Update README.md
Browse files
README.md
CHANGED
|
@@ -8,3 +8,25 @@ tags:
|
|
| 8 |
- video
|
| 9 |
---
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
- video
|
| 9 |
---
|
| 10 |
|
| 11 |
+
# AI Generated GTA V
|
| 12 |
+
|
| 13 |
+
A Deep Learning project that uses Diffusion transformers (DiT) to generate and Theft Auto V driving footage. This project is based on the [Open-Oasis Project](https://github.com/etched-ai/open-oasis)
|
| 14 |
+
|
| 15 |
+
Please see the GitHub repo for more info: https://github.com/ikergarcia1996/AI-Generated-GTAV
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
### Dit.safetensors
|
| 19 |
+
- Trained using 4xNvidia A100 80Gb in Bfloat16
|
| 20 |
+
- 64 batch size
|
| 21 |
+
- 1e-4 learning rate with constant scheduler and 5% warmup
|
| 22 |
+
- 1,610,000 steps
|
| 23 |
+
- ddim_noise_steps 50
|
| 24 |
+
- ctx noise increased from 0 to 40 during the first 50% of the training. Set to 40 during the remaining steps.
|
| 25 |
+
- No Action conditioning
|
| 26 |
+
|
| 27 |
+
### Dit_actions.safetensors
|
| 28 |
+
- Continual training of Dit.safetensors with action_conditioning for 210,000 steps.
|
| 29 |
+
- Trained using 4xNvidia A100 80Gb in Bfloat16
|
| 30 |
+
- 64 batch size
|
| 31 |
+
- 1e-4 learning rate with cosine scheduler to 1e-5 and 5% warmup
|
| 32 |
+
|