[feature] adding orthogononal finetuning (OFT) to llama factory (#8623)

Co-authored-by: Zeju <zqiu@g003.internal.cluster.is.localnet>
Co-authored-by: Zeju <zqiu@login2.is.localnet>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>

src/llamafactory/extras/constants.py

@@ -56,13 +56,13 @@ LAYERNORM_NAMES = {"norm", "ln"}
 
 LLAMABOARD_CONFIG = "llamaboard_config.yaml"
 
-METHODS = ["full", "freeze", "lora"]
+METHODS = ["full", "freeze", "lora", "oft"]
 
 MOD_SUPPORTED_MODELS = {"bloom", "falcon", "gemma", "llama", "mistral", "mixtral", "phi", "starcoder2"}
 
 MULTIMODAL_SUPPORTED_MODELS = set()
 
-PEFT_METHODS = {"lora"}
+PEFT_METHODS = {"lora", "oft"}
 
 RUNNING_LOG = "running_log.txt"