def train(trained_models_dir, estimator_model_dir, training_chunk_dir,
generation, params):
"""Train the latest model from gathered data.
Args:
trained_models_dir: Where to export the completed generation.
estimator_model_dir: tf.estimator model directory.
training_chunk_dir: Directory where gathered training chunks are.
generation: Which generation you are training.
params: A MiniGoParams instance of hyperparameters for the model.
"""
new_model_name = utils.generate_model_name(generation)
print('New model will be {}'.format(new_model_name))
new_model = os.path.join(trained_models_dir, new_model_name)
print('Training on gathered game data...')
tf_records = sorted(
tf.gfile.Glob(os.path.join(training_chunk_dir, '*'+_TF_RECORD_SUFFIX)))
tf_records = tf_records[
-(params.train_window_size // params.examples_per_chunk):]
print('Training from: {} to {}'.format(tf_records[0], tf_records[-1]))
with utils.logged_timer('Training'):
dualnet.train(estimator_model_dir, tf_records, generation, params)
dualnet.export_model(estimator_model_dir, new_model)
def train(trained_models_dir, estimator_model_dir, training_chunk_dir,
generation, params):
"""Train the latest model from gathered data.
Args:
trained_models_dir: Where to export the completed generation.
estimator_model_dir: tf.estimator model directory.
training_chunk_dir: Directory where gathered training chunks are.
generation: Which generation you are training.
params: A MiniGoParams instance of hyperparameters for the model.
"""
new_model_name = utils.generate_model_name(generation)
print('New model will be {}'.format(new_model_name))
new_model = os.path.join(trained_models_dir, new_model_name)
print('Training on gathered game data...')
tf_records = sorted(
tf.gfile.Glob(os.path.join(training_chunk_dir, '*'+_TF_RECORD_SUFFIX)))
tf_records = tf_records[
-(params.train_window_size // params.examples_per_chunk):]
print('Training from: {} to {}'.format(tf_records[0], tf_records[-1]))
with utils.logged_timer('Training'):
dualnet.train(estimator_model_dir, tf_records, generation, params)
dualnet.export_model(estimator_model_dir, new_model)
def train(trained_models_dir, estimator_model_dir, training_chunk_dir, params):
"""Train the latest model from gathered data.
Args:
trained_models_dir: Where to export the completed generation.
estimator_model_dir: tf.estimator model directory.
training_chunk_dir: Directory where gathered training chunks are.
params: An object of hyperparameters for the model.
"""
model_num, model_name = utils.get_latest_model(trained_models_dir)
print('Initializing from model {}'.format(model_name))
new_model_name = utils.generate_model_name(model_num + 1)
print('New model will be {}'.format(new_model_name))
save_file = os.path.join(trained_models_dir, new_model_name)
tf_records = sorted(
tf.gfile.Glob(os.path.join(training_chunk_dir,
'*' + _TF_RECORD_SUFFIX)))
tf_records = tf_records[-(params.train_window_size //
params.examples_per_chunk):]
print('Training from: {} to {}'.format(tf_records[0], tf_records[-1]))
with utils.logged_timer('Training'):
dualnet.train(estimator_model_dir, tf_records, model_num + 1, params)
dualnet.export_model(estimator_model_dir, save_file)
def test_inference(self):
with tempfile.TemporaryDirectory() as working_dir, \
tempfile.TemporaryDirectory() as export_dir:
dualnet.bootstrap(working_dir, model_params.DummyMiniGoParams())
exported_model = os.path.join(export_dir, 'bootstrap-model')
dualnet.export_model(working_dir, exported_model)
n1 = dualnet.DualNetRunner(
exported_model, model_params.DummyMiniGoParams())
n1.run(go.Position(utils_test.BOARD_SIZE))
n2 = dualnet.DualNetRunner(
exported_model, model_params.DummyMiniGoParams())
n2.run(go.Position(utils_test.BOARD_SIZE))
def test_inference(self):
with tempfile.TemporaryDirectory() as working_dir, \
tempfile.TemporaryDirectory() as export_dir:
dualnet.bootstrap(working_dir, model_params.DummyMiniGoParams())
exported_model = os.path.join(export_dir, 'bootstrap-model')
dualnet.export_model(working_dir, exported_model)
n1 = dualnet.DualNetRunner(exported_model,
model_params.DummyMiniGoParams())
n1.run(go.Position(utils_test.BOARD_SIZE))
n2 = dualnet.DualNetRunner(exported_model,
model_params.DummyMiniGoParams())
n2.run(go.Position(utils_test.BOARD_SIZE))
def bootstrap(estimator_model_dir, trained_models_dir, params):
"""Initialize the model with random weights.
Args:
estimator_model_dir: tf.estimator model directory.
trained_models_dir: Dir to save the trained models. Here to export the first
bootstrapped generation.
params: A MiniGoParams instance of hyperparameters for the model.
"""
bootstrap_name = utils.generate_model_name(0)
_ensure_dir_exists(trained_models_dir)
bootstrap_model_path = os.path.join(trained_models_dir, bootstrap_name)
_ensure_dir_exists(estimator_model_dir)
print('Bootstrapping with working dir {}\n Model 0 exported to {}'.format(
estimator_model_dir, bootstrap_model_path))
dualnet.bootstrap(estimator_model_dir, params)
dualnet.export_model(estimator_model_dir, bootstrap_model_path)
def bootstrap(estimator_model_dir, trained_models_dir, params):
"""Initialize the model with random weights.
Args:
estimator_model_dir: tf.estimator model directory.
trained_models_dir: Dir to save the trained models. Here to export the first
bootstrapped generation.
params: A MiniGoParams instance of hyperparameters for the model.
"""
bootstrap_name = utils.generate_model_name(0)
_ensure_dir_exists(trained_models_dir)
bootstrap_model_path = os.path.join(trained_models_dir, bootstrap_name)
_ensure_dir_exists(estimator_model_dir)
print('Bootstrapping with working dir {}\n Model 0 exported to {}'.format(
estimator_model_dir, bootstrap_model_path))
dualnet.bootstrap(estimator_model_dir, params)
dualnet.export_model(estimator_model_dir, bootstrap_model_path)