menoh 1.0.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +7 -0
- data/.gitignore +55 -0
- data/CODE_OF_CONDUCT.md +74 -0
- data/Dockerfile +49 -0
- data/Gemfile +5 -0
- data/LICENSE +21 -0
- data/README.md +80 -0
- data/Rakefile +18 -0
- data/Vagrantfile +46 -0
- data/bin/console +10 -0
- data/bin/setup +6 -0
- data/docs/tutorial.md +121 -0
- data/example/data/0.png +0 -0
- data/example/data/1.png +0 -0
- data/example/data/2.png +0 -0
- data/example/data/3.png +0 -0
- data/example/data/4.png +0 -0
- data/example/data/5.png +0 -0
- data/example/data/6.png +0 -0
- data/example/data/7.png +0 -0
- data/example/data/8.png +0 -0
- data/example/data/9.png +0 -0
- data/example/data/mnist.onnx +0 -0
- data/example/example_mnist.rb +74 -0
- data/example/example_mnist_with_block.rb +75 -0
- data/example/example_vgg16.rb +89 -0
- data/ext/menoh_native/extconf.rb +14 -0
- data/ext/menoh_native/menoh_ruby.c +328 -0
- data/ext/menoh_native/menoh_ruby.h +7 -0
- data/lib/menoh.rb +87 -0
- data/lib/menoh/version.rb +3 -0
- data/menoh.gemspec +29 -0
- metadata +146 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: ffe56775dcbde74f64fdeed4fa59bf6fcc539803
|
4
|
+
data.tar.gz: cc676fe5e5d2f2a2e591125fc6dcdfc9c4054962
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 6b1fb84c56bf469d76f97c86e0a0b0b0a832f146ff3ca52224d4983c429720c6dcb3c0191ca87fc1fb925234ea2d04bda2960a280265e52b7d790e270d73df4d
|
7
|
+
data.tar.gz: 62f09585697c3d536d55fef32f0477b289cb0593786e97ae7820759b87fc749d72446d8434abf23550c626e5737fc5e999c4375a4fa46807c31d85ddeeea3b14
|
data/.gitignore
ADDED
@@ -0,0 +1,55 @@
|
|
1
|
+
*.gem
|
2
|
+
*.rbc
|
3
|
+
/.vagrant/
|
4
|
+
/.config
|
5
|
+
/coverage/
|
6
|
+
/external/
|
7
|
+
/InstalledFiles
|
8
|
+
/pkg/
|
9
|
+
/spec/reports/
|
10
|
+
/spec/examples.txt
|
11
|
+
/test/tmp/
|
12
|
+
/test/version_tmp/
|
13
|
+
/tmp/
|
14
|
+
|
15
|
+
# Used by dotenv library to load environment variables.
|
16
|
+
# .env
|
17
|
+
|
18
|
+
## Specific to RubyMotion:
|
19
|
+
.dat*
|
20
|
+
.repl_history
|
21
|
+
build/
|
22
|
+
*.bridgesupport
|
23
|
+
build-iPhoneOS/
|
24
|
+
build-iPhoneSimulator/
|
25
|
+
|
26
|
+
## Specific to RubyMotion (use of CocoaPods):
|
27
|
+
#
|
28
|
+
# We recommend against adding the Pods directory to your .gitignore. However
|
29
|
+
# you should judge for yourself, the pros and cons are mentioned at:
|
30
|
+
# https://guides.cocoapods.org/using/using-cocoapods.html#should-i-check-the-pods-directory-into-source-control
|
31
|
+
#
|
32
|
+
# vendor/Pods/
|
33
|
+
|
34
|
+
## Documentation cache and generated files:
|
35
|
+
/.yardoc/
|
36
|
+
/_yardoc/
|
37
|
+
/doc/
|
38
|
+
/rdoc/
|
39
|
+
|
40
|
+
## Environment normalization:
|
41
|
+
/.bundle/
|
42
|
+
/vendor/bundle
|
43
|
+
/lib/bundler/man/
|
44
|
+
|
45
|
+
# for a library or gem, you might want to ignore these files since the code is
|
46
|
+
# intended to run in multiple environments; otherwise, check them in:
|
47
|
+
# Gemfile.lock
|
48
|
+
# .ruby-version
|
49
|
+
# .ruby-gemset
|
50
|
+
|
51
|
+
# unless supporting rvm < 1.11.0 or doing something fancy, ignore this:
|
52
|
+
.rvmrc
|
53
|
+
|
54
|
+
*.so
|
55
|
+
*.log
|
data/CODE_OF_CONDUCT.md
ADDED
@@ -0,0 +1,74 @@
|
|
1
|
+
# Contributor Covenant Code of Conduct
|
2
|
+
|
3
|
+
## Our Pledge
|
4
|
+
|
5
|
+
In the interest of fostering an open and welcoming environment, we as
|
6
|
+
contributors and maintainers pledge to making participation in our project and
|
7
|
+
our community a harassment-free experience for everyone, regardless of age, body
|
8
|
+
size, disability, ethnicity, gender identity and expression, level of experience,
|
9
|
+
nationality, personal appearance, race, religion, or sexual identity and
|
10
|
+
orientation.
|
11
|
+
|
12
|
+
## Our Standards
|
13
|
+
|
14
|
+
Examples of behavior that contributes to creating a positive environment
|
15
|
+
include:
|
16
|
+
|
17
|
+
* Using welcoming and inclusive language
|
18
|
+
* Being respectful of differing viewpoints and experiences
|
19
|
+
* Gracefully accepting constructive criticism
|
20
|
+
* Focusing on what is best for the community
|
21
|
+
* Showing empathy towards other community members
|
22
|
+
|
23
|
+
Examples of unacceptable behavior by participants include:
|
24
|
+
|
25
|
+
* The use of sexualized language or imagery and unwelcome sexual attention or
|
26
|
+
advances
|
27
|
+
* Trolling, insulting/derogatory comments, and personal or political attacks
|
28
|
+
* Public or private harassment
|
29
|
+
* Publishing others' private information, such as a physical or electronic
|
30
|
+
address, without explicit permission
|
31
|
+
* Other conduct which could reasonably be considered inappropriate in a
|
32
|
+
professional setting
|
33
|
+
|
34
|
+
## Our Responsibilities
|
35
|
+
|
36
|
+
Project maintainers are responsible for clarifying the standards of acceptable
|
37
|
+
behavior and are expected to take appropriate and fair corrective action in
|
38
|
+
response to any instances of unacceptable behavior.
|
39
|
+
|
40
|
+
Project maintainers have the right and responsibility to remove, edit, or
|
41
|
+
reject comments, commits, code, wiki edits, issues, and other contributions
|
42
|
+
that are not aligned to this Code of Conduct, or to ban temporarily or
|
43
|
+
permanently any contributor for other behaviors that they deem inappropriate,
|
44
|
+
threatening, offensive, or harmful.
|
45
|
+
|
46
|
+
## Scope
|
47
|
+
|
48
|
+
This Code of Conduct applies both within project spaces and in public spaces
|
49
|
+
when an individual is representing the project or its community. Examples of
|
50
|
+
representing a project or community include using an official project e-mail
|
51
|
+
address, posting via an official social media account, or acting as an appointed
|
52
|
+
representative at an online or offline event. Representation of a project may be
|
53
|
+
further defined and clarified by project maintainers.
|
54
|
+
|
55
|
+
## Enforcement
|
56
|
+
|
57
|
+
Instances of abusive, harassing, or otherwise unacceptable behavior may be
|
58
|
+
reported by contacting the project team at miyoshik@preferred.jp. All
|
59
|
+
complaints will be reviewed and investigated and will result in a response that
|
60
|
+
is deemed necessary and appropriate to the circumstances. The project team is
|
61
|
+
obligated to maintain confidentiality with regard to the reporter of an incident.
|
62
|
+
Further details of specific enforcement policies may be posted separately.
|
63
|
+
|
64
|
+
Project maintainers who do not follow or enforce the Code of Conduct in good
|
65
|
+
faith may face temporary or permanent repercussions as determined by other
|
66
|
+
members of the project's leadership.
