WISER_lora
Fine-tuned from Qwen/Qwen2.5-VL-7B-Instruct using supervised fine-tuning (SFT).
Training Results
| Metric | Value |
|---|---|
| epoch | 2.0 |
| total_flos | 2.0316819022261125e+18 |
| train_loss | 0.5175499592138373 |
| train_runtime | 21516.9615 |
| train_samples_per_second | 0.462 |
| train_steps_per_second | 0.058 |
Usage
from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration
model = Qwen2_5_VLForConditionalGeneration.from_pretrained("jaehojung25/WISER_lora")
processor = AutoProcessor.from_pretrained("jaehojung25/WISER_lora")
- Downloads last month
- 8
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for jaehojung25/WISER_lora
Base model
Qwen/Qwen2.5-VL-7B-Instruct