PicassoXS

EECS 441 @ UMich Project

Description

PicassoXS is your professional photo editing app that transfer your own photo into different painting

Repo Structure

.
├── README.md
├── doc                      # DOCUMENTATION
│   ├── AlgorithmSelection   # doc for algorithm selection 
│   └── PaintingSelection    # doc for painting selection 
└── src                # CODE 
    ├── backend        # code for BACKEND 
    │   ├── arbitrary_style_model       # model for user upload own style 
    │   │   ├── servable.py
    │   │   ├── infer.py
    │   │   ├── main.py
    │   │   ├── model.py
    │   │   ├── servable.py
    │   │   ├── train.py
    │   │   ├── utils.py
    │   │   └── * others
    │   ├── flask_app                  # flask web app 
    │   │   ├── __init__.py
    │   │   ├── config.py
    │   │   ├── * others
    │   │   └── views
    │   │       ├── __init__.py
    │   │       └── index.py 
    │   ├── gcloud                     # instructions to start cloud server 
    │   │   ├── Instruction.md
    │   │   ├── st_k8s.yaml
    │   │   └── * others
    │   ├── train                     # instructions on how to train model on server  
    │   │   ├── Instruction.md
    │   │   ├── tf1-gpu.yml
    │   │   ├── tf2-gpu.yml
    │   │   └── * others
    │   ├── general_model              # model for portrait mode model & general model 
    │   │   ├── README.md
    │   │   ├── StyleTransferrer.proto
    │   │   ├── img_augm.py
    │   │   ├── inference.sh
    │   │   ├── layers.py
    │   │   ├── main.py
    │   │   ├── module.py
    │   │   ├── prepare_dataset.py
    │   │   ├── setup.py
    │   │   ├── train.py
    │   │   └── train.sh
    │   └── tfserver                  # TensorFlow server : web app for serving model 
    │   │   ├── Instruction.md
    │   │   ├── PackDocker.md
    │   │   ├── QuickStart_ArbitaryStyle.md
    │   │   ├── QuickStart_ArbitaryStyleModel.md
    │   │   ├── QuickStart_GeneralModel.md
    │   │   ├── QuickPackServable.ipynb
    │   │   ├── servable_demo.ipynb
    │   │   ├── SendRequestArbitaryStyleModel_gRPC.py
    │   │   ├── SendRequestGeneralModel_REST.py
    │   │   ├── SendRequestGeneralModel_gRPC.py
    │   │   ├── models.config
    │   │   ├── *others 
    │   │   └──servable             # SERVABLE : packed model that can be used by TensorFlow Server
    │   │       └── * others 
    │   └── README.md
    │   └── app.yaml
    │   └── flask_run.sh
    │   └── requirements.txt
    │   └── setup.py
    └── frontend        # code for FRONTEND     
        └── PicassoXS

Contributors

Frontend

Zhijie Ren (uniqname: rzj)
Senhao Wang (uniqname: rogerogw)
Juye Xiao (uniqname: juyexiao)

Backend

Xiao Song (uniqname: xiaosx)
Enhao Zhang (uniqname: ehzhang)
Rena Xu (uniqname: renaxu)
Jason Zhao (uniqname: jasonyz)

Development Environment

IDE:

Frontend - Xcode
Backend - Visual Studio Code

Programming Languages:

Frontend - Swift
Backend - Python v.3.5 ~ v.3.7

Operating System:

Ubuntu 18.04 LTS

Third Party Libraries

1. TensorFlow

Release: 2.0

TensorFlow is an end-to-end open source platform for machine learning. We can use it for all model traning.

Installation

$ pip install tensorflow

TensorFlow intallation

TensorFlow documentation

2. Docker

Release: 19.03.5

Docker provides a way to run applications securely isolated in a container, packaged with all its dependencies and libraries. We can use Docker's containers to create virtual environments that isolate a TensorFlow installation from the rest of the system.

