Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,31 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model:
|
| 4 |
+
- genmo/mochi-1-preview
|
| 5 |
+
pipeline_tag: text-to-video
|
| 6 |
+
tags:
|
| 7 |
+
- infinite zoom
|
| 8 |
+
- art style
|
| 9 |
+
- mochi
|
| 10 |
+
- diffusion
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Fine-Tuning Mochi Text-to-Video: InfiniteZoom-Mochi
|
| 14 |
+
|
| 15 |
+
This project demonstrates the fine-tuning of the **Mochi Text-to-Video** model using a LoRA (Low-Rank Adaptation) approach, focusing on the **infinite zoom art style**.
|
| 16 |
+
|
| 17 |
+
## Training Details
|
| 18 |
+
|
| 19 |
+
- **Model Base**: Mochi Text-to-Video
|
| 20 |
+
- **Fine-Tuning Dataset**: 23 short video clips of infinite zoom art style
|
| 21 |
+
- **Training Settings :**: 37 frames
|
| 22 |
+
- **Training Hardware**: H100 GPU
|
| 23 |
+
- **Training Duration**: 2.2h
|
| 24 |
+
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
## Results
|
| 28 |
+
|
| 29 |
+
Below is an example of the generated video output:
|
| 30 |
+
|
| 31 |
+
### ** Samples (from training only)**
|