https://github.com/woctezuma/glide-text2im-colab
Colab notebook for openai/glide-text2im.
https://github.com/woctezuma/glide-text2im-colab
colab colab-notebook colab-notebooks colaboratory google-colab google-colab-gpu google-colab-notebook google-colab-notebooks google-colaboratory openai openai-glide openai-glide-text2im openai-text2im
Last synced: 4 months ago
JSON representation
Colab notebook for openai/glide-text2im.
- Host: GitHub
- URL: https://github.com/woctezuma/glide-text2im-colab
- Owner: woctezuma
- License: mit
- Created: 2021-12-21T09:14:34.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2023-11-06T12:46:58.000Z (almost 2 years ago)
- Last Synced: 2025-04-08T10:50:24.824Z (6 months ago)
- Topics: colab, colab-notebook, colab-notebooks, colaboratory, google-colab, google-colab-gpu, google-colab-notebook, google-colab-notebooks, google-colaboratory, openai, openai-glide, openai-glide-text2im, openai-text2im
- Language: Jupyter Notebook
- Homepage:
- Size: 22.5 KB
- Stars: 21
- Watchers: 2
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# GLIDE text2im on Colab
This repository provides a Colab notebook to produce images conditioned on text prompts with [GLIDE][openai-glide-code] [1].
## Usage
- Run [`text2im.ipynb`][colab-notebook-text2im]
[![Open In Colab][colab-badge]][colab-notebook-text2im]
Tip: press `` to run everything.
## Results
The process is based on the small, filtered-data GLIDE model, with classifier-free guidance.
Results consist of 64x64 images, and the corresponding 256x256 upsampled versions.
Expected run-time: 2m30s (for the one-time set-up), 1 min (64x64 sampling), 30 sec (256x256 upsampling).
---
Several uncurated samples obtained with the same prompt: "a magnificent French rooster singing".
## Safety considerations
The *small* model has 300 million parameters, compared to the unreleased 3.5 billion parameter model.
As described in Appendix F.1, the training dataset was *filtered* so that it would not contain:
- images of humans and human-like objects,
- images of violent objects,
- two prevalent hate symbols in America (swastika and confederate flag).
## References
[1] Alex Nichol, Prafulla Dhariwal, Aditya Ramesh, et al. *GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models*. [arXiv preprint 2112.10741][openai-glide-paper]. 2021.
[openai-glide-code]:
[openai-glide-paper]:
[colab-notebook-text2im]:
[colab-badge]: