8.1.1.9.2.3. blueoil.utils.executor

8.1.1.9.2.3.1. Module Contents

8.1.1.9.2.3.1.1. Functions

init_logging(config)

Init tensorflow logging level.

prepare_dirs(recreate=False)

Prepare config dirs

search_restore_filename(checkpoints_dir)

convert_variables_to_constants(sess, output_node_names=[‘output’])

save_pb_file(sess, output_dir, output_node_names=[‘output’], pb_name=’minimal_graph_with_shape.pb’)

metrics_summary_op(metrics_ops_dict)

Create summary_op for training metrics.

profile_train_step(step, sess, run_meta)

blueoil.utils.executor.init_logging(config)

Init tensorflow logging level.

https://github.com/tensorflow/tensorflow/blob/be52c5c09e39ac2df007fb2d62abe122d5ade6d0/tensorflow/core/platform/default/logging.h#L30

blueoil.utils.executor.prepare_dirs(recreate=False)

Prepare config dirs

When recreate is True, if previous execution exists, remove them and recreate. When recreate is False, remain previous execution.

blueoil.utils.executor.search_restore_filename(checkpoints_dir)
blueoil.utils.executor.convert_variables_to_constants(sess, output_node_names=['output'])
blueoil.utils.executor.save_pb_file(sess, output_dir, output_node_names=['output'], pb_name='minimal_graph_with_shape.pb')
blueoil.utils.executor.metrics_summary_op(metrics_ops_dict)

Create summary_op for training metrics.

Parameters

metrics_ops_dict (dict) – dict of name and metrics_op.

Returns

summary op of metrics.

Return type

metrics_summary_op

blueoil.utils.executor.profile_train_step(step, sess, run_meta)