Open yige0617 opened 1 year ago
I cannot feed model = AutoGPTQForCausalLM.from_quantized(quantized_model_dir, use_cuda_fp16=False,device="cuda:0") directly into below function keras_model = nobuco.pytorch_to_keras( model, ...... ) what should be done here? Thanks! What should I do to transform it into
Do we have any instruction to llama 2 conversion from pytorch to tensorflow?