|
67
|
+
|
68
|
+
## Attribution
|
69
|
+
|
70
|
+
This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
|
71
|
+
available at [http://contributor-covenant.org/version/1/4][version]
|
72
|
+
|
73
|
+
[homepage]: http://contributor-covenant.org
|
74
|
+
[version]: http://contributor-covenant.org/version/1/4/
|
data/Dockerfile
ADDED
@@ -0,0 +1,49 @@
|
|
1
|
+
FROM ubuntu:xenial-20180228
|
2
|
+
|
3
|
+
LABEL maintainer "Kunihiko Miyoshi <miyoshik@preferred.jp>"
|
4
|
+
LABEL OBJECT="Menoh Ruby Extension Reference Environment"
|
5
|
+
|
6
|
+
ENV INSTALL_PREFIX /usr/local
|
7
|
+
ENV LD_LIBRARY_PATH ${INSTALL_PREFIX}/lib
|
8
|
+
|
9
|
+
RUN apt-get update && apt-get install -y \
|
10
|
+
git \
|
11
|
+
gcc \
|
12
|
+
g++ \
|
13
|
+
cmake \
|
14
|
+
cmake-data \
|
15
|
+
libopencv-dev \
|
16
|
+
protobuf-compiler \
|
17
|
+
libprotobuf-dev \
|
18
|
+
ruby-dev \
|
19
|
+
ruby-rmagick \
|
20
|
+
ruby-bundler && \
|
21
|
+
rm -rf /var/lib/apt/lists/*
|
22
|
+
|
23
|
+
# MKL-DNN
|
24
|
+
RUN mkdir /opt/mkl-dnn
|
25
|
+
WORKDIR /opt/mkl-dnn
|
26
|
+
RUN git clone https://github.com/01org/mkl-dnn.git && \
|
27
|
+
cd mkl-dnn/scripts && bash ./prepare_mkl.sh && cd .. && \
|
28
|
+
sed -i 's/add_subdirectory(examples)//g' CMakeLists.txt && \
|
29
|
+
sed -i 's/add_subdirectory(tests)//g' CMakeLists.txt && \
|
30
|
+
mkdir -p build && cd build && cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_PREFIX .. && make && \
|
31
|
+
make install
|
32
|
+
|
33
|
+
# Menoh
|
34
|
+
WORKDIR /opt/
|
35
|
+
RUN git clone https://github.com/pfnet-research/menoh.git && \
|
36
|
+
cd menoh && \
|
37
|
+
sed -i 's/add_subdirectory(example)//g' CMakeLists.txt && \
|
38
|
+
sed -i 's/add_subdirectory(test)//g' CMakeLists.txt && \
|
39
|
+
mkdir build && \
|
40
|
+
cd build && \
|
41
|
+
cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_PREFIX .. && \
|
42
|
+
make install
|
43
|
+
|
44
|
+
# menoh-ruby
|
45
|
+
RUN gem install rake-compiler
|
46
|
+
RUN mkdir /opt/menoh-ruby
|
47
|
+
ADD . /opt/menoh-ruby
|
48
|
+
WORKDIR /opt/menoh-ruby
|
49
|
+
RUN rake && rake install
|
data/Gemfile
ADDED
data/LICENSE
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
The MIT License (MIT)
|
2
|
+
|
3
|
+
Copyright (c) 2018 Preferred Networks, Inc.
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
7
|
+
in the Software without restriction, including without limitation the rights
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
10
|
+
furnished to do so, subject to the following conditions:
|
11
|
+
|
12
|
+
The above copyright notice and this permission notice shall be included in
|
13
|
+
all copies or substantial portions of the Software.
|
14
|
+
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
21
|
+
THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,80 @@
|
|
1
|
+
# Menoh Ruby Extension
|
2
|
+
|
3
|
+
This is a Ruby extension of [Menoh](https://github.com/pfnet-research/menoh); an ONNX runtime engine developed by [@okdshin](https://github.com/okdshin) and their team [@pfnet-research](https://github.com/pfnet-research).
|
4
|
+
|
5
|
+
## Installation
|
6
|
+
|
7
|
+
You need `ruby-dev`, `bundler`, `rake-compiler` to install this extension.
|
8
|
+
|
9
|
+
```bash
|
10
|
+
$ sudo apt install ruby-dev
|
11
|
+
$ sudo gem install bundler
|
12
|
+
$ sudo gem install rake-compiler
|
13
|
+
```
|
14
|
+
|
15
|
+
And add this line to your application's Gemfile:
|
16
|
+
|
17
|
+
```ruby
|
18
|
+
gem 'menoh'
|
19
|
+
```
|
20
|
+
|
21
|
+
And then execute:
|
22
|
+
|
23
|
+
$ bundle
|
24
|
+
|
25
|
+
Or install it yourself as:
|
26
|
+
|
27
|
+
$ gem install menoh
|
28
|
+
|
29
|
+
## Usage
|
30
|
+
|
31
|
+
Please see [Menoh tutorial](https://github.com/pfnet-research/menoh/blob/master/docs/tutorial.md) and [menoh-ruby tutorial](https://github.com/pfnet-research/menoh-ruby/blob/master/docs/tutorial.md).
