list-dataloaders 1.0.1

This package contains a pure-python class that wraps a list of dataloaders and allows to iterate over them in a random order.

List dataloaders

This repository contains an hyper-simple wrapper class that allows to iterate over a list of dataloaders in a transparent way.

Installation

pip install list-dataloaders

or directly from the repository:

pip install git+https://github.com/MorenoLaQuatra/list-dataloaders.git

Usage

from list_dataloaders import ListDataLoaders

list_dataloaders = []
train_dl1 = DataLoader(...)
train_dl2 = DataLoader(...)
train_dl3 = DataLoader(...)
list_dataloaders.append(train_dl1)
list_dataloaders.append(train_dl2)
list_dataloaders.append(train_dl3)

train_dls = ListDataLoaders(list_dataloaders)

for batch in train_dls:
    print(batch)

Options and parameters

The main class ListDataLoaders has the following parameters:

• dataloaders: the list of dataloaders to iterate over
• verbose: if True, prints the number of samples in each dataloader (default: False)
• weight_by_num_samples: if True, the probability of sampling a dataloader is proportional to the number of samples in the dataloader (default: True)

License

This project is licensed under the terms of the MIT license.

Issues

If you encounter any problems, please file an issue along with a detailed description.