-
shows accuracy levels
Inspiration
This is the HackHer's Fiserv Challenge
What it does
It detects how likely it is that someone will default on their loans based on different factors like gender, age, marital status, etc
How we built it
python and google colab
Challenges we ran into
One challenge we ran into was editing the dataset to import it into our code and differentiating between the sections that are inputs and outputs. Another challenge is choosing best activation function (relu,sigmoid), loss function (mean squared error), optimizer (Adam) from keras. Our machine learning algorithm stopped at 77% when it should not have.
Accomplishments that we're proud of
This is the first Machine Learning project that Sam has done.
What we learned
We learned what format pandas data frame wants and how to change an excel file
What's next for Fiserv Challenge
Troubleshoot to figure out why our algorithm did not move past 77% accuracy
Log in or sign up for Devpost to join the conversation.