|
32
|
+
And we have [some examples](https://github.com/pfnet-research/menoh-ruby/blob/master/example/) on this repository.
|
33
|
+
|
34
|
+
## Development
|
35
|
+
|
36
|
+
After checking out the repo, run `bin/setup` to install ruby dependencies. Then, run `rake test` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
|
37
|
+
|
38
|
+
To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
|
39
|
+
|
40
|
+
### Docker
|
41
|
+
|
42
|
+
You can develop on docker. For details, please refer to [Dockerfile](Dockerfile).
|
43
|
+
|
44
|
+
```bash
|
45
|
+
$ export IMAGE_VERSION=0.0.0 # please specify version
|
46
|
+
$ sudo -E docker build -t menoh-ruby:$IMAGE_VERSION `pwd`
|
47
|
+
$ sudo -E docker run -it --name menoh-ruby-test -v $(pwd):/opt/menoh-ruby --entrypoint /bin/bash menoh-ruby:$IMAGE_VERSION
|
48
|
+
$ cd /opt/menoh-ruby
|
49
|
+
$ rake && rake install
|
50
|
+
|
51
|
+
```
|
52
|
+
|
53
|
+
#### attach after stop
|
54
|
+
|
55
|
+
```bash
|
56
|
+
sudo docker start menoh-ruby-test bash
|
57
|
+
sudo docker attach menoh-ruby-test
|
58
|
+
```
|
59
|
+
|
60
|
+
### Vagrant
|
61
|
+
|
62
|
+
You can also set up the development environment by using Vagrant. The details are available on [Vagrantfile](Vagrantfile).
|
63
|
+
|
64
|
+
```bash
|
65
|
+
$ vagrant up
|
66
|
+
$ vagrant ssh
|
67
|
+
$ cd /vagrant
|
68
|
+
```
|
69
|
+
|
70
|
+
## Contributing
|
71
|
+
|
72
|
+
Bug reports and pull requests are welcome on GitHub at https://github.com/pfnet-research/menoh-ruby. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [Contributor Covenant](http://contributor-covenant.org) code of conduct.
|
73
|
+
|
74
|
+
## License
|
75
|
+
|
76
|
+
The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
|
77
|
+
|
78
|
+
## Code of Conduct
|
79
|
+
|
80
|
+
Everyone interacting in the OnnxInstant project’s codebases, issue trackers, chat rooms and mailing lists is expected to follow the [code of conduct](https://github.com/pfnet-research/menoh-ruby/blob/master/CODE_OF_CONDUCT.md).
|
data/Rakefile
ADDED
@@ -0,0 +1,18 @@
|
|
1
|
+
require 'bundler/gem_tasks'
|
2
|
+
require 'rake/testtask'
|
3
|
+
|
4
|
+
Rake::TestTask.new(:test) do |t|
|
5
|
+
t.libs << 'test'
|
6
|
+
t.libs << 'lib'
|
7
|
+
t.test_files = FileList['test/**/*_test.rb']
|
8
|
+
end
|
9
|
+
|
10
|
+
require 'rake/extensiontask'
|
11
|
+
|
12
|
+
task build: :compile
|
13
|
+
|
14
|
+
Rake::ExtensionTask.new('menoh_native') do |ext|
|
15
|
+
ext.lib_dir = 'lib/menoh'
|
16
|
+
end
|
17
|
+
|
18
|
+
task default: %i[clobber compile test]
|
data/Vagrantfile
ADDED
@@ -0,0 +1,46 @@
|
|
1
|
+
# -*- mode: ruby -*-
|
2
|
+
# vi: set ft=ruby :
|
3
|
+
|
4
|
+
Vagrant.configure('2') do |config|
|
5
|
+
config.vm.box = 'ubuntu/xenial64'
|
6
|
+
|
7
|
+
config.vm.provider 'virtualbox' do |vb|
|
8
|
+
vb.memory = '4096'
|
9
|
+
vb.cpus = 2
|
10
|
+
end
|
11
|
+
|
12
|
+
config.vm.provision 'shell', inline: <<-SHELL
|
13
|
+
export INSTALL_PREFIX=/usr/local
|
14
|
+
sudo apt update
|
15
|
+
sudo apt install -y gcc g++ cmake cmake-data libopencv-dev
|
16
|
+
sudo apt install -y ruby-dev ruby-rmagick
|
17
|
+
sudo gem install bundler rake-compiler
|
18
|
+
|
19
|
+
# protobuf
|
20
|
+
sudo apt install -y protobuf-compiler libprotobuf-dev
|
21
|
+
|
22
|
+
# mkl-dnn
|
23
|
+
cd
|
24
|
+
git clone https://github.com/01org/mkl-dnn.git
|
25
|
+
cd mkl-dnn/scripts && bash ./prepare_mkl.sh && cd ..
