saipriya0209 / Project_Diabetes_Clinical_Text_Analysis_QBS101_5

0 stars 0 forks source link

Question regarding model types/classifiers #5

Closed fmalzy closed 1 year ago

fmalzy commented 1 year ago

It appears that you have employed various model types/classifiers (SVC, RandomForest, KNN, DecisionTree) for different disorders (such as diabetes and asthma). However, frankly speaking I find it somewhat confusing as to why you opted for these specific choices for each disorder...Also, since there are some redundancies in the code (data_model), one potential optimization suggestion in the future (given our current time constraints) is to implement a loop and define a training function that encompasses all disorders :)

saipriya0209 commented 1 year ago

thank you for the suggestion! I have a grid search for each disorder in the XG Boost model as my main analysis. The redundant models are just me comparing the performances of many models