Skip to content

Commit b1c642e

Browse files
Copilotanxiangsir
andcommitted
Add Training Loss section with dark/light theme GIF support
Co-authored-by: anxiangsir <31175974+anxiangsir@users.noreply.github.com>
1 parent 19f42a7 commit b1c642e

1 file changed

Lines changed: 10 additions & 0 deletions

File tree

README.md

Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -73,6 +73,16 @@ The visualization below demonstrates our complete video processing pipeline. The
7373
2. **Avoid direct supervision from existing models** - Indirect usage is preferred over direct distillation, which may limit scaling capabilities
7474
3. **Progressive training when resources are limited** - Start with low resolution/frame rate, then gradually fine-tune to higher settings (ref: CLIPA)
7575

76+
### Training Loss
77+
78+
<p align="center">
79+
<picture>
80+
<source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/anxiangsir/asset/main/OneVision/loss_github_dark.gif">
81+
<source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/anxiangsir/asset/main/OneVision/loss_github_light.gif">
82+
<img alt="Training Loss Visualization" src="https://raw.githubusercontent.com/anxiangsir/asset/main/OneVision/loss_github_light.gif" width="800" style="max-width: 100%;">
83+
</picture>
84+
</p>
85+
7686
---
7787

7888
### Attentive Probe Results

0 commit comments

Comments
 (0)