AI Expo Project and Presentation

About

This project uses AWS SageMaker to integrate AI capabilities into our website. Our team is developing a trip planner that utilizes AI to help plan. Users will be asked questions about their trip, and the AI will compile all the information into a summary.

How to use:

This tutorial assumes students are using AWS Learner Labs, but will work with any instance of AWS Sagemaker

Setting up SageMaker

1. Log into the AWS Console and search SageMaker
2. In the navigation console on the left, select "Notebooks"
3. Select "Launch Instance"
4. Name the instance, select the type (on the lab, you are limited to medium, large, and xlarge), make sure the IAM role you are using is the Lab Role
5. Leave all the other setting at their defaults and click "Create Notebook Instance"

The instance takes a few minutes to launch, you can press the refresh button to update the status.

While you wait, you can:

Download the code

1. In GitHub, click the Code button and copy the HTTPS URL
2. Open a Terminal in the directory or folder you want the project to be in
3. Run the command git clone https://github.com/byui-cloudsociety/Trip_Planner.git

OR

1. Click the Code button and "Download ZIP"
2. Select where you want the ZIP file to be saved, and wait for the download to complete
3. Unzip the file in a directory or folder where you want the project to be in

In the MS Teams chat, there is a shared file called poiTrainingData.csv, download that file somewhere you can find, we will be using it in the next step.

Upload content to the notebook

Once the instance has started, the status should be "Ready" or "InService"

1. Select the instance, and Open JupyterLabs. A new tab should open
2. In the top left on the new tab, click the button that is labelled "Upload Files" (next to the "+ New Launcher" button)
3. Find the folder where you downloaded the GitHub files. Open "notebook". Select and upload the following files:
   • sage_tfidf.py
   • sage_word2Vec.py
   • launch.ipynb
4. Find the folder where you downloaded the training data CSV file. Select and upload that file
5. In total you should have 4 files

Running the Code

Every time you launch or re-launch the instance, you will have to train the model with the data.

1. Double-click "launch.ipynb" from the JupyterLabs file browser
2. Click "Run", and "Run all Cells"

When the code process gets to the 5th cell, there will be some user interaction.

After the initial training, you can select the 5th cell and run it alone.

Next Steps

This code is available for anyone to use and change, locally.

If you are part of the BYU-I Cloud Computing Team, talk to a member of the presidency about how you can upload changes for the whole project.

There is a way to connect the frontend part of the website to the model in SageMaker, but that is for a future project.

Video Presentation

The video presentation made for the BYU-I AI Expo will go here