IR-QLoRA
IR-QLoRA copied to clipboard
This project is the official implementation of our Accurate LoRA-Finetuning Quantization of LLMs via Information Retention
有 requirements.txt 吗 Traceback (most recent call last): File "/model/irqlora.py", line 858, in train() File "/model/irqlora.py", line 721, in train model, tokenizer = get_accelerate_model(args, checkpoint_dir) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/model/irqlora.py", line 409,...
as mentioned above
hello, how much time need and what datasets are u used?
Hello! I got RuntimeError as follow. I found the shapes of `factor_best.reshape(-1, 256, 1)` and `absmax` are respectively [2048,256,1] and [1024,256,1]. ```python RuntimeError: The size of tensor a (2048) must...
What a great work and thanks for your sharing! But when I try to get start with this work, I find difficulty in creating the conda environment. I tried "pip...