diff --git a/src/functions/loss_functions.rs b/src/functions/loss_functions.rs
index a5d8a94..383ae21 100644
--- a/src/functions/loss_functions.rs
+++ b/src/functions/loss_functions.rs
@@ -1,4 +1,4 @@
-use ndarray::{Array1, ArrayView1};
+use ndarray::Array1;
 
 pub enum Type {
     MSE,
@@ -6,8 +6,8 @@ pub enum Type {
 }
 
 type LossFuncTuple = (
-    fn(ArrayView1<f64>, ArrayView1<f64>) -> f64,
-    fn(ArrayView1<f64>, ArrayView1<f64>) -> Array1<f64>,
+    fn(&Array1<f64>, &Array1<f64>) -> f64,
+    fn(&Array1<f64>, &Array1<f64>) -> Array1<f64>,
 );
 
 pub fn parse_type(t: Type) -> LossFuncTuple {
@@ -17,23 +17,23 @@ pub fn parse_type(t: Type) -> LossFuncTuple {
     }
 }
 
-pub fn mse(y_true: ArrayView1<f64>, y_pred: ArrayView1<f64>) -> f64 {
-    let mut temp = &y_true - &y_pred;
+pub fn mse(y_true: &Array1<f64>, y_pred: &Array1<f64>) -> f64 {
+    let mut temp = y_true - y_pred;
     temp.mapv_inplace(|x| x * x);
     let mut sum = 0.0;
-    for i in 0..temp.len() {
-        sum += temp.get(i).unwrap();
+    for entry in temp.iter() {
+        sum += entry;
     }
     sum / temp.len() as f64
 }
 
-pub fn mse_prime(y_true: ArrayView1<f64>, y_pred: ArrayView1<f64>) -> Array1<f64> {
-    let temp = &y_true - &y_pred;
+pub fn mse_prime(y_true: &Array1<f64>, y_pred: &Array1<f64>) -> Array1<f64> {
+    let temp = y_true - y_pred;
     temp / (y_true.len() as f64 / 2.0)
 }
 
-pub fn mae(y_true: ArrayView1<f64>, y_pred: ArrayView1<f64>) -> f64 {
-    let temp = &y_true - &y_pred;
+pub fn mae(y_true: &Array1<f64>, y_pred: &Array1<f64>) -> f64 {
+    let temp = y_true - y_pred;
     let mut sum = 0.0;
     for i in 0..temp.len() {
         sum += temp.get(i).unwrap().abs();
@@ -41,7 +41,7 @@ pub fn mae(y_true: ArrayView1<f64>, y_pred: ArrayView1<f64>) -> f64 {
     sum / temp.len() as f64
 }
 
-pub fn mae_prime(y_true: ArrayView1<f64>, y_pred: ArrayView1<f64>) -> Array1<f64> {
+pub fn mae_prime(y_true: &Array1<f64>, y_pred: &Array1<f64>) -> Array1<f64> {
     let mut result = Array1::zeros(y_true.raw_dim());
     for i in 0..result.len() {
         if y_true.get(i).unwrap() < y_pred.get(i).unwrap() {
diff --git a/src/layers/activation_layer.rs b/src/layers/activation_layer.rs
index 47a0c03..48e2502 100644
--- a/src/layers/activation_layer.rs
+++ b/src/layers/activation_layer.rs
@@ -1,4 +1,4 @@
-use ndarray::{arr1, Array1, ArrayView1};
+use ndarray::{arr1, Array1};
 
 use super::Layer;
 use crate::functions::activation_functions::*;
@@ -23,15 +23,15 @@ impl ActivationLayer {
 }
 
 impl Layer for ActivationLayer {
-    fn forward_pass(&mut self, input: ArrayView1<f64>) -> Array1<f64> {
-        self.input = input.to_owned();
+    fn forward_pass(&mut self, input: Array1<f64>) -> Array1<f64> {
+        self.input = input;
         // output isn't needed elsewhere
         // self.output = (self.activation)(&self.input);
         // self.output.clone()
         (self.activation)(&self.input)
     }
 
-    fn backward_pass(&mut self, output_error: ArrayView1<f64>, _learning_rate: f64) -> Array1<f64> {
+    fn backward_pass(&mut self, output_error: Array1<f64>, _learning_rate: f64) -> Array1<f64> {
         (self.activation_prime)(&self.input) * output_error
     }
 }
diff --git a/src/layers/fc_layer.rs b/src/layers/fc_layer.rs
index af0330d..8f4da68 100644
--- a/src/layers/fc_layer.rs
+++ b/src/layers/fc_layer.rs
@@ -1,6 +1,6 @@
 extern crate ndarray;
 
-use ndarray::{arr1, arr2, Array, Array1, Array2, ArrayView1, ShapeBuilder};
+use ndarray::{arr1, arr2, Array, Array1, Array2, ShapeBuilder};
 use ndarray_rand::rand_distr::{Normal, Uniform};
 use ndarray_rand::RandomExt;
 
