def main(
config: str = DEFAULT_YAML,
saved: str = None,
mxp: bool = False,
bs: int = None,
sentence_piece: bool = False,
subwords: bool = False,
device: int = 0,
cpu: bool = False,
output: str = "test.tsv",
):
assert saved and output
tf.random.set_seed(0)
tf.keras.backend.clear_session()
tf.config.optimizer.set_experimental_options({"auto_mixed_precision": mxp})
env_util.setup_devices([device], cpu=cpu)
config = Config(config)
speech_featurizer, text_featurizer = featurizer_helpers.prepare_featurizers(
config=config,
subwords=subwords,
sentence_piece=sentence_piece,
)
jasper = Jasper(**config.model_config,
vocabulary_size=text_featurizer.num_classes)
jasper.make(speech_featurizer.shape)
jasper.load_weights(saved, by_name=True)
jasper.summary(line_length=100)
jasper.add_featurizers(speech_featurizer, text_featurizer)
test_dataset = dataset_helpers.prepare_testing_datasets(
config=config,
speech_featurizer=speech_featurizer,
text_featurizer=text_featurizer)
batch_size = bs or config.learning_config.running_config.batch_size
test_data_loader = test_dataset.create(batch_size)
exec_helpers.run_testing(model=jasper,
test_dataset=test_dataset,
test_data_loader=test_data_loader,
output=output)
if args.tfrecords:
test_dataset = ASRTFRecordDataset(
speech_featurizer=speech_featurizer,
text_featurizer=text_featurizer,
**vars(config.learning_config.test_dataset_config))
else:
test_dataset = ASRSliceDataset(
speech_featurizer=speech_featurizer,
text_featurizer=text_featurizer,
**vars(config.learning_config.test_dataset_config))
# build model
jasper = Jasper(**config.model_config,
vocabulary_size=text_featurizer.num_classes)
jasper.make(speech_featurizer.shape)
jasper.load_weights(args.saved)
jasper.summary(line_length=100)
jasper.add_featurizers(speech_featurizer, text_featurizer)
batch_size = args.bs or config.learning_config.running_config.batch_size
test_data_loader = test_dataset.create(batch_size)
with file_util.save_file(file_util.preprocess_paths(args.output)) as filepath:
overwrite = True
if tf.io.gfile.exists(filepath):
overwrite = input(
f"Overwrite existing result file {filepath} ? (y/n): ").lower(
) == "y"
if overwrite:
results = jasper.predict(test_data_loader, verbose=1)
print(f"Saving result to {args.output} ...")
assert args.saved and args.output
config = Config(args.config)
speech_featurizer = TFSpeechFeaturizer(config.speech_config)
if args.subwords:
text_featurizer = SubwordFeaturizer(config.decoder_config)
else:
text_featurizer = CharFeaturizer(config.decoder_config)
# build model
jasper = Jasper(**config.model_config,
vocabulary_size=text_featurizer.num_classes)
jasper.make(speech_featurizer.shape)
jasper.load_weights(args.saved, by_name=True)
jasper.summary(line_length=100)
jasper.add_featurizers(speech_featurizer, text_featurizer)
concrete_func = jasper.make_tflite_function().get_concrete_function()
converter = tf.lite.TFLiteConverter.from_concrete_functions([concrete_func])
converter.experimental_new_converter = True
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.target_spec.supported_ops = [
tf.lite.OpsSet.TFLITE_BUILTINS, tf.lite.OpsSet.SELECT_TF_OPS
]
tflite_model = converter.convert()
args.output = file_util.preprocess_paths(args.output)
with open(args.output, "wb") as tflite_out:
tflite_out.write(tflite_model)
def main(
config: str = DEFAULT_YAML,
tfrecords: bool = False,
sentence_piece: bool = False,
subwords: bool = False,
bs: int = None,
spx: int = 1,
metadata: str = None,
static_length: bool = False,
devices: list = [0],
mxp: bool = False,
pretrained: str = None,
):
tf.keras.backend.clear_session()
tf.config.optimizer.set_experimental_options({"auto_mixed_precision": mxp})
strategy = env_util.setup_strategy(devices)
config = Config(config)
speech_featurizer, text_featurizer = featurizer_helpers.prepare_featurizers(
config=config,
subwords=subwords,
sentence_piece=sentence_piece,
)
train_dataset, eval_dataset = dataset_helpers.prepare_training_datasets(
config=config,
speech_featurizer=speech_featurizer,
text_featurizer=text_featurizer,
tfrecords=tfrecords,
metadata=metadata,
)
if not static_length:
speech_featurizer.reset_length()
text_featurizer.reset_length()
train_data_loader, eval_data_loader, global_batch_size = dataset_helpers.prepare_training_data_loaders(
config=config,
train_dataset=train_dataset,
eval_dataset=eval_dataset,
strategy=strategy,
batch_size=bs,
)
with strategy.scope():
jasper = Jasper(**config.model_config, vocabulary_size=text_featurizer.num_classes)
jasper.make(speech_featurizer.shape, batch_size=global_batch_size)
if pretrained:
jasper.load_weights(pretrained, by_name=True, skip_mismatch=True)
jasper.summary(line_length=100)
jasper.compile(
optimizer=config.learning_config.optimizer_config,
experimental_steps_per_execution=spx,
global_batch_size=global_batch_size,
blank=text_featurizer.blank,
)
callbacks = [
tf.keras.callbacks.ModelCheckpoint(**config.learning_config.running_config.checkpoint),
tf.keras.callbacks.experimental.BackupAndRestore(config.learning_config.running_config.states_dir),
tf.keras.callbacks.TensorBoard(**config.learning_config.running_config.tensorboard),
]
jasper.fit(
train_data_loader,
epochs=config.learning_config.running_config.num_epochs,
validation_data=eval_data_loader,
callbacks=callbacks,
steps_per_epoch=train_dataset.total_steps,
validation_steps=eval_dataset.total_steps if eval_data_loader else None,
)
speech_featurizer.reset_length()
text_featurizer.reset_length()
global_batch_size = args.bs or config.learning_config.running_config.batch_size
global_batch_size *= strategy.num_replicas_in_sync
train_data_loader = train_dataset.create(global_batch_size)
eval_data_loader = eval_dataset.create(global_batch_size)
with strategy.scope():
# build model
jasper = Jasper(**config.model_config,
vocabulary_size=text_featurizer.num_classes)
jasper.make(speech_featurizer.shape, batch_size=global_batch_size)
if args.pretrained:
jasper.load_weights(args.pretrained, by_name=True, skip_mismatch=True)
jasper.summary(line_length=100)
jasper.compile(optimizer=config.learning_config.optimizer_config,
experimental_steps_per_execution=args.spx,
global_batch_size=global_batch_size,
blank=text_featurizer.blank)
callbacks = [
tf.keras.callbacks.ModelCheckpoint(
**config.learning_config.running_config.checkpoint),
tf.keras.callbacks.experimental.BackupAndRestore(
config.learning_config.running_config.states_dir),
tf.keras.callbacks.TensorBoard(
**config.learning_config.running_config.tensorboard)
]