Train machine learning models on training data

This function can be used to train models on protein intensity data using different machine learning algorithms

Usage

train_models(
  split_df,
  resample_method = "repeatedcv",
  resample_iterations = 10,
  num_repeats = 3,
  algorithm_list,
  seed = NULL,
  ...
)

Arguments

split_df: A split_df object from performing split_data.
resample_method: The resampling method to use. Default is "repeatedcv" for repeated cross validation. See trainControl for details on other available methods.
resample_iterations: Number of resampling iterations. Default is 10.
num_repeats: The number of complete sets of folds to compute (For resampling method = "repeatedcv" only).
algorithm_list: A list of classification or regression algorithms to use. A full list of machine learning algorithms available through the caret package can be found here: http://topepo.github.io/caret/train-models-by-tag.html. See below for default options.
seed: Numerical. Random number seed. Default is NULL
...: Additional arguments to be passed on to train function in the caret package.

Value

A list of class train for each machine-learning algorithm. See train for more information on accessing different elements of this list.

Details

In the event that algorithm_list is not provided, a default list of four classification-based machine-learning algorithms will be used for building and training models. Default algorithm_list: "svmRadial", "rf", "glm", "xgbLinear, and "naive_bayes."
Note: Models that fail to build are removed from the output.
Make sure to fix the random number seed with seed for reproducibility

References

Kuhn, Max. "Building predictive models in R using the caret package." Journal of statistical software 28 (2008): 1-26.

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, seed = 8314)

## Fit models based on the default list of machine learning (ML) algorithms
covid_model_list1 <- train_models(split_df = covid_split_df, seed = 351)
#> 
#> Running svmRadial...
#> 
#> Running rf...
#> 
#> Running glm...
#> 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: 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: 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!

## Fit models using a user-specified list of ML algorithms.
covid_model_list2 <- train_models(
  covid_split_df,
  algorithm_list = c("svmRadial", "glmboost"),
  seed = 351
)
#> 
#> Running svmRadial...
#> 
#> Running glmboost...
#> glmboost failed.
#> Done!

## Change resampling method and resampling iterations.
covid_model_list3 <- train_models(
  covid_split_df,
  resample_method = "cv",
  resample_iterations = 50,
  seed = 351
)
#> Warning: `repeats` has no meaning for this resampling method.
#> 
#> Running svmRadial...
#> Warning: There were missing values in resampled performance measures.
#> 
#> Running rf...
#> Warning: There were missing values in resampled performance measures.
#> 
#> 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: 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: fitted probabilities numerically 0 or 1 occurred
#> Warning: There were missing values in resampled performance measures.
#> 
#> Running xgbLinear...
#> Warning: There were missing values in resampled performance measures.
#> 
#> Running naive_bayes...
#> Warning: There were missing values in resampled performance measures.
#> Done!
# }