return samples
samples = []
for csv, init_op in zip(test_csvs, test_init_ops):
print('Testing model on {}'.format(csv))
samples.extend(run_test(init_op, dataset=csv))
return samples
def main(_):
initialize_globals()
if not FLAGS.test_files:
log_error('You need to specify what files to use for evaluation via '
'the --test_files flag.')
exit(1)
from DeepSpeech import create_model, try_loading # pylint: disable=cyclic-import
samples = evaluate(FLAGS.test_files.split(','), create_model, try_loading)
if FLAGS.test_output_file:
# Save decoded tuples as JSON, converting NumPy floats to Python floats
json.dump(samples, open(FLAGS.test_output_file, 'w'), default=float)
if __name__ == '__main__':
create_flags()
tf.app.flags.DEFINE_string('test_output_file', '', 'path to a file to save all src/decoded/distance/loss tuples')
tf.app.run(main)
if FLAGS.test:
test()
elif FLAGS.job_name == 'worker':
# We are a worker and therefore we have to do some work.
# Assigns ops to the local worker by default.
with tf.device(
tf.train.replica_device_setter(
worker_device=Config.worker_device,
cluster=Config.cluster)):
# Do the training
train(server)
log_debug('Server stopped.')
# Are we the main process?
if Config.is_chief:
# Doing solo/post-processing work just on the main process...
# Exporting the model
if FLAGS.export_dir:
export()
if len(FLAGS.one_shot_infer):
do_single_file_inference(FLAGS.one_shot_infer)
if __name__ == '__main__':
create_flags()
tf.app.run(main)
