This section is still under construction. Sorry for the inconvenience
In general, OptimClassifier functions have an intuitive syntax, although some examples are shown below.
LM
First step charging the dataset and package
# Load the package
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model
linearcreditscoring <- Optim.LM(Y~., AustralianCredit, p = 0.7, seed=2018)
print(linearcreditscoring)
# Appears a warning because we don't choose threshold criteria
Warning: Thresholds' criteria not selected. The success rate is defined as the default.
3 successful models have been tested
Model rmse threshold success_rate ti_error tii_error
1 LM 0.3425880 1.6 0.8413 0.0192 0.1394231
2 SQRT.LM 0.4480197 1.2 0.8317 0.0144 0.1538462
3 LOG.LM 1.1473270 1.0 0.5961 0.4038 0.0000000
Do you want to see the coefficients of best model?
Simply, make a summary
summary(linearcreditscoring)
Call:
stats::lm(formula = ModelsTested$train_formula[[i]], data = training,
model = FALSE, x = FALSE, y = FALSE)
Residuals:
Min 1Q Median 3Q Max
-0.90688 -0.09413 0.01314 0.15060 1.07490
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 8.731e-01 9.800e-02 8.910 < 2e-16 ***
X11 1.261e-03 3.402e-02 0.037 0.970452
X2 -3.950e-04 1.521e-03 -0.260 0.795202
X3 -3.336e-03 3.297e-03 -1.012 0.312103
X42 4.622e-02 3.543e-02 1.304 0.192741
X43 4.618e-01 3.020e-01 1.529 0.126975
X52 3.150e-01 1.909e-01 1.651 0.099538 .
X53 3.183e-01 1.879e-01 1.694 0.090933 .
X54 2.765e-01 1.777e-01 1.556 0.120495
X55 3.118e-01 2.448e-01 1.273 0.203562
X56 2.742e-01 1.885e-01 1.455 0.146493
X57 3.970e-01 1.893e-01 2.098 0.036496 *
X58 3.621e-01 1.813e-01 1.998 0.046345 *
X59 3.638e-01 1.861e-01 1.955 0.051261 .
X510 4.024e-01 2.142e-01 1.879 0.060928 .
X511 3.762e-01 1.845e-01 2.039 0.042061 *
X512 3.440e-01 3.925e-01 0.876 0.381265
X513 4.934e-01 1.851e-01 2.665 0.007971 **
X514 5.414e-01 1.886e-01 2.870 0.004297 **
X62 -1.205e-01 2.641e-01 -0.456 0.648353
X63 -1.974e-01 2.726e-01 -0.724 0.469365
X64 -1.660e-01 1.749e-01 -0.949 0.343214
X65 -1.329e-01 1.839e-01 -0.723 0.470246
X67 -2.636e-02 2.293e-01 -0.115 0.908517
X68 -1.810e-01 1.765e-01 -1.026 0.305576
X69 -3.445e-01 2.545e-01 -1.353 0.176605
X7 1.208e-02 5.127e-03 2.356 0.018921 *
X81 6.131e-01 3.637e-02 16.859 < 2e-16 ***
X91 1.004e-01 3.994e-02 2.513 0.012335 *
X10 5.034e-03 3.524e-03 1.429 0.153840
X111 -4.244e-02 3.013e-02 -1.408 0.159694
X122 4.371e-02 5.719e-02 0.764 0.445102
X123 6.488e-01 1.666e-01 3.894 0.000113 ***
X13 -2.302e-04 9.932e-05 -2.317 0.020936 *
X14 4.384e-06 3.361e-06 1.304 0.192811
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.3136 on 447 degrees of freedom
Multiple R-squared: 0.6332, Adjusted R-squared: 0.6053
F-statistic: 22.7 on 34 and 447 DF, p-value: < 2.2e-16
GLM
# Load the package, if you still do not have it loaded
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model, if you had seen the LM example, this is very similar
creditscoring <- Optim.GLM(Y~., AustralianCredit, p = 0.7, seed=2018)
print(creditscoring)
# Appears a warning because we don't choose threshold criteria
Warning: Thresholds' criteria not selected. The success rate is defined as the default.
# Appear warnings generated by fitting low level function
Warning messages:
1: glm.fit: fitted probabilities numerically 0 or 1 occurred
2: glm.fit: fitted probabilities numerically 0 or 1 occurred
3: glm.fit: fitted probabilities numerically 0 or 1 occurred
# A table with all the models evaluated
7 successful models have been tested and 21 thresholds evaluated
Model rmse Threshold success_rate ti_error tii_error
1 poisson(log) 0.3389622 0.50 0.8365385 0.01442308 0.1490385
2 poisson(sqrt) 0.3409605 0.55 0.8365385 0.01442308 0.1490385
3 gaussian 0.3425880 0.60 0.8413462 0.01923077 0.1394231
4 poisson 0.3430554 0.55 0.8365385 0.01442308 0.1490385
5 binomial(cloglog) 0.3519640 0.35 0.8269231 0.01923077 0.1538462
6 binomial(probit) 0.3587217 0.45 0.8221154 0.02403846 0.1538462
7 binomial(logit) 0.3595036 0.35 0.8173077 0.01923077 0.1634615
LMM
First step charging the dataset and package
# Load the package, if you still do not have it loaded
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model, if you had seen the LM example, this is very similar
modelChooser <- Optim.LMM("Y", AustralianCredit, seed=2018)
print(modelChooser)
// TODO results for Optim.LMM example
DA
# Load the package, if you still do not have it loaded
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model, if you had seen the LM example, this is very similar
fit <- Optim.DA("Y~.", AustralianCredit,p=0.7 ,seed=2018)
print(fit)
// TODO results for Optim.DA example
CART
# Load the package, if you still do not have it loaded
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model, if you had seen the LM example, this is very similar
fit <- Optim.CART("Y~.", AustralianCredit,p=0.7 ,seed=2018)
print(fit)
// TODO results for Optim.CART example
NN
# Load the package, if you still do not have it loaded
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model, if you had seen the LM example, this is very similar
fit <- Optim.NN("Y~.", AustralianCredit,p=0.7 ,seed=2018)
print(fit)
// TODO results for Optim.NN example
SVM
# Load the package, if you still do not have it loaded
library(OptimClassifier)
# Load the dataset, AustralianCredit in this example
data("AustralianCredit")
# Let's go with the model, if you had seen the LM example, this is very similar
fit <- Optim.SVM("Y~.", AustralianCredit,p=0.7 ,seed=2018)
print(fit)
// TODO results for Optim.SVM example