Instructions to use Jeckmu/Qwen2-VL-2B-Instruct-GPTQ-Int4-lora-SurveillanceVideo-250210 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Jeckmu/Qwen2-VL-2B-Instruct-GPTQ-Int4-lora-SurveillanceVideo-250210 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-VL-2B-Instruct-GPTQ-Int4") model = PeftModel.from_pretrained(base_model, "Jeckmu/Qwen2-VL-2B-Instruct-GPTQ-Int4-lora-SurveillanceVideo-250210") - Notebooks
- Google Colab
- Kaggle
| { | |
| "epoch": 2.932862190812721, | |
| "num_input_tokens_seen": 1841024, | |
| "total_flos": 7451631889809408.0, | |
| "train_loss": 0.1802581724666414, | |
| "train_runtime": 2989.6538, | |
| "train_samples_per_second": 0.567, | |
| "train_steps_per_second": 0.035 | |
| } |