Note: Docker Desktop version is better, but can still install through command line

Installation

$ curl -fsSL https://get.docker.com -o get-docker.sh
$ sudo sh get-docker.sh

note: needs to log out and log back in for docker to work

Docker installation

Docker documentation

3. gRPC

Release: 1.27.1

Google's remote process call framwork. It can efficiently connect services in and across data centers with pluggable support for load balancing, tracing, health checking and authentication.

Installation

$ pip install grpcio

gRPC installation

gRPC documentation

Protocal Buffers

.proto extension
A language-neutral, platform-neutral, extensible way of serializing structured data for use in communications protocols, data storage, and more
Used with gRPC
Protocal Buffer documentation

4. AWS

EC2 (Elastic Compute Cloud):

Free to use
Designed to make web-scale cloud computing easier for developers. Amazon EC2’s simple web service interface allows you to obtain and configure capacity with minimal friction
EC2 documentation

ECS (Elastic Container Service):

A fully managed container orchestration service
Manage Docker containers on a cluster
ECS documentation

5. OpenCV

Release: 4.2.0

A library of programming functions mainly aimed at real-time computer vision.

Installation

$ pip install opencv-python

OpenCV installation

OpenCV documentation

6. SQLAlchemy

Release: 1.3.13

SQLAlchemy is the Python SQL toolkit and Object Relational Mapper that gives application developers the full power and flexibility of SQL.

Installation

$ pip install SQLAlchemy

SQLAlchemy installation

SQLAlchemy documentation

7. Tornado

Release: 6.0.3

Tornado is a Python web framework and asynchronous networking library

Installation

$ pip install tornado

Tornado installation

Tornado documentation

8. tqdm

Release: 4.43.0

A progress bar library with good support for nested loops

Installation

$ pip install tqdm

tqdm documentation

9. Pillow

Release: 2.2.2

Python Image Library. A free library for the Python programming language that adds support for opening, manipulating, and saving many different image file formats

Installation

$ pip install Pillow==2.2.2

Pillow documentation

10. SpiPy

Release: 1.4.1

A python-based ecosystem of open-source software for mathematics, science, and engineering

Installation

$ pip install scipy

SciPy documentation

11. imageio

Release: 2.8.0

Provides an easy interface to read and write a wide range of image data

Installation

$ pip install imageio

imageio documentation

12. Tensorflow Addons

Release: 0.8.2

TensorFlow Addons is a repository of contributions that conform to well- established API patterns, but implement new functionality not available in core TensorFlow

Installation

$ pip install tensorflow-addons

13. Pandas

Release: 0.24.0

Installation

$ pip install pandas

pandas documentation

Set Up Third Party Dependency

dep_install executable: one script that install dependencies completely
pip_setup executable: install python 3.7 and pip 20.0.2
requirements.txt: list thrid party libaries for pip to install

Note:

Your environment should be ubuntu 18.04 before you run the following command
You might have to press yes or Enter several times during installation to give permission to install
If you wish to install the libraries manually, be sure to run upgrade pip3
- $ pip3 install --upgrade pip

Library setup:

Method 1: Install with pip and requirements.txt

Upgrade python to python3 and upgrade pip to pip 20.0.2
- $ chmod +x pip_setup
- $ ./pip_setup
Install libraries with pip and requirements.txt
- $ pip install -r requirements.txt

Method 2: Install with shell script

Give the following shell script permission
- $ chmod +x dep_install
Run the script
- $ ./dep_install

Method 3: Install manually

Upgrade python to python3 and upgrade pip to pip 20.0.2
- $ chmod +x pip_setup
- $ ./pip_setup
Run each installment command in the Third Party Libraries section

DynamicProgrammingEECS441 / PicassoXS

readme

PicassoXS

Description

Repo Structure

Contributors

Frontend

Backend

Development Environment

IDE:

Programming Languages:

Operating System:

Third Party Libraries

1. TensorFlow

2. Docker

3. gRPC

Protocal Buffers

4. AWS

5. OpenCV

6. SQLAlchemy

7. Tornado

8. tqdm

9. Pillow

10. SpiPy

11. imageio

12. Tensorflow Addons

13. Pandas

Set Up Third Party Dependency