Artificial Intelligence using Python: Value added Course
Artificial Intelligence using Python is a 30 Hours of Hands On Value added course. This course is for FET Students only, register for the course. Course will be from 18-22 March 2024.
Note: The course will commence in March/April, 2024 . End Semester / Internal will be conducted as regular course.
Course Name: AI Using Python
MM: 100Time: 3 Hr.L T P2 0 1
Sessional: 30ESE: 70Credit : 0
Prerequisites:
Understanding ofBasic Programming Concept and Mathematics (probability and statistics).
Objectives:
Understand AI fundamentals and implement them using Python.Develop practical skills in AI through hands-on Python programming.Explore advanced AI topics and implement them using Python libraries.Apply AI techniques to real-world problems using Python. Cultivate collaboration and communication skills for AI projects using Python.
Course Coordinator
Prof. Mayank Aggarwal
Unit
Module
Course Content
No. of Hours
POs mapped
PSOsmapped
Unit-1
Module-1
Introduction to AI and PyTorch Fundamentals: Introduction to AI: (Definition, history, and applications, AI vs ML, Different types of AI: Supervised, Unsupervised, Discriminative, Generative Core Ideas), Introduction to Tensors, Creating Tensors, Accessing Data from Tensors, Manipulating Tensors, Handling Tensor Shapes, Indexing on Tensors, Interoperability between PyTorch and NumPy, Ensuring Reproducibility, Utilizing GPU for Tensor Operations.
06
PO1/PO2/PO3
PSO1/PSO2
Unit-2
Module-2
PyTorch Workflow fundamentals: Getting data ready, learning about linear regression, building a model, Fitting the model to data (training), Making predictions and evaluating a model (inference), Saving and loading a model, Putting it all together.
06
PO1/PO2/PO3
PSO1/PSO2
Unit-3
Module-3
PyTorch Neural Networks Classification: Architecture of a classification neural network, Getting binary classification data ready, Building a PyTorch classification model, Fitting the model to data (training), Making predictions and evaluating a model (inference), Improving a model (from a model perspective), Non-linearity, Replicating non-linear functions, Putting it all together with multi-class classification.
06
PO1/PO2/PO3
PSO1/PSO2
Unit-4
Module-4
Computer Vision and Custom Dataset: Model 0: Building a baseline model, Model 1: Adding non-linearity, Model 2: Convolutional Neural Network (CNN), Evaluating/Comparing our models, Transformation of Data, Model 0: TinyVGG without data augmentation, Exploring loss curves, Model 1: TinyVGG with data augmentation
06
PO1/PO2/PO3/PO4
PSO1/PSO2
Unit-5
Module-5
Transfer Learning, Going Modular, Experiment Tracking, Model Deployment: Get and customize a pretrained model, View our model’s results in TensorBoard, Creating a helper function to track experiments, View modeling experiments in TensorBoard, Learn Machine Learning Model Deployment, Ethics and concerns in AI
06
PO1/PO2/PO3/PO4
PSO1/PSO2
Total No. of Hours
30
Course Outcomes
Blooms Level
CO1
Understand AI history, applications, and types; grasp PyTorch basics including tensors, data manipulation, and GPU utilization.
Remember L1
CO2
Master PyTorch workflow from data preparation to model fitting, evaluation, and deployment.
Understand L2
CO3
Implement neural network classification in PyTorch, handling non-linearity and multi-class scenarios.
Apply L3
CO4
Apply computer vision techniques with custom datasets, including model evaluation and transfer learning.
Analyze L4
CO5
Utilize pretrained models, experiment tracking, and ethical considerations in AI deployment.
Create L5
Suggested books:
S. No.
Name of Authors /Books /Publisher/Year
1.
Artificial Intelligence with Python, Prateek Joshi., Packt Publishing Limited, 2017.
2.
Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, Aurelien Geron, O’Reilly, 2017.
3.
Deep Learning for Coders with fastai and PyTorch: AI Applications Without a PhD, Jeremy Howard and Sylvain Gugger, O’Reilly, 2020.
4.
Deep Learning with PyTorch: Build, train, and tune neural networks using Python tools, Eli Stevens, Luca Antiga and Thomas Viehmann, Manning Publishing Limited, 2020
5.
The elements of statistical learning, Friedman, Springer series in statistics, 2001.
