Scribd
Upload
15 views2 pages

Complex Engineering Activity

This document outlines an engineering activity focused on building an AI-powered sentiment analysis system using open-source tools like Hugging Face and Google Colab. Students will learn to fine-tune a pre-trained transformer model to classify customer reviews and deploy it for real-time use. The activity includes prerequisites, tools, and assessment criteria to ensure a comprehensive learning experience.

Uploaded by

Sajid Hussain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views2 pages

Complex Engineering Activity

This document outlines an engineering activity focused on building an AI-powered sentiment analysis system using open-source tools like Hugging Face and Google Colab. Students will learn to fine-tune a pre-trained transformer model to classify customer reviews and deploy it for real-time use. The activity includes prerequisites, tools, and assessment criteria to ensure a comprehensive learning experience.

Uploaded by

Sajid Hussain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Complex Engineering Activity

Title: Building an AI-Powered Sentiment Analysis System with Open-Source Models and Datasets Using
Colab and Hugging Face

Objective

The goal of this activity is to enable students to understand the end-to-end process of developing an AI
solution using open-source tools and platforms. Students will fine-tune a pre-trained transformer model
for sentiment analysis using Hugging Face and deploy it for real-time inference on Colab.

Activity Outline

1. Problem Statement

Develop a sentiment analysis system that can classify customer reviews as positive, negative, or neutral.
The system should be built using pre-trained transformer models from Hugging Face and trained on an
open-source dataset.

2. Learning Outcomes

By the end of this activity, students will:

1. Understand how to use open-source datasets and models from Hugging Face.

2. Fine-tune a pre-trained transformer model.

3. Evaluate and deploy the model for real-world applications.

4. Gain hands-on experience with Google Colab and Hugging Face's ecosystem.

3. Prerequisites

• Basic understanding of Python and deep learning.

• Familiarity with Google Colab and Hugging Face.

• Installed Hugging Face libraries (transformers and datasets) and PyTorch or TensorFlow.

4. Tools and Technologies

• Google Colab: Cloud-based notebook for coding.

• Hugging Face Transformers: Library for pre-trained models.

• Hugging Face Datasets: Repository of open-source datasets.


• Open-Source Dataset: IMDb or SST2 dataset for sentiment analysis.

5: Test and Debug

• Test the model with real-world examples.

• Debug issues related to tokenization, data imbalance, or overfitting.

6. Assessment Criteria

1. Correct implementation of dataset loading and preprocessing.

2. Successful fine-tuning of the transformer model.

3. Effective evaluation and interpretation of the model's performance.

4. Deployment of the model for inference.

5. Clarity of code, documentation, and explanation of steps.

You might also like