unsloth/tests/saving/language_models/test_merge_model_perplexity_llama-3.2.py

from unsloth import FastLanguageModel, FastVisionModel, UnslothVisionDataCollator
from unsloth.chat_templates import get_chat_template
from trl import SFTTrainer, SFTConfig
from transformers import (
    DataCollatorForLanguageModeling,
    DataCollatorForSeq2Seq,
    TrainingArguments,
)
from datasets import load_dataset, Dataset
import torch
from tqdm import tqdm
import pandas as pd
import multiprocessing as mp
from multiprocessing import Process, Queue
import gc

# ruff: noqa
import sys
from pathlib import Path


REPO_ROOT = Path(__file__).parents[3]
sys.path.insert(0, str(REPO_ROOT))


from tests.utils.cleanup_utils import safe_remove_directory
from tests.utils.perplexity_eval import (
    ppl_model,
    add_to_comparison,
    print_model_comparison,
)


def formatting_prompts_func(examples):
    convos = examples["messages"]
    texts = [
        tokenizer.apply_chat_template(convo, tokenize = False, add_generation_prompt = False)
        for convo in convos
    ]
    return {"text": texts}


def load_and_compute_8bit_ppl(
    result_queue,
    load_in_4bit = False,
    load_in_8bit = False,
):
    """Load model and compute perplexity in subprocess"""
    from unsloth import FastLanguageModel
    from unsloth.chat_templates import get_chat_template
    from tests.utils.perplexity_eval import ppl_model

    merged_model, merged_tokenizer = FastLanguageModel.from_pretrained(
        model_name = "./unsloth_out/merged_llama_text_model",
        max_seq_length = 2048,
        load_in_4bit = load_in_4bit,
        load_in_8bit = load_in_8bit,
    )
    merged_tokenizer = get_chat_template(
        merged_tokenizer,
        chat_template = "llama-3.1",
    )

    # Load dataset fresh in subprocess
    dataset_ppl = load_dataset("allenai/openassistant-guanaco-reformatted", split = "eval")

    def formatting_prompts_func(examples):
        convos = examples["messages"]
        texts = [
            merged_tokenizer.apply_chat_template(convo, tokenize = False, add_generation_prompt = False)
            for convo in convos
        ]
        return {"text": texts}

    dataset_ppl = dataset_ppl.map(formatting_prompts_func, batched = True)

    ppl_value = ppl_model(merged_model, merged_tokenizer, dataset_ppl)

    # Coerce to a plain Python float for cross-process transfer.
    if torch.is_tensor(ppl_value):
        ppl_value = ppl_value.cpu().item()
    elif hasattr(ppl_value, "item"):
        ppl_value = ppl_value.item()
    else:
        ppl_value = float(ppl_value)

    result_queue.put(ppl_value)

    del merged_model
    del merged_tokenizer
    del dataset_ppl
    torch.cuda.empty_cache()
    gc.collect()


if __name__ == "__main__":
    mp.set_start_method("spawn", force = True)

    from unsloth import is_bfloat16_supported
    from unsloth.models._utils import HAS_FLASH_ATTENTION

    compute_dtype = torch.bfloat16 if is_bfloat16_supported() else torch.float16
    attn_implementation = "flash_attention_2" if HAS_FLASH_ATTENTION else "sdpa"

    model, tokenizer = FastLanguageModel.from_pretrained(
        model_name = "unsloth/Llama-3.2-3B-Instruct",
        max_seq_length = 2048,
        dtype = compute_dtype,
        load_in_4bit = True,
        load_in_8bit = False,
        full_finetuning = False,
        attn_implementation = attn_implementation,
    )

    tokenizer = get_chat_template(
        tokenizer,
        chat_template = "llama-3.1",
    )

    from unsloth.chat_templates import standardize_sharegpt

    dataset_train = load_dataset("allenai/openassistant-guanaco-reformatted", split = "train")
    dataset_ppl = load_dataset("allenai/openassistant-guanaco-reformatted", split = "eval")

    dataset_train = dataset_train.map(formatting_prompts_func, batched = True)
    dataset_ppl = dataset_ppl.map(formatting_prompts_func, batched = True)

    add_to_comparison("Base model 4 bits", ppl_model(model, tokenizer, dataset_ppl))

    model = FastLanguageModel.get_peft_model(
        model,
        r = 16,
        target_modules = [
            "k_proj",
            "q_proj",
            "v_proj",
            "o_proj",
            "gate_proj",
            "down_proj",
            "up_proj",
        ],
        lora_alpha = 16,
        lora_dropout = 0,
        bias = "none",
        use_gradient_checkpointing = "unsloth",
        random_state = 3407,
        use_rslora = False,
        loftq_config = None,
    )

    from unsloth import is_bfloat16_supported

    trainer = SFTTrainer(
        model = model,
        tokenizer = tokenizer,
        train_dataset = dataset_train,
        dataset_text_field = "text",
        max_seq_length = 2048,
        data_collator = DataCollatorForSeq2Seq(tokenizer = tokenizer),
        dataset_num_proc = 2,
        packing = False,
        args = TrainingArguments(
            per_device_train_batch_size = 2,
            gradient_accumulation_steps = 4,
            warmup_ratio = 0.1,
            max_steps = 10,
            learning_rate = 2e-4,
            fp16 = not is_bfloat16_supported(),
            bf16 = is_bfloat16_supported(),
            logging_steps = 50,
            optim = "adamw_8bit",
            lr_scheduler_type = "linear",
            seed = 3407,
            output_dir = "outputs",
            report_to = "none",
        ),
    )

    from unsloth.chat_templates import train_on_responses_only

    trainer = train_on_responses_only(
        trainer,
        instruction_part = "<|start_header_id|>user<|end_header_id|>\n\n",
        response_part = "<|start_header_id|>assistant<|end_header_id|>\n\n",
    )

    trainer_stats = trainer.train()

    add_to_comparison("Qlora model", ppl_model(model, tokenizer, dataset_ppl))

    print("merge and save to local disk")
    model.save_pretrained_merged(
        save_directory = "./unsloth_out/merged_llama_text_model", tokenizer = tokenizer
    )

    # print("cleaning")
    # del model
    # del tokenizer
    # torch.cuda.empty_cache()
    # gc.collect()

    print("Loading merged model in 4 bit for perplexity test")
    merged_model, merged_tokenizer = FastLanguageModel.from_pretrained(
        model_name = "./unsloth_out/merged_llama_text_model",
        max_seq_length = 2048,
        load_in_4bit = True,
        load_in_8bit = False,
    )

    add_to_comparison(
        "merged model load 4bit", ppl_model(merged_model, merged_tokenizer, dataset_ppl)
    )

    print("Computing 8-bit model perplexity in subprocess...")
    result_queue = mp.Queue()
    p = mp.Process(target = load_and_compute_8bit_ppl, args = (result_queue, False, True))
    p.start()
    p.join()

    ppl_8bit = result_queue.get()
    add_to_comparison("merged model loaded 8bits", ppl_8bit)

    print("Loading merged model in 16 bit for perplexity test")
    merged_model, merged_tokenizer = FastLanguageModel.from_pretrained(
        model_name = "./unsloth_out/merged_llama_text_model",
        max_seq_length = 2048,
        load_in_4bit = False,
        load_in_8bit = False,
    )

    add_to_comparison(
        "merged model loaded 16bits",
        ppl_model(merged_model, merged_tokenizer, dataset_ppl),
    )

    print_model_comparison()

    safe_remove_directory("./outputs")
    safe_remove_directory("./unsloth_compiled_cache")
    safe_remove_directory("./unsloth_out")