CricketIQ

• 

  Homepage or Initial page

• 

  Dropdowns and options to select

• 

  Result after clicking "Predict"

• 

• 

• 

• 

• 

• 

Inspiration

Being born and brought up in India I have always heard about cricket from my family, friends everyone and the craze of cricket in India, you can't compare it. So the Idea of CricketIQ came up cause why not.

What it does

CricketIQ is an IPL match prediction and analytics dashboard. It predicts the likelihood of winning a match based on: -The teams involved -The venue -The toss winner and what decision they make(ball or chase)

How we built it

I started building the project by first collecting the historical IPL data from Kaggle. Then came the main part the cleaning and standardizing part. The ML model was trained using Scikit-learn and saved using Joblib. Then finally a Streamlit dashboard was created using Streamlit and Plotly was used for visualization.

Challenges we ran into

The biggest challenge was preparing the dataset as it has inconsistencies. Another challenge was integrating the trained model with the Streamlit application while ensuring that the same encoders used during training were also used during prediction.

What we learned

Working on CricketIQ helped me understand the complete machine learning workflow, from raw data to deployment. Some of the key skills I gained include:

• Data preprocessing and cleaning
• Training and evaluating machine learning models
• Model serialization using Joblib
• Building interactive dashboards with Streamlit
• Data visualization using Plotly
• Organizing a machine learning project into reusable modules Beyond the technical aspects, this project reinforced the importance of clean data and thoughtful feature selection in building reliable machine learning applications.

What's next for CricketIQ

There are several directions in which CricketIQ can be expanded:

• Live IPL match prediction using real-time APIs
• Player-level performance analysis
• More advanced machine learning models
• Additional match statistics and visualizations
• Cloud deployment for public access

Built With

• git
• github
• joblib
• pandas
• plotly
• python
• scikit-learn
• streamlit