def __init__(self, tarball_path):
# """Creates and loads pretrained deeplab model."""
self.graph = tf.Graph()
graph_def = None
# Extract frozen graph from tar archive.
tar_file = tarfile.open(tarball_path)
for tar_info in tar_file.getmembers():
if self.FROZEN_GRAPH_NAME in os.path.basename(tar_info.name):
file_handle = tar_file.extractfile(tar_info)
graph_def = GraphDef.FromString(file_handle.read())
break
tar_file.close()
if graph_def is None:
raise RuntimeError('Cannot find inference graph in tar archive.')
with self.graph.as_default():
tf.import_graph_def(graph_def, name = '')
self.sess = Session(graph = self.graph)
executor.benchmark_layout_transform(min_exec_num=exec_num)
executor.run()
executor.write_opt_sch2record_file(opt_sch_file)
if __name__ == '__main__':
# logging.getLogger('autotvm').setLevel(logging.DEBUG)
# 1.load img
img_path = '/opt/dataset/tr2_cropped/data/1.png'
image = Image.open(img_path).resize((1024, 1024))
x = np.array(image)
# 2.load graph
GRAPH_PB_PATH = './frozen'
# graph_def = tf.compat.v1.get_default_graph().as_graph_def(add_shapes=True)
with tf.io.gfile.GFile('./frozen/frozen_model_fixed.pb', 'rb') as f:
graph_def = GraphDef.FromString(f.read())
#call the utility to import the graph definition into default graph.
graph_def = tf_testing.ProcessGraphDefParam(graph_def)
# 3. tvm frontend
shape_dict = {"input_1": (1, 1024, 1024, 3)} # change shape
mod, params = relay.frontend.from_tensorflow(graph_def, shape_dict)
# meta-data部份
data_shape = (1, 1024, 1024, 3) # maybe
output_shape = (1, 1024, 1024, 4)
batch_size = 1
dtype = "float32"
model_name = "unet_cpu_12_thread"
log_file = "%s.log" % model_name
graph_opt_sch_file = "%s_graph_opt_1000.log" % model_name
input_name = "x" #这是和后面的输入名字一样的