ML Regression using Feature Input¶
ML Regression using Feature Input is a floe that predicts property from a generic pretrained model on custom feature input. It runs a Tensorflow based fully-connected neural network Regression model which needs to be provided by user. Model can be generated using Regression Custom Input builder floe. The molecules predicted need to contain a FloatVec field of same length as the ones used for training. In addition, the floe also uses Convex Box and Monte Carlo Dropout to determine DOA. Finally, it uses Lime, a model agnostic system, to explain the predictions on the molecule. Very cheap and quick. Takes about a cent for property prediction of 20 molecules.
Name |
Description |
Type |
---|---|---|
Input Small Molecule(s) Dataset
to predict property of
|
The dataset(s) to read records from |
Molecule Dataset |
Input tensorflow Model |
Machine Learning model to predict property |
Machine Learing Tensorflow Model Dataset |
Name |
Description |
Type |
---|---|---|
Model ID of which Tensorflow model to use to predict |
Which model to select. Make sure this matches with input Model ID |
Int |
Preprocess Molecule |
Preprocess by Neutral Ph, Largest Mol, Blockbuster Filter |
Bool |
Apply Blockbuster filter |
For every molecule, stores only largest component, adjusts ionization to Neutral Ph |
Bool |
Number of features to explain |
Number of top features to provide LIME explanations for |
Int |
Name |
Description |
Type |
---|---|---|
Property Validation Field |
If the dataset has a baseline, the floe reports
a comparison between prediction in Floereport
|
Float |
Custom Feature |
Field containing feature vector to train model on.
|
FloatVec |
Name |
Description |
Type |
---|---|---|
Output Property |
Output dataset to write to |
Dataset |
Failure Property |
Output dataset to write to |
Dataset |