Fixes to code examples in section 5.2 (#1594)

* Fixes to code examples in section 5.2

* A more generic way to get a device for code examples in Burn book section 5.2

* Change run-checks instruction + fix comment spacing

---------

Co-authored-by: hrishim <[email protected]>
Co-authored-by: Guillaume Lagrange <[email protected]>

CONTRIBUTING.md

@@ -45,7 +45,7 @@ Be sure to follow the coding standards and guidelines used in the rest of the pr
 
 ### Step 6: Run the Pre-Pull Request Script
 
-Before you open a pull request, please run the [`run-checks.sh`](/run-before-pr.sh) script. This
+Before you open a pull request, please run [`./run-checks.sh all`](/run-checks.sh). This
 will ensure that your changes are in line with our project's standards and guidelines. You can run
 this script by opening a terminal, navigating to your local project directory, and typing
 `./run-checks`.

burn-book/src/building-blocks/tensor.md

@@ -20,10 +20,13 @@ The actual shape of the tensor is inferred from its initialization. For example,
 ```rust, ignore
 let floats = [1.0, 2.0, 3.0, 4.0, 5.0];
 
+// Get the default device
+let device = Default::default();
+
 // correct: Tensor is 1-Dimensional with 5 elements
-let tensor_1 = Tensor::<Backend, 1>::from_floats(floats);
+let tensor_1 = Tensor::<Backend, 1>::from_floats(floats, &device);
 
-// incorrect: let tensor_1 = Tensor::<Backend, 5>::from_floats(floats);
+// incorrect: let tensor_1 = Tensor::<Backend, 5>::from_floats(floats, &device);
 // this will lead to an error and is for creating a 5-D tensor
 ```
 
@@ -38,19 +41,19 @@ a tensor from different inputs.
 ```rust, ignore
 
 // Initialization from a given Backend (Wgpu)
-let tensor_1 = Tensor::<Wgpu, 1>::from_data([1.0, 2.0, 3.0]);
+let tensor_1 = Tensor::<Wgpu, 1>::from_data([1.0, 2.0, 3.0], &device);
 
 // Initialization from a generic Backend
-let tensor_2 = Tensor::<Backend, 1>::from_data(Data::from([1.0, 2.0, 3.0]).convert());
+let tensor_2 = Tensor::<Backend, 1>::from_data(Data::from([1.0, 2.0, 3.0]).convert(), &device);
 
 // Initialization using from_floats (Recommended for f32 ElementType)
 // Will be converted to Data internally.
 // `.convert()` not needed as from_floats() defined for fixed ElementType
-let tensor_3 = Tensor::<Backend, 1>::from_floats([1.0, 2.0, 3.0]);
+let tensor_3 = Tensor::<Backend, 1>::from_floats([1.0, 2.0, 3.0], &device);
 
 // Initialization of Int Tensor from array slices
 let arr: [i32; 6] = [1, 2, 3, 4, 5, 6];
-let tensor_4 = Tensor::<Backend, 1, Int>::from_data(Data::from(&arr[0..3]).convert());
+let tensor_4 = Tensor::<Backend, 1, Int>::from_data(Data::from(&arr[0..3]).convert(), &device);
 
 // Initialization from a custom type
 
@@ -66,7 +69,7 @@ let bmi = BodyMetrics{
         weight: 80.0
     };
 let data  = Data::from([bmi.age as f32, bmi.height as f32, bmi.weight]).convert();
-let tensor_5 = Tensor::<Backend, 1>::from_data(data);
+let tensor_5 = Tensor::<Backend, 1>::from_data(data, &device);
 
 ```
 
@@ -84,7 +87,7 @@ times will necessitate cloning it. Let's look at an example to understand the ow
 cloning better. Suppose we want to do a simple min-max normalization of an input tensor.
 
 ```rust, ignore
-let input = Tensor::<Wgpu, 1>::from_floats([1.0, 2.0, 3.0, 4.0]);
+let input = Tensor::<Wgpu, 1>::from_floats([1.0, 2.0, 3.0, 4.0], &device);
 let min = input.min();
 let max = input.max();
 let input = (input - min).div(max - min);
@@ -98,7 +101,7 @@ available for further operations. Burn Tensors like most complex primitives do n
 doing min-max normalization with cloning.
 
 ```rust, ignore
-let input = Tensor::<Wgpu, 1>::from_floats([1.0, 2.0, 3.0, 4.0]);
+let input = Tensor::<Wgpu, 1>::from_floats([1.0, 2.0, 3.0, 4.0], &device);
 let min = input.clone().min();
 let max = input.clone().max();
 let input = (input.clone() - min.clone()).div(max - min);