brittlewis12 / autogguf Goto Github PK
View Code? Open in Web Editor NEWEasily convert HuggingFace models to GGUF-format for llama.cpp
License: MIT License
Easily convert HuggingFace models to GGUF-format for llama.cpp
License: MIT License
While autoguff worked for the first model I tried, it hasn't worked for all the the ones I tried after.
This is probably due to the models I'm trying to convert more than anything, but just to be sure I though I'd share:
Traceback (most recent call last):
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1466, in <module>
main()
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1402, in main
model_plus = load_some_model(args.model)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1278, in load_some_model
models_plus.append(lazy_load_file(path))
^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 887, in lazy_load_file
return lazy_load_torch_file(fp, path)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 843, in lazy_load_torch_file
model = unpickler.load()
^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 832, in find_class
return self.CLASSES[(module, name)]
~~~~~~~~~~~~^^^^^^^^^^^^^^^^
KeyError: ('torch', 'ByteStorage')
Error: failed with return code 1
and
Traceback (most recent call last):
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1466, in <module>
main()
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1402, in main
model_plus = load_some_model(args.model)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1278, in load_some_model
models_plus.append(lazy_load_file(path))
^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 890, in lazy_load_file
return lazy_load_safetensors_file(fp, path)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 869, in lazy_load_safetensors_file
model = {name: convert(info) for (name, info) in header.items() if name != '__metadata__'}
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 869, in <dictcomp>
model = {name: convert(info) for (name, info) in header.items() if name != '__metadata__'}
^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 857, in convert
data_type = SAFETENSORS_DATA_TYPES[info['dtype']]
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^
KeyError: 'U8'
Error: failed with return code 1
and
Traceback (most recent call last):████████████████████████████████████████████████████████████| 551M/551M [02:04<00:00, 8.42MB/s]
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1466, in <module>
main()
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1402, in main
model_plus = load_some_model(args.model)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/elonmusk/code/llama.cpp/convert.py", line 1271, in load_some_model
raise Exception(f"Found multiple models in {path}, not sure which to pick: {files}")
Exception: Found multiple models in awesome_recipes_exp, not sure which to pick: [PosixPath('awesome_recipes_exp/optimizer.pt'), PosixPath('awesome_recipes_exp/scheduler.pt'), PosixPath('awesome_recipes_exp/pytorch_model.bin')]
Error: failed with return code 1
Hey. Can i use this to convert a local fine tunbed model (LORA)?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.