USPS Dataset

Estimated reading: 3 minutes

Visualization of USPS dataset in the Deep Lake UI

What is USPS Dataset?

The USPS dataset showcases a broad range of font styles. The dataset was made by automatically scanning envelopes by the U.S. Postal Service. The dataset consists of 9,298 16×16 pixel grayscale samples. Each image in the dataset is centered and normalized.

Download USPS Dataset in Python

Instead of downloading the USPS dataset in Python, you can effortlessly load it in Python via our Deep Lake open-source with just one line of code.

Load MNIST Dataset Training Subset in Python

				
					import deeplake
ds = deeplake.load('hub://activeloop/usps-train')

Load USPS Dataset Testing Subset in Python

				
					import deeplake
ds = deeplake.load('hub://activeloop/usps-test')

USPS Dataset Structure

USPS Data Fields

images: tensor containing the image.
labels: tensor to represent image label categories.

USPS Data Splits

The USPS dataset training set is composed of 7291 images.
The USPS dataset testing set is composed of 2007 images.

How to use USPS Dataset with PyTorch and TensorFlow in Python

Train a model on USPS dataset with PyTorch in Python

Let’s use Deep Lake built-in PyTorch one-line dataloader to connect the data to the compute:

				
					dataloader = ds.pytorch(num_workers=0, batch_size=4, shuffle=False)

Train a model on the USPS dataset with TensorFlow in Python

				
					dataloader = ds.tensorflow()

Additional Information about USPS Dataset

USPS Dataset Description

Homepage:https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#usps
Paper: J. J. Hull, “A database for handwritten text recognition research,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16, no. 5, pp. 550-554, May 1994, doi: 10.1109/34.291440.
Point of Contact: N/A

USPS Dataset Curators

J. J. Hull

USPS Dataset Licensing Information

Deep Lake users may have access to a variety of publicly available datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have a license to use the datasets. It is your responsibility to determine whether you have permission to use the datasets under their license.

If you’re a dataset owner and do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thank you for your contribution to the ML community!

USPS Dataset Citation Information

				
					@ARTICLE{291440,
  author={Hull, J.J.},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
  title={A database for handwritten text recognition research}, 
  year={1994},
  volume={16},
  number={5},
  pages={550-554},
  doi={10.1109/34.291440}}

USPS Dataset FAQs

What is the USPS dataset for Python?

The USPS dataset is often used in computer vision, pattern recognition, and handwritten text recognition research. USPS is often used as a benchmark dataset as it showcases a broad range of font styles. Also, the images in the dataset are 16×16 pixel grayscale samples, and each image in the dataset is centered and normalized.

How to download the USPS dataset in Python?

You can load the USPS dataset fast with one line of code using the open-source package Activeloop Deep Lake in Python. See detailed instructions on how to load the USPS dataset training subset and testing subset in Python.

How can I use the USPS dataset in PyTorch or TensorFlow?

You can stream the USPS dataset while training a model in PyTorch or TensorFlow with one line of code using the open-source package Activeloop Deep Lake in Python. See detailed instructions on how to train a model on the USPS dataset with PyTorch in Python or train a model on the USPS dataset with TensorFlow in Python.