README.md

---
license: mit
pretty_name: Raw Explicit Image Dataset 1 Split 2
size_categories:
- 1K<n<10K
extra_gated_prompt: This dataset contains NSFW content that has not undergone manual
  review and may include material unsuitable for all audiences. It is intended exclusively
  for research and educational purposes within the context of studying human behavior,
  societal norms, and artistic expression. By accessing this dataset, you acknowledge
  and agree that you will use it solely for legitimate academic or research purposes
  and handle the information responsibly and ethically, considering the potential
  impact of your work. If you utilize this dataset to train an AI model, you agree
  and understand that this dataset was not manually reviewed, and that Interformed
  is not responsible for the outputs of your model.
dataset_info:
  features:
  - name: file_name
    dtype: string
  - name: subreddit
    dtype: string
  - name: sort_type
    dtype: string
  - name: image
    dtype: image
  splits:
  - name: raw
    num_bytes: 1267017218.875
    num_examples: 1297
  download_size: 1265041248
  dataset_size: 1267017218.875
configs:
- config_name: default
  data_files:
  - split: raw
    path: data/nsfw_bw/raw-*
---

# Explicit Image Dataset 1
Explicit Image Dataset 1 (EXIM1) is a collection of user-generated content from a variety of publicly-availably sources across the internet. EXIM1's images primarily come from Reddit.

Most, if not all of the state-of-the-art diffusion and image generation models come 'aligned' out-of-the-box. While AI safety is an important aspect of this profession, such alignment often results in less than ideal outputs from the model. The goal of this dataset is to provide a comprehensive collection of anatomically accurate images in a variety of settings, clothing, etc.

EXIM1_RAW is nothing more than a __raw__ dataset. This means that none of the images have been quality-checked, captioned, or verified. We intend to make ready-to-go datasets shortly, but we also provide the raw versions of our datasets. This dataset, along with all of Interformed's datasets are unfinished subject to change at __any__ time.

## Dataset Information
Due to the large size of the data, EXIM1 is divided into parts. 
- Part 1 ([EXIM1-1](https://huggingface.co/datasets/Interformed/EXIM1-1_RAW)) contains a collection of user-generated images from subreddits in alphabetical order #-E.
- Part 2 ([EXIM1-2](https://huggingface.co/datasets/Interformed/EXIM1-2_RAW)) contains a collection of user-generated images from subreddits in alphabetical order F-L.
- Part 3 ([EXIM1-3](https://huggingface.co/datasets/Interformed/EXIM1-3_RAW)) contains a collection of user-generated images from subreddits in alphabetical order M-?.

## Dataset Structure

### Dataset Splits

| Split | Approx. Size | Description |
| ----- | ------------ | ----------- |
| raw | 246GB | The one split for this dataset. Since this is purely raw data, there is not a need to validate the images or caption them yet. |

### Dataset Fields
| Field | Type | Description |
| ----- | ---- | ----------- |
| file_name | string | The image file's name (typically formatted with subredditname_sort_timestamp_number.ext) |
| subreddit | string | For images acquired through Reddit, the subreddit name where the image originated from. |
| sort_type | string | For images acquired through Reddit, the sorting column from which the image originated from (hot, top, controversial, etc.) |
| image | Image | The image... |

## Notes
- This dataset is not finished. It is being uploaded in alphabetical order. Please refrain from requesting a subreddit/booru/any other image board's contents before that letter has been uploaded. You will not get a response.
- The license for this dataset is subject to change. Please be mindful of the agreements of both the gated repository agreement and the dataset license as you use it.