I’m trying to load a HF model (Mistral-7B-v0.1) and tokenizer offline, from local files and using transformers library. I’ve downloaded the whole repository (config, model, weights, tokenizers, etc.) and I’m using the following code to load them which outputs : “name ‘tokenizer’ is not defined”

Thank you

import os
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig

os.environ['TRANSFORMERS_OFFLINE']='1'

model_path = "/media/models/mistral-7v1/"

try:
    config = AutoConfig.from_pretrained(model_path, local_files_only=True)
    tokenizer = AutoTokenizer.from_pretrained(model_path, config=config, local_files_only=True)
    model = AutoModelForCausalLM.from_pretrained(model_path, config=config, local_files_only=True)

except torch.cuda.OutOfMemoryError as e:
    print(f"Error: {e}")
    show_gpu()
except Exception as e:
    print(f"Error loading the model or tokenizer: {e}")


def generate_text(prompt, max_length=50):
    if tokenizer is None or model is None:
        print("Model or tokenizer not loaded properly.")
        return None

    try:
        # Encode the prompt
        inputs = tokenizer(prompt, return_tensors="pt")
        # Generate the text
        outputs = model.generate(inputs["input_ids"], max_length=max_length, num_return_sequences=1)
        # Decode the generated text
        text = tokenizer.decode(outputs[0], skip_special_tokens=True)
        return text
    except torch.cuda.OutOfMemoryError as e:
        print(f"Memory error: {e}")
    except Exception as e:
        print(f"Error generating text: {e}")


prompt = "Once upon a time"
try:
    generated_text = generate_text(prompt)
    if generated_text:
        print(f"Prompt: {prompt}")
        print(f"Generated Text: {generated_text}")
except Exception as e:
    print(f"Error in text generation: {e}")