Gluoncv SSD coco models compilation takes 4 min 30 sec on i7

apivovarov · May 21, 2019, 7:38pm

I tried to run deploy_ssd_gluoncv.py tutorial code for ssd_512_resnet50_v1_coco and ssd_512_mobilenet1.0_coco models.

github.com

dmlc/tvm/blob/master/tutorials/frontend/deploy_ssd_gluoncv.py

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
"""
Deploy Single Shot Multibox Detector(SSD) model
===============================================
**Author**: `Yao Wang <https://github.com/kevinthesun>`_

This file has been truncated. show original

The compilation takes 4 min 30 sec. `target=‘llvm’, ctx=‘cpu(0)’
The last log lines before it hangs for 4 min are:

[12:34:08] /home/ubuntu/tvm/src/relay/op/type_relations.cc:111: In1:TensorType([1, 6132, 1], float32),In2:TensorType([1, 6132, 1], float32),Out:IncompleteTypeNode(0, 0x1e982ad0)

[12:34:08] /home/ubuntu/tvm/src/relay/op/type_relations.cc:111: In1:TensorType([1, 6132, 1], float32),In2:TensorType([1, 6132, 1], float32),Out:IncompleteTypeNode(0, 0x2388a0d0)

[12:34:08] /home/ubuntu/tvm/src/schedule/bound.cc:129: not in feed graph consumer = hybrid(hybrid_nms, 0x1f6ce710)
[12:34:08] /home/ubuntu/tvm/src/schedule/bound.cc:129: not in feed graph consumer = hybrid(hybrid_nms, 0x1f6ce710)