TableShift

https://tableshift.org

Admin

TableShift is a benchmark and associated Python library for machine learning with tabular data under...

Tokens:565

Snippets:7

Trust Score:5.2

Update:4 months ago

Show doc for...

Tokens:

Raw

### Fetch and Run TableShift Docker Image

Source: https://tableshift.org

Use these commands to pull the latest TableShift Docker image and run it to test your setup. The run command automatically launches example scripts.

```bash
# fetch the docker image
docker pull ghcr.io/jpgard/tableshift:latest

# run it to test your setup; this automatically launches examples/run_expt.py
docker run ghcr.io/jpgard/tableshift:latest --model xgb

```

--------------------------------

### Access TableShift Datasets in Python

Source: https://tableshift.org

This Python code demonstrates how to access a TableShift dataset for model training. It shows how to fetch data as a Pandas DataFrame and as a PyTorch DataLoader.

```python
# To run this Python code using the Docker environment from above, run:
# docker run -it --entrypoint=/usr/local/bin/python ghcr.io/jpgard/tableshift:latest

from tableshift import get_dataset

dataset_name = "diabetes_readmission"
dset = get_dataset(dataset_name)

# Fetch a pandas DataFrame of the training set
X_tr, y_tr, _, _ = dset.get_pandas("train")

# Fetch and use a pytorch DataLoader
train_loader = dset.get_dataloader("train", batch_size=1024)

for X, y, _, _ in train_loader:
    ...

```

--------------------------------

### TableShift Citation

Source: https://tableshift.org

This is the citation information for the TableShift paper, to be used when referencing the software or benchmark.

```bibtex
@article{gardner2023tableshift,
  title={Benchmarking Distribution Shift in Tabular Data with TableShift},
  author={Gardner, Josh and Popovic, Zoran and Schmidt, Ludwig},
  journal={Advances in Neural Information Processing Systems},
  year={2023}
}

```

Tokens:

Raw