Inspiration
Data is very huge in now days and the use of data rapidly evolved field . Building something that helps community and put positive impact on them is a great success and it is actually a good thing . When i started learning and studied that how we use past data and build applications , models and automate our daily tasks in every field so I started working on it and build models , applications .
What it does
*Diabetes Prediction model is based on Machine Learning model that build on past data . After training on huge amount of data and securing a valuable evaluation matrices results we deployed it . In real world scenarios it takes data from person who want to know he/she is falling into diabetic or not . Based on features that requires putting into model and get results . *
How we built it
** We need to have data, resources that allow you to use their data take data from that . After doing all data analysis and data engineering steps time comes to use algorithm . Try Machine Learning algorithms as well as Deep Learning neural network architecture and get results . After that decides which of the algorithm works good and make predictions accurately with high accuracy if data is balanced and perform good by checking other evaluation metrics .**
Challenges we ran into
** Yah we faced challenges, First is that finding useful data gathering data that should useful for model that we use for predictions . and Secondly choosing right model consider all challenges like, computational resources , best performance . **
Accomplishments that we're proud of
** When we help community by resolving their problems is actual accomplishment . Now in AI era we can automate our tasks and make more productive peoples **
What we learned
** We learned lot of things while gathering data and building algorithms **
What's next for Diabetes Prediction
** There are some projects , I am working on it some NLP tasks , LLMS as well**
Built With
- ipython
- joblib
- matplotlib
- numpy
- python
- scikit-learn
- scipy
- seaborn
Log in or sign up for Devpost to join the conversation.