|
26
|
+
sed -i 's/add_subdirectory(examples)//g' CMakeLists.txt
|
27
|
+
sed -i 's/add_subdirectory(tests)//g' CMakeLists.txt
|
28
|
+
mkdir -p build && cd build && cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_PREFIX .. && make
|
29
|
+
sudo make install
|
30
|
+
|
31
|
+
# Menoh
|
32
|
+
cd
|
33
|
+
git clone https://github.com/pfnet-research/menoh.git
|
34
|
+
cd menoh
|
35
|
+
sed -i 's/add_subdirectory(example)//g' CMakeLists.txt
|
36
|
+
sed -i 's/add_subdirectory(test)//g' CMakeLists.txt
|
37
|
+
mkdir -p build && cd build && cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_PREFIX .. && make
|
38
|
+
sudo make install
|
39
|
+
|
40
|
+
# menoh-ruby
|
41
|
+
export LD_LIBRARY_PATH=${INSTALL_PREFIX}/lib:$LD_LIBRARY_PATH
|
42
|
+
cd /vagrant
|
43
|
+
rake
|
44
|
+
sudo rake install
|
45
|
+
SHELL
|
46
|
+
end
|
data/bin/console
ADDED
data/bin/setup
ADDED
data/docs/tutorial.md
ADDED
@@ -0,0 +1,121 @@
|
|
1
|
+
# Tutorial
|
2
|
+
|
3
|
+
In this tutorial, we are going to make a CNN model inference software.
|
4
|
+
This tutorial is based on [Menoh](https://github.com/pfnet-research/menoh)'s [original tutorial](https://github.com/pfnet-research/menoh/blob/master/docs/tutorial.md).
|
5
|
+
|
6
|
+
This script loads `data/VGG16.onnx` and takes input image, then outputs classification result.
|
7
|
+
|
8
|
+
## Setup model
|
9
|
+
|
10
|
+
For gettinig ONNX model's named variables, please refer to [Menoh Tutorial](https://github.com/pfnet-research/menoh/blob/master/docs/tutorial.md).
|
11
|
+
|
12
|
+
VGG16 has one input and one output. So now we can check that the input name is *140326425860192* (input of 0:Conv) and the output name is *140326200803680* (output of 39:Softmax).
|
13
|
+
|
14
|
+
Some of we are interested the feature vector of input image. So in addition, we are going to take the output of 32:FC(fc6, which is the first FC layer after CNNs) named *140326200777584*.
|
15
|
+
|
16
|
+
We define name aliases for convenience:
|
17
|
+
|
18
|
+
```ruby
|
19
|
+
CONV1_1_IN_NAME = '140326425860192'.freeze
|
20
|
+
FC6_OUT_NAME = '140326200777584'.freeze
|
21
|
+
SOFTMAX_OUT_NAME = '140326200803680'.freeze
|
22
|
+
```
|
23
|
+
|
24
|
+
To build model, we load model data from ONNX file:
|
25
|
+
|
26
|
+
```ruby
|
27
|
+
onnx_obj = Menoh::Menoh.new './data/VGG16.onnx'
|
28
|
+
```
|
29
|
+
|
30
|
+
Now let's build the model.
|
31
|
+
|
32
|
+
```ruby
|
33
|
+
# data shape of input images
|
34
|
+
input_shape = {
|
35
|
+
channel_num: 3,
|
36
|
+
width: 224,
|
37
|
+
height: 224
|
38
|
+
}
|
39
|
+
# model options for model
|
40
|
+
model_opt = {
|
41
|
+
backend: 'mkldnn',
|
42
|
+
input_layers: [
|
43
|
+
{
|
44
|
+
name: CONV1_1_IN_NAME,
|
45
|
+
dims: [
|
46
|
+
image_list.length,
|
47
|
+
input_shape[:channel_num],
|
48
|
+
input_shape[:width],
|
49
|
+
input_shape[:height]
|
50
|
+
]
|
51
|
+
}
|
52
|
+
],
|
53
|
+
output_layers: [FC6_OUT_NAME, SOFTMAX_OUT_NAME]
|
54
|
+
}
|
55
|
+
# make model for inference under 'model_opt'
|
56
|
+
model = onnx_obj.make_model model_opt
|
57
|
+
```
|
58
|
+
|
59
|
+
## Preprocessing dataset
|
60
|
+
|
61
|
+
Before running the inference, the preprocessing of input dataset is required. `data/VGG16.onnx` takes 3 channels 224 x 224 sized image but input image is not always sized 224x224. So we use Imagemagick's `resize_to_fill` method for resizing.
|
62
|
+
|
63
|
+
`VGG16.onnx`'s input layer *140326425860192* takes images as NCHW format (N x Channels x Height x Width). But RMagick's image array has alternately flatten values for each channel. So next we call `export_pixels` method for each channels `['B', 'G', 'R']`, then `flatten` arrays.
|
64
|
+
|
65
|
+
```ruby
|
66
|
+
image_list = [
|
67
|
+
'./data/Light_sussex_hen.jpg',
|
68
|
+
'./data/honda_nsx.jpg',
|
69
|
+
]
|
70
|
+
image_set = [
|
71
|
+
{
|
72
|
+
name: CONV1_1_IN_NAME,
|
73
|
+
data: image_list.map do |image_filepath|
|
74
|
+
image = Magick::Image.read(image_filepath).first
|
75
|
+
image = image.resize_to_fill(input_shape[:width], input_shape[:height])
|
76
|
+
'BGR'.split('').map do |color|
|
77
|
+
image.export_pixels(0, 0, image.columns, image.rows, color).map { |pix| pix / 256 }
|
78
|
+
end.flatten
|
79
|
+
end.flatten
|
80
|
+
}
|
81
|
+
]
|
82
|
+
```
|
83
|
+
|
84
|
+
In current case, the range of pixel value `data/VGG16.onnx` taking is [0, 256]. On the other hand RMagick's image array takes [0, 65536]. So we have to scale the values by dividing 256.
|
85
|
+
|
86
|
+
And sometimes model takes values scaled in range [0, 1] or something. In that case, we can scale values here:
|
87
|
+
|
88
|
+
```ruby
|
89
|
+
image_set = [
|
90
|
+
{
|
91
|
+
name: CONV1_1_IN_NAME,
|
92
|
+
data: image_list.map do |image_filepath|
|
93
|
+
image = Magick::Image.read(image_filepath).first
|
94
|
+
image = image.resize_to_fill(input_shape[:width], input_shape[:height])
|
95
|
+
'BGR'.split('').map do |color|
|
96
|
+
image.export_pixels(0, 0, image.columns, image.rows, color).map { |pix| pix / 65536 }
|
97
|
+
end.flatten
|
98
|
+
end.flatten
|
99
|
+
}
|
100
|
+
]
|
101
|
+
```
|
102
|
+
|
103
|
+
## Run inference and get results
|
104
|
+
|
105
|
+
Now we can run the inference.
|
106
|
+
|
107
|
+
```ruby
|
108
|
+
# execute inference
|
109
|
+
inferenced_results = model.run image_set
|
110
|
+
```
|
111
|
+
|
112
|
+
The `inferenced_results` is the array that contains the hash of results of `output_layers`. So you can get each value as follows.
|
113
|
+
|
114
|
+
```ruby
|
115
|
+
fc6_out = inferenced_results.find { |x| x[:name] == FC6_OUT_NAME }
|
116
|
+
softmax_out = inferenced_results.find { |x| x[:name] == SOFTMAX_OUT_NAME }
|
117
|
+
```
|
118
|
+
|
119
|
+
That's it.
|
120
|
+
|
121
|
+
The full code is available at [VGG16 example](https://github.com/pfnet-research/menoh-ruby/blob/master/example/example_vgg16.rb).
|