diff --git a/R/Estimator.R b/R/Estimator.R
index 45cccaf..44cfac7 100644
--- a/R/Estimator.R
+++ b/R/Estimator.R
@@ -320,7 +320,7 @@ gridCvDeep <- function(mappedData,
 
 
   # save torch code here
-  estimatorFile <- estimator$save(modelLocation, "DeepEstimatorModel.pt")
+  estimator$save(modelLocation, "DeepEstimatorModel.pt")
 
   return(
     list(
diff --git a/R/ResNet.R b/R/ResNet.R
index 86bb140..fec94b5 100644
--- a/R/ResNet.R
+++ b/R/ResNet.R
@@ -30,8 +30,6 @@
 #' @param hiddenFactor      How much to grow the amount of neurons in each ResLayer, default: 1:4
 #' @param residualDropout   How much dropout to apply after last linear layer in ResLayer, default: seq(0, 0.3, 0.05)
 #' @param hiddenDropout     How much dropout to apply after first linear layer in ResLayer, default: seq(0, 0.3, 0.05)
-#' @param normalization     Which type of normalization to use. Default: 'Batchnorm'
-#' @param activation        What kind of activation to use. Default: 'RelU'
 #' @param sizeEmbedding     Size of embedding layer, default: 2^(6:9) (64 to 512)
 #' @param weightDecay       Weight decay to apply, default: c(1e-6, 1e-3)
 #' @param learningRate      Learning rate to use. default: c(1e-2, 1e-5)
@@ -39,7 +37,7 @@
 #' @param hyperParamSearch  Which kind of hyperparameter search to use random sampling or exhaustive grid search. default: 'random'
 #' @param randomSample      How many random samples from hyperparameter space to use
 #' @param device            Which device to run analysis on, either 'cpu' or 'cuda', default: 'cpu'
-#' @param batchSize        Size of batch, default: 1024
+#' @param batchSize         Size of batch, default: 1024
 #' @param epochs            Number of epochs to run, default: 10
 #'
 #' @export
@@ -48,8 +46,6 @@ setResNet <- function(numLayers = c(1:8),
                       hiddenFactor = c(1:4),
                       residualDropout = c(seq(0, 0.5, 0.05)),
                       hiddenDropout = c(seq(0, 0.5, 0.05)),
-                      normalization = c("BatchNorm"),
-                      activation = c("RelU"),
                       sizeEmbedding = c(2^(6:9)),
                       weightDecay = c(1e-6, 1e-3),
                       learningRate = c(1e-2, 3e-4, 1e-5),
diff --git a/vignettes/BuildingDeepModels.Rmd b/vignettes/BuildingDeepModels.Rmd
index 360065e..8ef5b5c 100644
--- a/vignettes/BuildingDeepModels.Rmd
+++ b/vignettes/BuildingDeepModels.Rmd
@@ -281,7 +281,7 @@ paper](https://arxiv.org/abs/2106.11959).
 
 ### Example
 
-#### Set Fuction
+#### Set Function
 
 To use the package to fit a ResNet model you can use the `setResNet()`
 function to specify the hyperparameter settings for the network.
@@ -299,11 +299,6 @@ function to specify the hyperparameter settings for the network.
 `residualDropout` and`hiddenDropout` : How much dropout to apply in
 hidden layer or residual connection
 
-`normalization` : which type of normalization to use (batch
-normalization or layer normalization)
-
-`activation` : which activation function to use
-
 `sizeEmbedding` : The size of the initial embedding layer
 
 ##### Training process inputs: