Project Owner: Tanya Tandon QA Owner: Saurabh Annadate

Dear Genie - Kickstarter

Vision: In the history of the world, today is the best time for a dream to come true. With distances in the world shrinking by the internet, Kickstarter is one of the fastest and easiest ways to kickstart a great idea. This project aligns with Kickstarters mission to "help bring creative projects to life" by predicting if a kickstarter campaign will be successful or not. It will help any and everyone launching a campaign on Kickstarter. Users with failed campaigns can leverage the platform by exploring alternate goals as well as alternate completion dates to understand the tweaks needed for a more likely success.

Mission: Developing a classifier algorithm that predicts whether a Kickstarter project will be successful or not. This will help users evaluate whether their campaign will be able to raise money or not even before they do it. It will further suggest alternate completion dates as well as alternate goals to users that will be more likely to succeed. This will help users modify their campaign for a sure shot success.

Success criteria:

• Accuracy of at least 60% ( Using AUC or Misclassification rate)
• Inbound traffic of at least 100 Kickstarter campaign starters
• Above 70 % positive feedback on UI design

Planned work

Theme 1: Model Development

Epic 1: Design data set that best represents all aspects needed to classify a Kickstarter project

• Story 1: Downloading the relevant dataset from Kaggle keeping in mind that it would be possible to join them later
• Story 2: Connect all these data sets
• Story 3: Design new features

Epic 2: Data preparation

• Story 1: Exploratory Data Analysis
• Story 2: Data Cleaning

Epic 3: Model Development

• Story 1: Build logistic regression model
• Story 2: Build boosted trees
• Story 3: Build Random Forest
• Story 4: Compare measures of success (AOC, missclass rate, recall, etc) (%) for each on new data
• Story 5: Reiterate the model for improvement

Epic 4: Create unit test scripts to test model functionality

Theme2 : App Development

Epic 1: Setup Deployment pipeline

• Story 1: Setting up the requirements.txt from the current environment state to make reproducible environment
• Story 2: Set up S3 instance
• Story 3: Initialize RDS database
• Story 4: Deploy the flask app

Epic 2: Setup front end interface UI where the users will interact with the model built

• Story1: Setup a basic front end interface with all the required options
• Story 2: Improve the basic front end interface to include engaging gifs on the Home page

Epic 3: Create UAT test scripts to test that UI is working as expected to intake selections and output accordingly

• Story1: Check for model functionality
• Story 2: Check for exception handling by UI

Epic 4: Create logging for easy debugging and error notifications

Backlog

Sprint Sizing Legend:

• 0 points - quick chore
• 1 point ~ 1 hour (small)
• 2 point ~ 2.5 hour ( slightly bigger than small)
• 3 points ~ 1/2 day (medium)
• 4 points ~ 1 day (large)
• 8 points - big and needs to be broken down more when it comes to execution (okay as placeholder for future work though)

1. Theme1.epic1.story1 (2pts) - (completed)

2. Theme1.epic1.story2 (2pts) - (completed)

3. Theme1.epic1.story3 (2pts) - (completed)

4. Theme1.epic2.story1 (2pts) - (completed)

5. Theme1.epic2.story2 (2pts) - (completed)

6. Theme1.epic3.story1 (2pts) - (completed)

7. Theme1.epic3.story2 (2pts) - (completed)

8. Theme1.epic3.story3 (2pts) - (completed)

9. Theme1.epic3.story4 (1pts) - (completed)

10. Theme1.epic3.story5 (2pts) - (completed)

11. Theme1.epic4 (3pts)

12. Theme2.epic1.story1 (2pts) (completed)

13. Theme2.epic1.story2 (1pts) (completed)

14. Theme2.epic1.story3 (1pts) (completed)

15. Theme2.epic1.story4 (1pts) (completed)

16. Theme2.epic1.story5 (1pts) (completed)

17. Theme2.epic2.story1 (3pts) (completed)

18. Theme2.epic2.story2 (4pts) (completed)

19. Theme2.epic3.story1 (1pts) (completed)

20. Theme2.epic3.story2 (1pts) (completed)

21. Theme2.epic4.story1 (2pts) (completed)

22. Theme2.epic4 (2pts)

Icebox

Epic 1 : Develop a recommendation bar to suggest alternate goals and alternate completion dates for success of the projects

