DisgustingOzil/qalaw-mistral-model

Project Name

Model Description

this model is just trained till 100 steps. It is basically traind on law related question answer especially UK Child Laws.

Requirements

!pip install -U xformers --index-url https://download.pytorch.org/whl/cu121
!pip install "unsloth[kaggle-new] @ git+https://github.com/unslothai/unsloth.git"

import os
os.environ["WANDB_DISABLED"] = "true"

Inference

load_in_4bit = True
from peft import AutoPeftModelForCausalLM
from transformers import AutoTokenizer
import torch
alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
Generate Answer of the question asked : 

### Input:
{}

### Response:
{}"""
model = AutoPeftModelForCausalLM.from_pretrained(
        "DisgustingOzil/qalaw-mistral-model", # YOUR MODEL YOU USED FOR TRAINING
        load_in_4bit = load_in_4bit,
        torch_dtype=torch.float16,
    )
tokenizer = AutoTokenizer.from_pretrained("DisgustingOzil/qalaw-mistral-model")

from unsloth import FastLanguageModel
FastLanguageModel.for_inference(model) # Enable native 2x faster inference

inputs = tokenizer(
[
    alpaca_prompt.format(
        "Explain Children and Young Persons Act", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
tokenizer.batch_decode(outputs)