Image-to-Text
PEFT
Safetensors
code-generation
multimodal
vision-encoder-decoder
lora
swin
qwen2.5-coder
code-trainer-v6
Instructions to use cmndcntrlcyber/code-trainer-vision-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use cmndcntrlcyber/code-trainer-vision-adapter with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Sync model card from docs/model_cards/code-trainer-vision-adapter.md
Browse files
README.md
CHANGED
|
@@ -152,5 +152,5 @@ print(model.generate(image, max_new_tokens=512))
|
|
| 152 |
python -m src.phase3_vision_model.scripts.launch_vision_training \
|
| 153 |
--config src/config/v6_config.yaml --wait
|
| 154 |
```
|
| 155 |
-
* **W&B project:** `rtpi-phase3-vision`.
|
| 156 |
* **Cost:** approximately $18 on `a100-large` (~5.5 h training + ~20 min eval).
|
|
|
|
| 152 |
python -m src.phase3_vision_model.scripts.launch_vision_training \
|
| 153 |
--config src/config/v6_config.yaml --wait
|
| 154 |
```
|
| 155 |
+
* **W&B project:** [`rtpi-phase3-vision`](https://wandb.ai/cmndcntrlcyber-c3s-consulting/rtpi-phase3-vision).
|
| 156 |
* **Cost:** approximately $18 on `a100-large` (~5.5 h training + ~20 min eval).
|