Supervised classification | Morphometric analyzer - NeuroSuites

Supervised classification

1. Features selection

Target variable:

Select class variables:

No data available. If you want to upload your dataset please go here and then come back here.

Manual predictor feature selection:

No data available. If you want to upload your dataset please go here and then come back here.


Feature pre-processing:

Fill missing values (if any):

This algorithm requires a
discrete class feature


Automatic predictor feature selection:

Feature selection algorithm:

Filtering criteria:


Dimensionality reduction:

Dimensionality reduction algorithm:

When to apply:


Loading dataset...

Estimated time to finish:

Data uploaded

2. Training-validation-test sets configuration

Select the k for the k-fold cross-validation:

Select the % of instances to be in the test set:
(Set to 0% if you want to use all the data set for training)

%

Loading...

Sets created

3. Supervised learning algorithm

Select the learning algorithm:

Multi label learning method:

Selected algorithms:

Multi-label K Nearest neighbors algorithm hyperparameters



Backend:

Multi-label SVM algorithm hyperparameters



Backend:

Binary relevance algorithm hyperparameters

Classifier chain algorithm hyperparameters

Label powerset algorithm hyperparameters

RAKELd algorithm hyperparameters

k Nearest neighbors algorithm hyperparameters



Backend:

Rule induction (CN2) algorithm hyperparameters



Backend:

Decision tree algorithm hyperparameters


Criterion:

Max features:


Backend:

Random forest algorithm hyperparameters


Criterion:

Max features:


Backend:

SVM algorithm hyperparameters


Kernel:


Backend:

Neural network (Multilayer perceptron classifier) algorithm hyperparameters


Activation function:


Backend:

Linear discriminant analysis (LDA) algorithm hyperparameters



Backend:

Quadratic discriminant analysis (QDA) algorithm hyperparameters



Backend:

Logistic regression algorithm hyperparameters


Penalty:


Backend:

Naive Bayes algorithm hyperparameters



Backend:

Tree Augmented Naive Bayes algorithm hyperparameters



Backend:

Bagging metaclassifier algorithm hyperparameters


Base estimator (it must have been selected first in the left side):
(If none is selected, a decision tree will be used)


Backend:

Boosting (AdaBoost) metaclassifier algorithm hyperparameters


Base estimator (it must have been selected first in the left side):
(If none is selected, a decision tree will be used)

Algorithm:


Backend:

Stacking metaclassifier algorithm hyperparameters


Classifiers (it must have been selected first in the left side):
(If none is selected, a decision tree will be used)


Meta-classifier (it must have been selected first in the left side):
(If none is selected, logistic regression will be used)


Backend:

Running learning algorithms...

Estimated time to finish:

Learning phase done
Check the results below.

4. Performance evaluation

Training-validation set

Test set

5. Prediction in new data set

Upload the new dataset in CSV or Apache Parquet format (.parquet.gzip).

Important note: the data set must have the same predictor features as the training data set.

GsVeJwK2tgeZ1U58OVbA0taceHY2qP7uhv2VXOEdaYcbVjR1GFjaSrvSFFcR1IXi

Loading

Upload complete.


Download predictions

Select the learned algorithms:

Error downloading the predictions. Try again later.