while os.path.basename(__exec_dir) != "src":
__exec_dir = os.path.dirname(__exec_dir)
sys.path.insert(0, __exec_dir)
from datasets.common.utils import get_dataset
from utils import ArgumentList, Run
from utils.common.files import get_full_path
from utils.common.logging import logging_info
from utils.common.terminal import query_yes_no
from utils.tfu import tfu_load_graph, tfu_set_logging
from utils.visualization import draw_single_box, get_distinct_colors
if __name__ == "__main__":
# parse arguments
argument_list = ArgumentList(
description="Apply an exported model on a single image.")
argument_list.add_image_filename_argument("The input image filename.",
required=True)
argument_list.add_model_argument("The model used for training.",
default=None,
required=True)
argument_list.add_model_name_argument("The exported model name.",
required=True)
argument_list.add_dataset_argument("The dataset used for training.",
default=None)
argument_list.add_tf_verbosity_argument("Tensorflow verbosity.",
default="info")
argument_list.add_tf_min_log_level_argument(
"Tensorflow minimum log level.", default=3)
arguments = argument_list.parse()
# update image
image.set_data(frame)
def handle_key_press_event(event):
"""Handle any key press events in the Matplotlib window.
"""
if event.key == "q":
plt.close(event.canvas.figure)
if __name__ == "__main__":
# parse arguments
argument_list = ArgumentList(
description=
"Apply an exported model on live data like a video or a webcam.")
argument_list.add_video_filename_argument("The input video filename.",
required=False)
argument_list.add_model_argument("The model used for training.",
default=None,
required=True)
argument_list.add_model_name_argument("The exported model name.",
required=True)
argument_list.add_dataset_argument("The dataset used for training.",
default=None)
argument_list.add_tf_verbosity_argument("Tensorflow verbosity.",
default="info")
argument_list.add_tf_min_log_level_argument(
"Tensorflow minimum log level.", default=3)
arguments = argument_list.parse()
# skip if run does not exist
if not os.path.exists(run.base_path):
return
# ask user to keep the run
should_keep_run = query_yes_no("Should the run '{}' be kept?".format(
os.path.basename(run.base_path)),
default="yes")
if not should_keep_run:
shutil.rmtree(run.base_path)
if __name__ == "__main__":
# parse arguments
argument_list = ArgumentList(
description="Train a model for object detection.")
argument_list.add_model_argument("The model to use for training.",
default="ssd_vgg_300")
argument_list.add_dataset_argument("The dataset to use for training.",
default="voc2007")
argument_list.add_dataset_split_argument(
"The dataset split to use for training.",
default="train",
required=False)
argument_list.add_random_seed_argument(
"The global random seed used for determinism.", default=1807241)
argument_list.add_op_random_seed_argument(
"The operation random seed used for determinism.", default=1807242)
argument_list.add_num_parallel_calls_argument(
"Number of parallel calls for preprocessing the data.", default=6)
argument_list.add_prefetch_buffer_size_argument(
__exec_dir = os.path.dirname(__exec_dir)
sys.path.insert(0, __exec_dir)
from data import DataProvider
from data.preprocessors import BBoxPreprocessor, DefaultPreprocessor, ImagePreprocessor
from datasets.common.utils import get_dataset
from models.ssd.common.utils import get_model
from utils import ArgumentList, AveragePrecision, Run
from utils.common.logging import logging_error, logging_info, logging_eval
from utils.common.terminal import query_yes_no
from utils.tfu import tfu_get_uninitialized_variables, tfu_set_logging
if __name__ == "__main__":
# parse arguments
argument_list = ArgumentList(
description="Evaluate model from a run and compute mAP.")
argument_list.add_run_argument("The run from which to evaluate the model.",
required=True)
argument_list.add_dataset_argument("The dataset to use for evaluation.",
default="voc2007")
argument_list.add_dataset_split_argument(
"The dataset split to use for evaluation.", default="test")
argument_list.add_num_parallel_calls_argument(
"Number of parallel calls for preprocessing the data.", default=6)
argument_list.add_prefetch_buffer_size_argument(
"Buffer size for prefetching the data.", default=2)
argument_list.add_batch_size_argument("Batch size for evaluation.",
default=32)
argument_list.add_input_device_argument("Device for processing inputs.",
default="/cpu:0")
argument_list.add_inference_device_argument("Device for inference.",
__exec_dir = sys.path[0]
while os.path.basename(__exec_dir) != "src":
__exec_dir = os.path.dirname(__exec_dir)
sys.path.insert(0, __exec_dir)
from utils import ArgumentList, Run
from utils.common.files import get_full_path
from utils.common.logging import logging_error, logging_info
from utils.common.terminal import query_yes_no
from utils.tfu import tfu_set_logging
if __name__ == "__main__":
# parse arguments
argument_list = ArgumentList(
description="Export the model from a run for inference.")
argument_list.add_run_argument("The run from which to export the model.",
required=True)
argument_list.add_model_name_argument("The output model name.",
required=True)
argument_list.add_tf_verbosity_argument("Tensorflow verbosity.",
default="info")
argument_list.add_tf_min_log_level_argument(
"Tensorflow minimum log level.", default=3)
arguments = argument_list.parse()
# load run
run = Run(run_id=arguments.run)
if not run.open():
logging_error("There is no run '{}'.".format(arguments.run))