Save and load the state of preprocessed text

You've finished preprocessing a dataset of agricultural questions and answers to help farmers look up common issues in the field. To save your progress, you decide to create and load checkpoints of the preprocessed text data.

Some data has been preloaded:

dataloader contains a preprocessed dataset of agricultural questions and answers
Accelerator has been imported from accelerate

This exercise is part of the course

Efficient AI Model Training with PyTorch

View Course

Exercise instructions

Save a checkpoint of the preprocessed data in checkpoint_dir.
Load a checkpoint from checkpoint_dir.

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

accelerator = Accelerator()

dataloader = accelerator.prepare(dataloader)

checkpoint_dir = Path("preprocess_checkpoint")

# Save a checkpoint of the data
____.____(checkpoint_dir)

# Load from the checkpoint
____.____(checkpoint_dir)

Edit and Run Code