| Using Lora to train opt-6.7b | |
| import torch | |
| from peft import PeftModel, PeftConfig | |
| from transformers import AutoModelForCausalLM, AutoTokenizer | |
| peft_model_id = "sayril007/opt_lora-7b-lora-pretrained" | |
| config = PeftConfig.from_pretrained(peft_model_id) | |
| model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, return_dict=True, load_in_8bit=True, device_map='auto') | |
| tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path) | |
| # Load the Lora model | |
| model = PeftModel.from_pretrained(model, peft_model_id) | |
| batch = tokenizer("Two things are infinite: ", return_tensors='pt') | |
| with torch.cuda.amp.autocast(): | |
| output_tokens = model.generate(**batch, max_new_tokens=50) | |
| print('\n\n', tokenizer.decode(output_tokens[0], skip_special_tokens=True)) |