Update Use_GPU_In_PyTorch.md

contrib/machine-learning/Use_GPU_In_PyTorch.md

@@ -22,8 +22,11 @@ To check if you've got access to a Nvidia GPU, you can run `!nvidia-smi` where t
 
 ```python
 !nvidia-smi
+```
 
-output -> /bin/bash: line 1: nvidia-smi: command not found
+#### Output
+```
+/bin/bash: line 1: nvidia-smi: command not found
 ```    
 
 As you can see that it  is showing that  `command not found error` it means currently we do'nt have colab GPU access.
@@ -40,8 +43,9 @@ Now to check again run the command
 
 ```python
 !nvidia-smi
+```
 
-output -> Fri May 31 04:01:18 2024       
+#### Output -> Fri May 31 04:01:18 2024       
     +---------------------------------------------------------------------------------------+
     | NVIDIA-SMI 535.104.05             Driver Version: 535.104.05   CUDA Version: 12.2     |
     |-----------------------------------------+----------------------+----------------------+
@@ -61,7 +65,6 @@ output -> Fri May 31 04:01:18 2024
     |=======================================================================================|
     |  No running processes found                                                           |
     +---------------------------------------------------------------------------------------+
- ```   
 
 Whoo!!! now we have GPU access
 
@@ -80,8 +83,10 @@ You can test if PyTorch has access to a GPU using `torch.cuda.is_available()`.
 # Check for GPU
 import torch
 torch.cuda.is_available()
-
-output -> True
+```
+#### Output
+```
+True
 ```
 
 
@@ -97,8 +102,11 @@ Let's create a device variable to store what kind of device is available.
 ```python
 device = "cuda" if torch.cuda.is_available else "cpu"
 device
+```
 
-output -> 'cuda'
+#### Output
+```
+'cuda'
 ```
 
 
@@ -110,8 +118,11 @@ You can count the number of GPUs PyTorch has access to using `torch.cuda.device_
 
 ```python
 torch.cuda.device_count()
+```
 
-output -> 1
+#### Output
+```
+1
 ```
 
 
@@ -123,8 +134,11 @@ output -> 1
 tensor = torch.tensor([1,2,3])
 
 print(f"Tensor is running on the :{tensor.device}")
+```
 
-output -> Tensor is running on the :cpu
+#### Output
+```
+Tensor is running on the :cpu
 ```    
 
 Note: By default tensors run on the 'CPU'
@@ -135,8 +149,11 @@ Note: By default tensors run on the 'CPU'
 tensor_on_gpu = tensor.to(device)
 
 print(f"Tensor is running on:{tensor_on_gpu.device}")
+```
 
-output -> Tensor is running on:cuda:0
+#### Output
+```
+Tensor is running on:cuda:0
 ```    
 
 Notice the second tensor has device=`'cuda:0'`, this means it's stored on the 0th GPU available (GPUs are 0 indexed, if two GPUs were available, they'd be `'cuda:0'` and `'cuda:1'` respectively, up to `'cuda:n'`).
@@ -153,8 +170,9 @@ Let's try using the `torch.Tensor.numpy()` method on our `tensor_on_gpu`.
 ```python
 # If tensor is on GPU, can't transform it to NumPy (this will error)
 tensor_on_gpu.numpy()
+```
 
-output -> 
+#### Output 
 
 
     ---------------------------------------------------------------------------
@@ -168,7 +186,6 @@ output ->
 
     TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
     
-```
 Instead, to get a tensor back to CPU and usable with NumPy we can use Tensor.cpu().
 
 This copies the tensor to CPU memory so it's usable with CPUs
@@ -178,7 +195,10 @@ This copies the tensor to CPU memory so it's usable with CPUs
 # Instead, copy the tensor back to cpu
 tensor_back_on_cpu = tensor_on_gpu.cpu().numpy()
 tensor_back_on_cpu
-
-output -> array([1, 2, 3])
+```
+
+#### Output
+```
+array([1, 2, 3])
 ```