encoder and
o A Decoder
7. Machine Learning
• Machine Learning is a way to teach computers to
learn from experience and improve their
performance over time and make predictions.
• Machine learning systems first takes input to learn
and then make useful predictions and decision
about unseen data.
11. Classification
• Primary objective of a classification
model in machine learning is to
categorize or label input data into
predefined classes or categories.
• Some Classification Algorithms :
Logistic Regression, KNN, Naïve Bayes,
SVM, etc.
12. Regression
• Regression analysis is used when
the goal is to understand the
relationship between dependent
and independent variables.
• Some Regression Algorithms :
Linear regression, Decision Tree,
Random Forest regression, etc.
13. Clustering
• Primary objective of clustering is to
gather similar points together, forming
distinct clusters.
• Some Clustering Algorithms :
K-means, Hierarchical clustering,
DBSCAN, etc.
14. Type of ML model should be used ?
2. For suggesting
spelling corrections
1. For Labelling Email as
spam or not-spam
3. For Estimating the
arrival of bus, based on
the time and traffic.
16. Various Python Libraries for ML
• NumPy: Numerical computing and efficient mathematical functions
• Pandas: Data manipulation and analysis, providing data structures
like Data Frames.
• Matplotlib: 2D plotting library for creating static, animated, and
interactive visualizations.
• Scikit-learn: For building and evaluating predictive models.
• TensorFlow: For building and training deep learning models.
• Pytorch: Deep learning library with a dynamic computational graph.