ECFP with SVM
Having good fun with clustering analysis, gonna move to doing some machine learning.
Again, so simple to do so with KNIME.
File Reader: Reads csv
Column Filter: Remove unnecessary columns
RDKit Fingerpring: Rocks!
Expand Bit Vector: Do I need to do this? I dunno
k-Means: was there from last blog post, didn't need this time
SVM Learner: SVM Learns... (Used default)
SVM Predictor: SVM Predicts...
Scorer: Scores...
Nice neatly aligned hey? coz I used this!
Auto Lay-Out!!! I didn't know this till today, just one click and kindly aligns for you. Why didn't I know this... But does cause a problem with big workflows so probaly only with simple ones for now, there should be "Fix In This Position" option for nodes too (maybe there is one already...).
Anyway, took a bit of time to learn but it's SVM, it's magical, it's strong, it will give a great result!!!!
Haa? what is this...
Would a parameter tuning work? I don't know... I think I'm not doing anything wrong. I have to see this. Maybe Random Forest next? (So added Baysian Paramter Optimization post. Works so much better now)