Thank you for this high quality material. I am running this on colab pro with L4 GPU. I am getting following error on trainer.train() line:
/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in __getattr__(self, name)
1929 if name in modules:
1930 return modules[name]
-> 1931 raise AttributeError(
1932 f"'{type(self).__name__}' object has no attribute '{name}'"
1933 )
AttributeError: 'Qwen2Model' object has no attribute 'model' (I am getting the same error for Meta-Llama-3.1-8B-Instruct). Is it some thing to do with correct version of torch ???
Thank you for this high quality material. I am running this on colab pro with L4 GPU. I am getting following error on trainer.train() line:
/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py in __getattr__(self, name)
1929 if name in modules:
1930 return modules[name]
-> 1931 raise AttributeError(
1932 f"'{type(self).__name__}' object has no attribute '{name}'"
1933 )
AttributeError: 'Qwen2Model' object has no attribute 'model' (I am getting the same error for Meta-Llama-3.1-8B-Instruct). Is it some thing to do with correct version of torch ???
Sorry, I made some mistake in wrapping base model with peft using get_peft_model. You code is perfect !!!! Working smoothly. Thank you for sharing
Excited to see that everything works smoothly. Enjoy 🤘
Loving your content. Also I'm loving the podcast feature. I can read at times and listen when I'm on the go!
I'm curious, did you build that into a ZenML pipeline? 😯