Build Pulse

Bar Chart with maintenance cost
Pie Chart Asset Type and Number Of Repairs
Line Chart Number of Repairs categorized by manufacturer and asset type
Radar Chart
Data Prediction and Anamoly Detection
Instance Creation In Google Cloud
Database creation in pgAdmin
Database Query in Google shell
Code of Connecting to Cloud and Creation of API
Successful creation of API and retrieval of required data from Google Cloud

Build Pulse 🏢

Inspiration

🚀 Our goal is to provide property managers with top-end visualizations, allowing them to gain meaningful insights from their data.

What it does

Build Pulse is a property maintenance software with three powerful modules:

Visualization Module 📊
- Create customizable visualizations with ease.
- Choose from various chart types like pie, line, or bar charts.
- Select the time duration to analyze data.
- Filter by manufacturer or specific asset for tailored insights.
Prediction Analysis Module 🧠
- Leverage machine learning to predict future repairs.
- Input manufacturer, room, and floor data to get expected repair estimates.
Ticket Status Module 🎫
- Get real-time updates on all open maintenance tickets.

How we built it

🔧 We built Build Pulse using the following technologies:

Frontend: React and Recharts for visualizations.
Backend: Python for machine learning models.
Database: Google Cloud PostgreSQL.
Model Development: Jupyter Notebook.

Challenges we ran into

🤯 Integrating the frontend with the backend was a significant challenge.

Accomplishments that we're proud of

🎉 We're proud of successfully completing the entire hackathon as a team. It was the first in-person hackathon experience for all of us, and it pushed us to our limits.

What we learned

🧠 Throughout this project, we learned the importance of teamwork and collaboration. We also gained valuable experience with new technologies, including Google Cloud and Recharts.

What's next for Build Pulse

🚀 The journey doesn't end here! We plan to enhance Build Pulse further by adding features like data export downloading options, and advanced predictive analytics.

Built With

css
flask
google-cloud-postgresql
html
javascript
jupyter
keras
opencv
pgadmin
python
randomforest
recat
recharts
restapi
tensorflow

Submitted to

HackSMU V

Created by

I worked on the frontend using ReactJS as well as ReCharts to provide users with a user friendly way of creating the visualization they want.

Prabhleen Saini
I worked on Predictive Analysis and build a sutiable model to predict the no of repairs that might occur based on manufacturer , floor, room as inputs for the five types of assets. I worked on the given fictional dataset. based on analysis i found that the data is not coreleated and othere anomylies so i worked on the data and generated random values based on the analysis on the given data. and corelated the data without harming the classes. Once i was satisfied with the corelation matrix. I have rand the data on diffrent ml models and came with the conclusion that the Random Forest Classifier gave me the highest accuracy of 50 percent. and saved the checkpoint model and pushed it to api to display on the website

Pratyush Varma
I worked on the database - google cloud postgresql. Added dummy dataset(~1000) to train the ML model accurately, normalized and created database. It was my first time using google cloud, had lot of challenges to connect it locally with postgresql local server - pgAdmin, once that was achieved we also had difficulties integrating flask. Fortunately, I was able to figure out the connection setup of flask in the last 5 hours and we successfully managed to create required APIs.

Tweensi Jasani
I worked on building the back end of the project while building it I faced lots of challenges, one of the biggest challenges was connecting to Google Cloud using PostgreSQL and building API's for the front end. It took lot of time to resolve as i had to explore many technologies and read the documentation. I have used technologies like Python, Postman, Rest API, Flask, and Sql. Later, we all resolved the issue together and worked very hard to build the project.

venkat biyyapu

Updates

Prabhleen Saini started this project — Sep 17, 2023 12:32 PM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.