phi/run_eval.py

import argparse
import os
import torch
from transformers import AutoModelForCausalLM, AutoProcessor, StoppingCriteria, StoppingCriteriaList
import evaluate
from normalizer import data_utils
import time
from tqdm import tqdm

wer_metric = evaluate.load("wer")
torch.set_float32_matmul_precision('high')

class MultipleTokenBatchStoppingCriteria(StoppingCriteria):
    """Stopping criteria capable of receiving multiple stop-tokens and handling batched inputs."""

    def __init__(self, stop_tokens: torch.LongTensor, batch_size: int = 1) -> None:
        """Initialize the multiple token batch stopping criteria.

        Args:
            stop_tokens: Stop-tokens.
            batch_size: Batch size.

        """

        self.stop_tokens = stop_tokens
        self.max_stop_tokens = stop_tokens.shape[-1]
        self.stop_tokens_idx = torch.zeros(batch_size, dtype=torch.long, device=stop_tokens.device)

    def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor, **kwargs) -> bool:
        # Only gather the maximum number of inputs compatible with stop tokens
        # and checks whether generated inputs are equal to `stop_tokens`
        generated_inputs = torch.eq(input_ids[:, -self.max_stop_tokens :].unsqueeze(1), self.stop_tokens)
        equal_generated_inputs = torch.all(generated_inputs, dim=2)

        # Mark the position where a stop token has been produced for each input in the batch,
        # but only if the corresponding entry is not already set
        sequence_idx = torch.any(equal_generated_inputs, dim=1)
        sequence_set_mask = self.stop_tokens_idx == 0
        self.stop_tokens_idx[sequence_idx & sequence_set_mask] = input_ids.shape[-1]

        return torch.all(self.stop_tokens_idx)


def main(args):
    model = AutoModelForCausalLM.from_pretrained(
        args.model_id,
        trust_remote_code=True,
        torch_dtype="auto",
        _attn_implementation="flash_attention_2",
    ).to(args.device)
    model.eval()
    processor = AutoProcessor.from_pretrained(args.model_id, trust_remote_code=True)

    user = "<|user|>"
    assistant = "<|assistant|>"
    prompt_suffix = "<|end|>"

    prompt = f"{user}<|audio_1|>{args.user_prompt}{prompt_suffix}{assistant}"

    gen_kwargs = {"max_new_tokens": args.max_new_tokens, "num_beams": args.num_beams}

    stop_tokens = [prompt_suffix, processor.tokenizer.eos_token]
    stop_tokens_ids = processor.tokenizer(stop_tokens, add_special_tokens=False, padding="longest", return_tensors="pt")["input_ids"]
    stop_tokens_ids = stop_tokens_ids.to(model.device)

    def benchmark(batch, min_new_tokens=None):
        # Load audio inputs
        audios = [(audio["array"], audio["sampling_rate"]) for audio in batch["audio"]]
        minibatch_size = len(audios)
        gen_kwargs["stopping_criteria"] = StoppingCriteriaList(
            [MultipleTokenBatchStoppingCriteria(stop_tokens_ids, batch_size=args.num_beams * minibatch_size)]
        )

        # START TIMING
        start_time = time.time()

        with torch.autocast(model.device.type, enabled=True):
            inputs = processor(text=[prompt] * minibatch_size, audios=audios, return_tensors="pt").to(args.device)

            # Model Inference
            pred_ids = model.generate(
                **inputs,
                pad_token_id=processor.tokenizer.pad_token_id,
                eos_token_id=processor.tokenizer.eos_token_id,
                **gen_kwargs,
                min_new_tokens=min_new_tokens,
            )

        # Gather the sequence index of the stop token
        stop_tokens_idx = gen_kwargs["stopping_criteria"][0].stop_tokens_idx.reshape(minibatch_size, -1)[:, 0]

        # If a stop token was produced, we need to remove its length from the found index,
        # however there might be a chance that the stop token was not produced and the index
        # returned is the length of the generated sequence
        stop_tokens_idx = torch.where(
            stop_tokens_idx > 0,
            stop_tokens_idx - stop_tokens_ids.shape[-1],
            pred_ids.shape[-1],
        )

        # Convert token ids to text transcription
        pred_text = [
            processor.decode(_pred_ids[inputs["input_ids"].shape[1] : _stop_tokens_idx], skip_special_tokens=True, clean_up_tokenization_spaces=False)
            for _pred_ids, _stop_tokens_idx in zip(pred_ids, stop_tokens_idx)
        ]

        # END TIMING
        runtime = time.time() - start_time

        # normalize by minibatch size since we want the per-sample time
        batch["transcription_time_s"] = minibatch_size * [runtime / minibatch_size]

