Metadata-Version: 2.1
Name: pandatorch
Version: 1.0.3
Summary: A flexible simple library that makes it easier to use the extrememly popular pandas package with the other extremely popular framework PyTorch.
Home-page: UNKNOWN
Author: Ashwin U Iyer
Author-email: ashwiniyer1706@gmail.com
License: UNKNOWN
Platform: UNKNOWN
Requires: numpy
Requires: pandas
Requires: pandoc
Requires: pytest
Requires: pylint
Requires: sklearn
Description-Content-Type: text/markdown

# PandaTorch 

A flexible simple library that makes it easier to use the extrememly popular pandas package with the other extremely popular framework PyTorch. 

## Functions

1. Converts a Pandas DataFrame into a usable PyTorch dataset.
2. Allows use of all usual Pandas functions

## Usage
```
import pandas as pd
from pandatorch import data
df=pd.read_csv("path_to_dataset")
torch_df=data.DataFrame(X=df.drop("<Target Column>",axis=1),y=df["<Target Column>"])
```

**Note:** Check out the notebooks folder for a full end-to-end training example of a tabular dataset using PandaTorch


