PyTorch DataLoader
Good job defining the Dataset class! The WaterDataset
you just created is now available for you to use.
The next step in preparing the training data is to set up a DataLoader
. A PyTorch DataLoader
can be created from a Dataset
to load data, split it into batches, and perform transformations on the data if desired. Then, it yields a data sample ready for training.
In this exercise, you will build a DataLoader
based on the WaterDataset
. The DataLoader
class you will need has already been imported for you from torch.utils.data
. Let's get to it!
This exercise is part of the course
Intermediate Deep Learning with PyTorch
Exercise instructions
- Create an instance of
WaterDataset
fromwater_train.csv
, assigning it todataset_train
. - Create
dataloader_train
based ondataset_train
, using a batch size of two and shuffling the samples. - Get a batch of features and labels from the DataLoader and print them.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create an instance of the WaterDataset
dataset_train = ____(____)
# Create a DataLoader based on dataset_train
dataloader_train = ____(
____,
batch_size=____,
shuffle=____,
)
# Get a batch of features and labels
features, labels = ____
print(features, labels)