Datasets:
Nick Padgett
commited on
Commit
•
fc433d8
1
Parent(s):
02dd306
Updating README and adding tutorials.
Browse files- README.md +43 -3
- assets/spawning-logo.png +3 -0
- tutorials/images.md +29 -0
- tutorials/metadata.md +30 -0
README.md
CHANGED
@@ -1,3 +1,43 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Spawning-15m
|
2 |
+
|
3 |
+
![Spawning-15m](./assets/spawning-logo.png)
|
4 |
+
|
5 |
+
# Summary
|
6 |
+
**Spawning-15m** dataset is a collection of about 15 million CC0/PD image-caption pairs for the purpose of training generative image models.
|
7 |
+
|
8 |
+
# About
|
9 |
+
Training a state-of-the-art generative image model typically requires vast amounts of images from across the internet. Training with images from across the web introduces several data quality issues: the presence of copyright material, low quality images and captions, violent or nsfw content, PII, decaying dataset quality via broken links, etc. Additionally, downloading from the original image hosts introduces an undue burden to those hosts, impacting services for legitimate users.
|
10 |
+
|
11 |
+
The Spawning-15m dataset aims to resolve these issues through collecting only public domain and cc0 licensed images, automated recaptioning of image data, quality and safety filtering, and hosting the images in the dataset on dedicated cloud storage separate from the original image hosts. These innovations make Spawning-15m the largest safe and reliable public image dataset available.
|
12 |
+
|
13 |
+
Built and curated with [Source.Plus](https://source.plus).
|
14 |
+
|
15 |
+
# Overview
|
16 |
+
This dataset has two components. The first is the `metadata`, which contains the image urls, captions, image dimensions, etc. The second component are the `images`.
|
17 |
+
|
18 |
+
## Metadata
|
19 |
+
The metadata is made available through a series of parquet files with the following schema:
|
20 |
+
- `id`: A unique identifier for the image.
|
21 |
+
- `url`: The URL of the image.
|
22 |
+
- `s3_key`: The S3 file key of the image.
|
23 |
+
- `caption`: A caption for the image.
|
24 |
+
- `md5_hash`: The MD5 hash of the image file.
|
25 |
+
- `mime_type`: The MIME type of the image file.
|
26 |
+
- `width`: The width of the image in pixels.
|
27 |
+
- `height`: The height of the image in pixels.
|
28 |
+
- `license_type`: The URL of the license.
|
29 |
+
|
30 |
+
## Images
|
31 |
+
The image files are all hosted in the AWS S3 bucket `spawning-15m`. The URLs to the images files are all maintained in the metadata files.
|
32 |
+
|
33 |
+
# Tutorials
|
34 |
+
|
35 |
+
[Working with the Metadata](./tutorials/metadata.md)
|
36 |
+
|
37 |
+
[Downloading Images](./tutorials/images.md)
|
38 |
+
|
39 |
+
# License
|
40 |
+
The dataset is licensed under the [CDLA-Permissive-2.0](https://cdla.dev/permissive-2-0/).
|
41 |
+
|
42 |
+
# Reporting Issues
|
43 |
+
We've gone through great lengths to ensure the dataset is free from objectionable and infringing content. If you find any issues or have any concerns, please report them to us at [[email protected]](mailto:[email protected]), along with the id of the relevant item.
|
assets/spawning-logo.png
ADDED
Git LFS Details
|
tutorials/images.md
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Downloading Images
|
2 |
+
Once you have the URLs or S3 file keys from the metadata ([follow the steps here here]("./metadata.md")), you can download the images through any standard means.
|
3 |
+
|
4 |
+
#### cURL
|
5 |
+
Download an image from a url to a local image file with the name `image.png`:
|
6 |
+
```bash
|
7 |
+
curl -O image.png https://spawning-15m.s3.us-west-2.amazonaws.com/image.png
|
8 |
+
```
|
9 |
+
#### Python
|
10 |
+
Download an image from a url to a local image file with the name `image.png`:
|
11 |
+
```python
|
12 |
+
import requests
|
13 |
+
|
14 |
+
url = "https://spawning-15m.s3.us-west-2.amazonaws.com/image.png"
|
15 |
+
response = requests.get(url)
|
16 |
+
with open('image.png', 'wb') as f:
|
17 |
+
f.write(response.content)
|
18 |
+
```
|
19 |
+
#### img2dataset
|
20 |
+
You can also use the `img2dataset` tool to quickly download images from a metadata file. The tool is available [here](https://github.com/rom1504/img2dataset). The example below will download all the images to a local `images` directory.
|
21 |
+
```bash
|
22 |
+
img2dataset download --url_list spawning-15m-metadata.001.parquet --input_format parquet --url_col url --caption_col caption --output-dir images/
|
23 |
+
```
|
24 |
+
|
25 |
+
#### S3 CLI
|
26 |
+
Download an image from an S3 bucket to an image with the name `image.png`:
|
27 |
+
```bash
|
28 |
+
aws s3 cp s3://spawning-15m/image.png image.png
|
29 |
+
```
|
tutorials/metadata.md
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Working with the Metadata
|
2 |
+
The metadata files are in parquet format, and contain the following attributes:
|
3 |
+
- `id`: A unique identifier for the image.
|
4 |
+
- `url`: The URL of the image.
|
5 |
+
- `s3_key`: The S3 file key of the image.
|
6 |
+
- `caption`: A caption for the image.
|
7 |
+
- `md5_hash`: The MD5 hash of the image file.
|
8 |
+
- `mime_type`: The MIME type of the image file.
|
9 |
+
- `width`: The width of the image in pixels.
|
10 |
+
- `height`: The height of the image in pixels.
|
11 |
+
- `license_type`: The URL of the license.
|
12 |
+
|
13 |
+
#### Open a metadata file
|
14 |
+
The files are in parquet format, and can be opened with a tool like `pandas` in Python.
|
15 |
+
```python
|
16 |
+
import pandas as pd
|
17 |
+
df = pd.read_parquet('spawning-15m-metadata.001.parquet')
|
18 |
+
```
|
19 |
+
|
20 |
+
#### Get URLs from metadata
|
21 |
+
Once you have opened a maetadata file with pandas, you can get the URLs of the images with the following command:
|
22 |
+
```python
|
23 |
+
urls = df['url']
|
24 |
+
```
|
25 |
+
|
26 |
+
#### Get S3 File Keys from metadata
|
27 |
+
You can also get the S3 file keys, which can be used to download the images using the S3 CLI:
|
28 |
+
```python
|
29 |
+
s3_keys = df['s3_key']
|
30 |
+
```
|