Machine Learning

Complete the form to build your model.

Loading form...

Type

Algorithm Regression algorithm used to build the model.

Algorithm Classification algorithm used to build the model.

Algorithm Clustering algorithm used to build the model.

Algorithm Dimensionality reduction algorithm used to build the model.

Degree Maximum degree of the polynomial.

Estimators Number of trees in the forest. Enter an integer between 1 and 1,000.

Max features Number of features to consider when looking for the best split. Enter an integer between 1 and the number of X variables selected.

Min leaf samples Minimum number of samples required to be at a leaf node. Enter an integer between 1 and the number of train samples.

Max tree depth Maximum depth of each tree. Enter an integer between 1 and 20.

Selection method Method used to calculate a prediction from the output of each tree.

Distribution Probability distribution used in the model.

Neighbors Number of nearest neighbors used to make a prediction. Enter an integer between 1 and the number of train samples.

Clusters Number of clusters to form.

Iterations Number of iterations used to train the model. Enter an integer between 1 and 100,000.

Activation Activation function used in each hidden layer.

Hidden layers

Layer 1 nodes

Number of hidden layers in the model.

Initialization Method used to select the initial position of each cluster centroid.

Distance Metric used to compute distance.

Seed Creates the random number generator used to build the model. Enter an integer between 1 and 1,000,000,000 for a deterministic build. Leave the input blank for a random build.

Learning rate

0.01

Rate at which the model learns the weight and bias parameters.

Regularization

0.0001

Regularization prevents overfitting the model to the data.

Tolerance

0.0001

Error tolerance for the model.

Compute intercept term

Whether to calculate the intercept for the model.

Use bagging

Whether bootstrap aggregating is used to build the trees.

Center data

Centering subtracts the mean from each data point.

Scale data

Scaling divides each data point by the standard deviation.

Variable

Build

Preview

Chart

Page of

Split samples

Train

Test

Split the samples into train and test subsets. Train samples build the model. Test samples evaluate the performance of the model.

Shuffle samples

Shuffling changes the order of the samples in a random way.

Shuffle seed Creates the random number generator used to shuffle the samples. Enter an integer between 1 and 1,000,000,000 for a deterministic shuffle. Leave the input blank for a random shuffle.

Load data to build a model.

Type:

Algorithm:

Date:

Time:

Dataset:

Header:

Rows:

Columns:

X variables:

Y variable:

Classes:

Degree:

Estimators:

Max features:

Min leaf samples:

Max tree depth:

Selection method:

Distribution:

Neighbors:

Clusters:

Iterations:

Activation:

Hidden layers:

Initialization:

Distance:

Seed:

Learning rate:

Regularization:

Tolerance:

Intercept:

Bagging:

Center data:

Scale data:

Shuffle samples:

Shuffle seed:

Power:

Y = A X ^B

Linear:

log(Y) = B log(X) + log(A)

Exponential:

Y = A e ^{B X}

Linear:

log(Y) = B X + log(A)

	coef	std err	t	P > \|t\|

	coef

DF residuals:

DF model:

F-statistic:

Prob (F-statistic):

R-squared:

Adjusted R-squared:

Residual std err:

Residuals:

Min

Median

Max

Tree #

Structure:

Tree height:

Internal nodes:

Leaf nodes:

Gain function:

Split function:

Show tree

Parent node

Current node

Height:

Depth:

Left:

= <

Right:

= ≥

Left child node

Right child node


Prior probability

Means

Standard deviations

Conditional probabilities


Bias

Weights


Cluster size
Cluster error

Iterations completed:

Converged:


Eigenvalues
Explained variance
Cumulative variance

Eigenvectors

Offsets

Scaling factors

Train samples:

OOB

Max error

MAE

MedAE

MAPE

MSE

RMSE

MSLE

Confusion matrix:

		Predicted

OOB

Accuracy

F1 Score

TPR

TNR

PPV

NPV

Out-of-bag samples:

Model

Baseline

Max error

MAE

MedAE

MAPE

MSE

RMSE

MSLE

Confusion matrix:

		Predicted
	Baseline

Model

Baseline

Accuracy

F1 Score

TPR

TNR

PPV

NPV

Test samples:

Input:

Output:

Model

Baseline

Build a model to view results.