Efficiently Extract, Transform, and Load your dataset into PyTorch models
Project description
TorchETL
If you're working on classification problem, with dataset that is available in their native format (jpg, bmp, etc) and have PyTorch in your arsenal, you'll most likely feel that the DatasetFolder or ImageFolder is not good enough. So does vanilla torch.utils.data.Dataset. This library attempts to bridge that gap to effectively Extract, Transform, and Load your data by extending torch.utils.data.Dataset.
Main Features
Extract class would partition your dataset into train, validation, and test csv
TransformAndLoad class would Transform and consume your dataset efficiently
Prerequisites
Python 3.7.2 (other versions might work if type checking is supported)
torch
torchvision
numpy
pandas
opencv-python
sklearn
Or simply download requirements.txt and fire 'pip3 install -r requirements.txt'
Installing
pip3 install torchetl
Tutorial
See tutorial/Tutorial.ipynb
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
| Filename, size | File type | Python version | Upload date | Hashes |
|---|---|---|---|---|
| Filename, size torchetl-0.3.9-py3-none-any.whl (8.6 kB) | File type Wheel | Python version py3 | Upload date | Hashes View |
| Filename, size torchetl-0.3.9.tar.gz (4.8 kB) | File type Source | Python version None | Upload date | Hashes View |