xls-r-300m-lm-fr / eval_lm.py

pascal lim

update eval script with lm

9e32bde over 2 years ago

No virus

4.49 kB

	#!/usr/bin/env python3
	import argparse
	import re
	from typing import Dict

	from datasets import Audio, Dataset, load_dataset, load_metric
	import torch
	from transformers import AutoFeatureExtractor, pipeline, Wav2Vec2ProcessorWithLM


	def log_results(result: Dataset, args: Dict[str, str]):
	"""DO NOT CHANGE. This function computes and logs the result metrics."""

	log_outputs = args.log_outputs
	dataset_id = "_".join(args.dataset.split("/") + [args.config, args.split])

	# load metric
	wer = load_metric("wer")
	cer = load_metric("cer")

	# compute metrics
	wer_result = wer.compute(references=result["target"], predictions=result["prediction"])
	cer_result = cer.compute(references=result["target"], predictions=result["prediction"])

	# print & log results
	result_str = f"WER: {wer_result}\n" f"CER: {cer_result}"
	print(result_str)

	with open(f"{dataset_id}_eval_results.txt", "w") as f:
	f.write(result_str)

	# log all results in text file. Possibly interesting for analysis
	if log_outputs is not None:
	pred_file = f"log_{dataset_id}_predictions.txt"
	target_file = f"log_{dataset_id}_targets.txt"

	with open(pred_file, "w") as p, open(target_file, "w") as t:

	# mapping function to write output
	def write_to_file(batch, i):
	p.write(f"{i}" + "\n")
	p.write(batch["prediction"] + "\n")
	t.write(f"{i}" + "\n")
	t.write(batch["target"] + "\n")

	result.map(write_to_file, with_indices=True)


	def normalize_text(text: str) -> str:
	"""DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""

	chars_to_ignore_regex = '[^a-zàâäçéèêëîïôöùûüÿ\'’ ]' # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training

	text = re.sub(chars_to_ignore_regex, "", text.lower()).replace('’', "'")

	# In addition, we can normalize the target text, e.g. removing new lines characters etc...
	# note that order is important here!
	token_sequences_to_ignore = ["\n\n", "\n", " ", " "]

	for t in token_sequences_to_ignore:
	text = " ".join(text.split(t))

	return text

	def evaluate_with_lm(batch):
	inputs = processor(batch["audio"]["array"], sampling_rate=16_000, return_tensors="pt", padding=True)

	with torch.no_grad():
	logits = model(**inputs.to('cuda')).logits
	int_result = processor.batch_decode(logits.cpu().numpy())

	batch["prediction"] = int_result.text
	batch["target"] = normalize_text(batch["sentence"])

	del int_result
	torch.cuda.empty_cache()

	return batch

	def main(args):
	# load dataset
	dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
	# resample audio
	dataset = dataset.cast_column("audio", Audio(sampling_rate=16_000))

	# for testing: only process the first two examples as a test
	# dataset = dataset.select(range(10))

	# load processor
	processor = Wav2Vec2ProcessorWithLM.from_pretrained("./")

	model = Wav2Vec2ForCTC.from_pretrained("./")
	model.to('cuda')

	# run inference on all examples
	result = dataset.map(evaluate_with_lm, remove_columns=dataset.column_names)

	# compute and log_results
	# do not change function below
	log_results(result, args)


	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",
	type=str,
	required=True,
	help="Dataset name to evaluate the `model_id`. Should be loadable with 🤗 Datasets",
	)
	parser.add_argument(
	"--config", type=str, required=True, help="Config of the dataset. E.g. `'en'` for Common Voice"
	)
	parser.add_argument("--split", type=str, required=True, help="Split of the dataset. E.g. `'test'`")
	parser.add_argument(
	"--chunk_length_s", type=float, default=None, help="Chunk length in seconds. Defaults to 5 seconds."
	)
	parser.add_argument(
	"--stride_length_s", type=float, default=None, help="Stride of the audio chunks. Defaults to 1 second."
	)
	parser.add_argument(
	"--log_outputs", action="store_true", help="If defined, write outputs to log file for analysis."
	)
	args = parser.parse_args()

	main(args)