KLR_predict: KLR_predict

In mrecos/klrfome: Kernel Logistic Regression with Focal Mean Embeddings

KLR_predict

Description

'KLR_predict()' is a function to predict the probability of site presence to a new list of data based on the fitted alpha parameters returned from the 'KLR()' funtion.

Usage

KLR_predict(
  test_data,
  train_data,
  alphas_pred,
  sigma,
  progress = TRUE,
  dist_metric = "euclidean"
)

Arguments

test_data	- [list] Training data used to create similarity kernel matrix
train_data	- [list] Testing data to predict class
alphas_pred	- [vector] Numeric vector of alpha parameters from KLR function
sigma	- [scaler] Smoothing parameter for RBF kernel
progress	- [logical] False = no progress bar; 1 = show progress bar
dist_metric	[character] One of the distance methods from rdist::cdist. Default = "euclidean". see ?rdist::cdist

Details

This function takes a list of the 'test_data', a list of the 'train_data', a vector of the approximated alpha parameters as 'alpha_pred', a scalar value for the 'sigma' kernel hyperparameter, and a distance method (deafult = "Euclidean"). This function predicts the probability of site presence for new observations based on the training data and 'alphas' parameters. This is accomplished by building the 'k*k' kernel matrix as the similarity between the training test data then computing the inverse logit of 'k*k

Value

- [vector] - predicted probabiity of positive class

Examples

## Not run: 
sim_data <- get_sim_data(site_samples = 800, N_site_bags = 75,
sites_var1_mean = 80, sites_var1_sd   = 10,
sites_var2_mean = 5,  sites_var2_sd   = 2,
backg_var1_mean = 100,backg_var1_sd   = 20,
backg_var2_mean = 6,  backg_var2_sd   = 3)
formatted_data <- format_site_data(sim_data, N_sites=10, train_test_split=0.8,
                                   sample_fraction = 0.9, background_site_balance=1)
train_data <- formatted_data[["train_data"]]
train_presence <- formatted_data[["train_presence"]]
test_presence <- formatted_data[["test_presence"]]

##### Logistic Mean Embedding KLR Model
#### Build Kernel Matrix
K <- build_K(train_data, sigma = sigma, dist_metric = dist_metric)
#### Train
train_log_pred <- KLR(K, train_presence, lambda, 100, 0.001, verbose = 2)
#### Predict
test_log_pred <- KLR_predict(test_data, train_data, dist_metric = dist_metric,
                            train_log_pred[["alphas"]], sigma)

## End(Not run)

mrecos/klrfome documentation built on April 6, 2022, 8:02 p.m.