Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- image-to-image
|
| 5 |
+
- text-to-image
|
| 6 |
+
tags:
|
| 7 |
+
- ai-generated
|
| 8 |
+
- image-generation
|
| 9 |
+
- prompts
|
| 10 |
+
- telegram
|
| 11 |
+
- midjourney
|
| 12 |
+
- stable-diffusion
|
| 13 |
+
- dalle
|
| 14 |
+
pretty_name: Telegram AI-Generated Images and Prompts
|
| 15 |
+
size_categories:
|
| 16 |
+
- 10K<n<100K
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
# Telegram AI-Generated Images and Prompts
|
| 20 |
+
|
| 21 |
+
A dataset of 28,118 AI-generated images paired with their generation prompts,
|
| 22 |
+
collected from a Telegram channel. Each record contains the image bytes
|
| 23 |
+
embedded in the `image` column plus the generation prompt/caption text.
|
| 24 |
+
|
| 25 |
+
The channel features AI-generated artwork from various text-to-image models
|
| 26 |
+
including Midjourney, Stable Diffusion, DALL-E, and others.
|
| 27 |
+
|
| 28 |
+
## Usage
|
| 29 |
+
|
| 30 |
+
```python
|
| 31 |
+
from datasets import load_dataset
|
| 32 |
+
ds = load_dataset("GCStream/telegram-channel-dataset")
|
| 33 |
+
```
|