自己量化操作时报错

[bopresslte]

Loading model file ..\models\CodeShell-7B-Chat\pytorch_model-00001-of-00002.bin Loading model file ..\models\CodeShell-7B-Chat\pytorch_model-00001-of-00002.bin Loading model file ..\models\CodeShell-7B-Chat\pytorch_model-00002-of-00002.bin Traceback (most recent call last): File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 1193, in main() File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 1134, in main model_plus = load_some_model(args.model) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 1056, in load_some_model model_plus = merge_multifile_models(models_plus) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 589, in merge_multifile_models
model = merge_sharded([mp.model for mp in models_plus]) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 568, in merge_sharded return {name: convert(name) for name in names} ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 568, in return {name: convert(name) for name in names} ^^^^^^^^^^^^^ File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 543, in convert lazy_tensors: list[LazyTensor] = [model[name] for model in models] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\AutoAI\llama_cpp_for_codeshell-master\convert.py", line 543, in lazy_tensors: list[LazyTensor] = [model[name] for model in models]

KeyError: 'transformer.wte.weight'
PS D:\AutoAI\llama_cpp_for_codeshell-master> python .\convert.py ..\models\CodeShell-7B-Chat\

[chiranko]

You may consider using our provided script "convert-codeshell-hf-to-gguf.py" for the conversion.