Winnie Yeung

Identify trainable layers

1

Hi I am trying to apply StyleGAN-NADA Clip method onto OASIS as suggested by the author, can you help me to identify which layer of OASIS is trainable?

Don’t know how to ingest data to use the model

1

In the markdown, you mentioned `loss=trainer(data)` . As a new user, I’m still figuring out how to use the models with custom data. Maybe I can add a toy_problem.py as...

`│ /usr/lib/python3/dist-packages/llmfoundry/models/hf/hf_causal_lm.py:119 in │ │ __init__ │ │ │ │ 116 │ │ ] │ │ 117 │ │ │ │ 118 │ │ # Construct the Hugging Face config...