@@ -71,30 +71,25 @@ impl FCLayer {
 }
 
 impl Layer for FCLayer {
-    fn forward_pass(&mut self, input: ArrayView1<f64>) -> Array1<f64> {
+    fn forward_pass(&mut self, input: Array1<f64>) -> Array1<f64> {
         if !self.is_initialized {
             self.initialize(input.len());
         }
 
-        self.input = input.to_owned();
+        self.input = input;
         // output isn't needed elsewhere
         // self.output = self.input.dot(&self.weights) + &self.biases;
         // self.output.clone()
         self.input.dot(&self.weights) + &self.biases
     }
 
-    fn backward_pass(&mut self, output_error: ArrayView1<f64>, learning_rate: f64) -> Array1<f64> {
+    fn backward_pass(&mut self, output_error: Array1<f64>, learning_rate: f64) -> Array1<f64> {
         let input_error = output_error.dot(&self.weights.t());
         let delta_weights = self
             .input
-            .to_owned()
-            .into_shape((self.input.len(), 1usize))
+            .to_shape((self.input.len(), 1usize))
             .unwrap()
-            .dot(
-                &output_error
-                    .into_shape((1usize, output_error.len()))
-                    .unwrap(),
-            );
+            .dot(&output_error.to_shape((1usize, output_error.len())).unwrap());
         self.weights = &self.weights + learning_rate * &delta_weights;
         self.biases = &self.biases + learning_rate * &output_error;
         input_error
diff --git a/src/layers/mod.rs b/src/layers/mod.rs
index 53c3eee..cbcfb28 100644
--- a/src/layers/mod.rs
+++ b/src/layers/mod.rs
@@ -1,9 +1,9 @@
-use ndarray::{Array1, ArrayView1};
+use ndarray::Array1;
 
 pub mod activation_layer;
 pub mod fc_layer;
 
 pub trait Layer {
-    fn forward_pass(&mut self, input: ArrayView1<f64>) -> Array1<f64>;
-    fn backward_pass(&mut self, output_error: ArrayView1<f64>, learning_rate: f64) -> Array1<f64>;
+    fn forward_pass(&mut self, input: Array1<f64>) -> Array1<f64>;
+    fn backward_pass(&mut self, output_error: Array1<f64>, learning_rate: f64) -> Array1<f64>;
 }
diff --git a/src/lib.rs b/src/lib.rs
index bbff9d5..2b3f2af 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -3,12 +3,12 @@ pub mod layers;
 
 use functions::loss_functions::{self, parse_type};
 use layers::*;
-use ndarray::{Array1, ArrayView1};
+use ndarray::Array1;
 
 pub struct Network {
     layers: Vec<Box<dyn Layer>>,
-    loss: fn(ArrayView1<f64>, ArrayView1<f64>) -> f64,
-    loss_prime: fn(ArrayView1<f64>, ArrayView1<f64>) -> Array1<f64>,
+    loss: fn(&Array1<f64>, &Array1<f64>) -> f64,
+    loss_prime: fn(&Array1<f64>, &Array1<f64>) -> Array1<f64>,
 }
 
 impl Network {
@@ -30,10 +30,9 @@ impl Network {
         let mut result = vec![];
 
         for input in inputs.iter() {
-            let mut output = Array1::default(inputs[0].raw_dim());
-            output.assign(input);
+            let mut output = input.to_owned();
             for layer in &mut self.layers {
-                output = layer.forward_pass(output.view());
+                output = layer.forward_pass(output);
             }
             result.push(output);
         }
@@ -57,22 +56,21 @@ impl Network {
             let mut err = 0.0;
             for j in 0..num_samples {
                 // forward propagation
-                let mut output = Array1::default(x_train[0].raw_dim());
-                output.assign(&x_train[j]);
+                let mut output = x_train[j].to_owned();
                 for layer in self.layers.iter_mut() {
-                    output = layer.forward_pass(output.view());
+                    output = layer.forward_pass(output);
                 }
 
                 // compute loss
-                err += (self.loss)(y_train[j].view(), output.view());
+                err += (self.loss)(&y_train[j], &output);
 
                 // backward propagation
-                let mut error = (self.loss_prime)(y_train[j].view(), output.view());
+                let mut error = (self.loss_prime)(&y_train[j], &output);
                 for layer in self.layers.iter_mut().rev() {
                     if trivial_optimize {
-                        error = layer.backward_pass(error.view(), learning_rate / (i + 1) as f64);
+                        error = layer.backward_pass(error, learning_rate / (i + 1) as f64);
                     } else {
-                        error = layer.backward_pass(error.view(), learning_rate);
+                        error = layer.backward_pass(error, learning_rate);
                     }
                 }
             }