- I tried to convert the non-numeric data into categorical one-hot vector data where needed
model.fit()
has been unable to accept such data for now- So I am currently just settling with
sklearn
'sLabelEncoder
and converting categorical data into numbers instead of one-hot versions
- Increase test accuracy --> improve model
- Data processing:
convert_to_categorical
- Create a passenger who would likely survive or die on Titanic
- Maybe, based on user characteristics, predict if user would survive the titanic