Here is 1 way to prepare your data using DataLoader in Pytorch.
1. Create Custom Dataset Class
When using a custom dataset the following 3 functions has to be overloaded.
class Dataset:
#Initialize the dataset with the input data and the corresponding labels
def __init__(self):
self.data=[1,2,3,4,5,6]
self.label=[1,0,0,0,1,1]
# This method is called whenever you would use object[index] to access any element in the dataset
def __getitem__(self,index):
return self.data[index],self.label[index]
# Method to simply return the number of training samples
def __len__(self):
#Now you can create an instance of your dataset
prepared_dataset = Dataset()
2. Load Data Using DataLoader
# These are some examples of possible arguments
loaded_data = DataLoader(prepared_dataset, batch_size=1, shuffle=False, sampler=None,
batch_sampler=None, num_workers=0, collate_fn=None,
pin_memory=False, drop_last=False, timeout=0,
worker_init_fn=None, *, prefetch_factor=2,
persistent_workers=False)
References:
Blog Post by Manpreet Singh Minhas