        # normalize transcriptions with English normalizer
        batch["predictions"] = [data_utils.normalizer(pred) for pred in pred_text]
        batch["references"] = batch["norm_text"]
        return batch

    if args.warmup_steps is not None:
        dataset = data_utils.load_data(args)
        dataset = data_utils.prepare_data(dataset)

        num_warmup_samples = args.warmup_steps * args.batch_size
        if args.streaming:
            warmup_dataset = dataset.take(num_warmup_samples)
        else:
            warmup_dataset = dataset.select(range(min(num_warmup_samples, len(dataset))))
        warmup_dataset = iter(warmup_dataset.map(benchmark, batch_size=args.batch_size, batched=True, fn_kwargs={"min_new_tokens": args.max_new_tokens}))

        for _ in tqdm(warmup_dataset, desc="Warming up..."):
            continue

    dataset = data_utils.load_data(args)
    if args.max_eval_samples is not None and args.max_eval_samples > 0:
        print(f"Subsampling dataset to first {args.max_eval_samples} samples!")
        if args.streaming:
            dataset = dataset.take(args.max_eval_samples)
        else:
            dataset = dataset.select(range(min(args.max_eval_samples, len(dataset))))
    dataset = data_utils.prepare_data(dataset)

    dataset = dataset.map(
        benchmark, batch_size=args.batch_size, batched=True, remove_columns=["audio"],
    )

    all_results = {
        "audio_length_s": [],
        "transcription_time_s": [],
        "predictions": [],
        "references": [],
    }
    result_iter = iter(dataset)
    for result in tqdm(result_iter, desc="Samples..."):
        for key in all_results:
            all_results[key].append(result[key])

    # Write manifest results (WER and RTFX)
    manifest_path = data_utils.write_manifest(
        all_results["references"],
        all_results["predictions"],
        args.model_id,
        args.dataset_path,
        args.dataset,
        args.split,
        audio_length=all_results["audio_length_s"],
        transcription_time=all_results["transcription_time_s"],
    )
    print("Results saved at path:", os.path.abspath(manifest_path))

    wer = wer_metric.compute(
        references=all_results["references"], predictions=all_results["predictions"]
    )
    wer = round(100 * wer, 2)
    rtfx = round(sum(all_results["audio_length_s"]) / sum(all_results["transcription_time_s"]), 2)
    print("WER:", wer, "%", "RTFx:", rtfx)


if __name__ == "__main__":
    parser = argparse.ArgumentParser()

    parser.add_argument(
        "--model_id",
        type=str,
        required=True,
        help="Model identifier. Should be loadable with 🤗 Transformers",
    )
    parser.add_argument(
        "--dataset_path",
        type=str,
        default="esb/datasets",
        help="Dataset path. By default, it is `esb/datasets`",
    )
    parser.add_argument(
        "--dataset",
        type=str,
        required=True,
        help="Dataset name. *E.g.* `'librispeech_asr` for the LibriSpeech ASR dataset, or `'common_voice'` for Common Voice. The full list of dataset names "
        "can be found at `https://huggingface.co/datasets/esb/datasets`",
    )
    parser.add_argument(
        "--split",
        type=str,
        default="test",
        help="Split of the dataset. *E.g.* `'validation`' for the dev split, or `'test'` for the test split.",
    )
    parser.add_argument(
        "--device",
        type=int,
        default=-1,
        help="The device to run the pipeline on. -1 for CPU (default), 0 for the first GPU and so on.",
    )
    parser.add_argument(
        "--batch_size",
        type=int,
        default=16,
        help="Number of samples to go through each streamed batch.",
    )
    parser.add_argument(
        "--num_beams",
        type=int,
        default=1,
        help="Number of beams for beam search.",
    )
    parser.add_argument(
        "--max_eval_samples",
        type=int,
        default=None,
        help="Number of samples to be evaluated. Put a lower number e.g. 64 for testing this script.",
    )
    parser.add_argument(
        "--no-streaming",
        dest="streaming",
        action="store_false",
        help="Choose whether you'd like to download the entire dataset or stream it during the evaluation.",
    )
    parser.add_argument(
        "--max_new_tokens",
        type=int,
        default=None,
        help="Maximum number of tokens to generate (for auto-regressive models).",
    )
    parser.add_argument(
        "--warmup_steps",
        type=int,
        default=2,
        help="Number of warm-up steps to run before launching the timed runs.",
    )
    parser.add_argument(
        "--user_prompt",
        type=str,
        default="Transcribe the audio clip into text.",
        help="User prompt string.",
    )
    args = parser.parse_args()
    parser.set_defaults(streaming=False)

    main(args)