| # # model_quantizer.py | |
| # from transformers import AutoModelForSequenceClassification | |
| # from optimum.onnxruntime import ORTOptimizer, ORTModelForSequenceClassification | |
| # from optimum.onnxruntime.configuration import OptimizationConfig | |
| # model = ORTModelForSequenceClassification.from_pretrained( | |
| # "Essay-Grader/roberta-ai-detector-20250401_232702", | |
| # from_transformers=True | |
| # ) | |
| # optimizer = ORTOptimizer.from_pretrained(model) | |
| # optimization_config = OptimizationConfig( | |
| # optimization_level=99, | |
| # enable_transformers_specific_optimizations=True, | |
| # optimize_for_gpu=True, | |
| # fp16=True | |
| # ) | |
| # optimizer.optimize( | |
| # save_dir="./optimized_model", | |
| # optimization_config=optimization_config | |
| # ) |