Test machine learning models on test data

This function can be used to predict test data using models generated by different machine learning algorithms

Usage

test_models(
  model_list,
  split_df,
  type = "prob",
  save_confusionmatrix = FALSE,
  file_path = NULL,
  ...
)

Arguments

model_list: A model_list object from performing train_models.
split_df: A split_df object from performing split_data.
type: Type of output. Set type as "prob" (default) to output class probabilities, and "raw" to output class predictions.
save_confusionmatrix: Logical. If TRUE, a tab-delimited text file ("Confusion_matrices.txt") with confusion matrices in the long-form data format will be saved in the directory specified by file_path. See below for more details.
file_path: A string containing the directory path to save the file.
...: Additional arguments to be passed on to predict.

Value

prediction_list: If type = "raw", a list of factors containing class predictions for each method will be returned.

Details

Setting type = "raw" is required to obtain confusion matrices.
Setting type = "prob" (default) will output a list of probabilities that can be used to generate ROC curves using roc_plot.

Author

Chathurani Ranathunge

Examples

# \donttest{
## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)
#> Total number of differentially expressed proteins (8) is less than n_top.
#> None of the proteins show high pair-wise correlation.
#> 
#> No highly correlated proteins to be removed.

## Split the data frame into training and test data sets
covid_split_df <- split_data(covid_model_df)

## Fit models using the default list of machine learning (ML) algorithms
covid_model_list <- train_models(covid_split_df)
#> 
#> Running svmRadial...
#> 
#> Running rf...
#> 
#> Running glm...
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> 
#> Running xgbLinear...
#> 
#> Running naive_bayes...
#> Done!

# Test a list of models on a test data set and output class probabilities,
covid_prob_list <- test_models(model_list = covid_model_list, split_df = covid_split_df)
#> 
#> Testing svmRadial...
#> 
#> Testing rf...
#> 
#> Testing glm...
#> 
#> Testing xgbLinear...
#> 
#> Testing naive_bayes...
#> 
#> Done!
# }

if (FALSE) {
# Save confusion matrices in the working directory and output class predictions
covid_pred_list <- test_models(
  model_list = covid_model_list,
  split_df = covid_split_df,
  type = "raw",
  save_confusionmatrix = TRUE,
  file_path = "."
)
}

Test machine learning models on test data

Usage

Arguments

Value

Details

See also

Author

Examples