Morteza Heidari

icon indicating a website link https://morteza89.github.io/

icon company Intel icon location NY icon location Staff machine learning engineer @Intel

Results 1 comments of Morteza Heidari

PEFT to ONNX conversion

I'm attempting to convert a fine-tuned TinyLlama model, which has been fine-tuned using a LoRA configuration. The PEFT model is approximately 2.1 GB on disk. When I load it for...