issues
search
amritbhanu
/
EDM591_Hyperparameter
Hyper Parameter Optimization for Education Mining
0
stars
1
forks
source link
Results to see the performance improvement
#17
Open
amritbhanu
opened
6 years ago
amritbhanu
commented
6 years ago
For Classification
Conclusion: We see dramatic improvement for SVM classification especially for min-max preprocessing ones.
For other learners we will need to plot them side by side to see the comparison. Pick 1 preprocessed dataset.
Untuned:
Tuned:
amritbhanu
commented
6 years ago
For Regression
Conclusion: We see dramatic improvement for SVM regression especially for min-max preprocessing ones.
For other learners we will need to plot them side by side to see the comparison. Pick 1 preprocessed dataset.
Untuned:
Tuned:
For Classification
Untuned:
Tuned: