README.md

---
license: mit
language:
- en
pretty_name: HW1 Image Dataset
---


# Dataset Card for {{ pretty_name | default("Dataset Name", true) }}

This dataset covers 32 original photos of 6 landmarks at Carnegie Mellon University along with 320 pieces of artifial data. 
This could be used for image identification tasks or geolocation tasks. 

## Dataset Details

### Dataset Description

- **Curated by:** Carnegie Mellon University: 24-679
- **Shared by [optional]:** Devin DeCosmo
- **Language(s) (NLP):** English
- **License:** MIT

### Dataset Sources [optional]

<!-- Provide the basic links for the dataset. -->

- **Repository:** {{ repo | default("[More Information Needed]", true)}}


## Uses

The main use was to train tabular machine learning models to predict what landmark is being shown or to predict the GPS location
of a landmark in an image. 

### Direct Use

The direct use would be location or geolocal positioning tasks. 

### Out-of-Scope Use



## Dataset Structure

This dataset consists of two splits
An original split with 32 photos
An artificial split with 320 photos

The tasks fall into 6 categories based on the building pictured
1. Arts Building
2. Football Stadium
3. Gates Center
4. Hamerschlag Hall
5. Scaife Hall
6. Staircase to the Sky

## Dataset Creation

### Source Data

Source data is photos from a Moto 5G around CMU campus

#### Data Collection and Processing

Data for this was collected by the owner using a personal phone

#### Who are the source data producers?

Data was initially produced by the owner. 

## Bias, Risks, and Limitations

This is a very small data set and will likely have issues with training and fitting, especially for more complex identification problems. 

### Recommendations

This dataset probably has limited accuracy as a first draft but may be useful for learning how to train models.