• Story 1: Integrate the model with this functionality
• Story 2: Integrate the UI with this functionality

Repo Structure

├── README.md                         <- You are here
│
├── app
│   ├── static/                       <- CSS, JS files that remain static 
│   ├── templates/                    <- HTML (or other code) that is templated and changes based on a set of inputs
│   ├── models.py                     <- Creates the data model for the database connected to the Flask app 
│   ├── __init__.py                   <- Initializes the Flask app and database connection
│
├── config                            <- Directory for yaml configuration files for model training, scoring, etc
│   ├── logging/                      <- Configuration files for python loggers
│
├── data                              <- Folder that contains data used or generated. Only the external/ and sample/ subdirectories are tracked by git. 
│   ├── archive/                      <- Place to put archive data is no longer usabled. Not synced with git. 
│   ├── external/                     <- External data sources, will be synced with git
│   ├── sample/                       <- Sample data used for code development and testing, will be synced with git
│
├── docs                              <- A default Sphinx project; see sphinx-doc.org for details.
│
├── figures                           <- Generated graphics and figures to be used in reporting.
│
├── models                            <- Trained model objects (TMOs), model predictions, and/or model summaries
│   ├── archive                       <- No longer current models. This directory is included in the .gitignore and is not tracked by git
│
├── notebooks
│   ├── develop                       <- Current notebooks being used in development.
│   ├── deliver                       <- Notebooks shared with others. 
│   ├── archive                       <- Develop notebooks no longer being used.
│   ├── template.ipynb                <- Template notebook for analysis with useful imports and helper functions. 
│
├── src                               <- Source data for the project 
│   ├── archive/                      <- No longer current scripts.
│   ├── helpers/                      <- Helper scripts used in main src files 
│   ├── sql/                          <- SQL source code
│   ├── add_songs.py                  <- Script for creating a (temporary) MySQL database and adding songs to it 
│   ├── ingest_data.py                <- Script for ingesting data from different sources 
│   ├── generate_features.py          <- Script for cleaning and transforming data and generating features used for use in training and scoring.
│   ├── train_model.py                <- Script for training machine learning model(s)
│   ├── score_model.py                <- Script for scoring new predictions using a trained model.
│   ├── postprocess.py                <- Script for postprocessing predictions and model results
│   ├── evaluate_model.py             <- Script for evaluating model performance 
│
├── test                              <- Files necessary for running model tests (see documentation below) 

├── run.py                            <- Simplifies the execution of one or more of the src scripts 
├── app.py                            <- Flask wrapper for running the model 
├── config.py                         <- Configuration file for Flask app
├── requirements.txt                  <- Python package dependencies 

Instructions to run the application

Ths application can be run on both local system as well as on AWS. Steps on how to deploy the app for both settings is given below.

1. Setting up AWS

Either you run on AWS or Local, you need to set up your aws account. There are two ways to do it.

1.1. Download .pem file from AWS

You can check if you own it by going to your terminal and typing cd ~/.aws . If it exists, then you are good otherwise, follow 1.2.

1.2. Set up configure using terminal

pip install awscli
aws configure

Input values in AWS Access key ID and AWS Secret Access Key. Press enter on region and output format.

2. Specify in configure file(s) if the app will be run locally or on AWS

2.1. Running on a local machine

config/config.py

• Make sure that the WHERE variable is set to Local
• Ensure that the HOST variable is set to 127.0.0.1
• Keep the default PORT of 3000 ; However, change it if this port is not available

Makefile

• Ensure that the WHERE variable is set to Local
• Ensure that the variable BUCKET is set to Scipts if running on Windows, else bin if on Linux or MacOS

2.2. Running on AWS

RDS configurations In order to use RDS, the RDS credentials need to be added to the os environment. Add the following variables to the environment:

• MYSQL_USER : Username to access the RDS instance
• MYSQL_PASSWORD : Password to access the RDS instance
• MYSQL_HOST : RDS instance endpoinr
• MYSQL_PORT : Port number to access the instance
• MYSQL_DB : Name of the database

Makefile

• Make sure that the WHERE variable is set to AWS
• Set the variable YOUR_S3_BUCKET to the S3 bucket that you intend to use
• Make sure that the variable BUCKET is set to AWS if running on Windows server, else "bin" if on Linux server

3. Run all the functions in one step

In command line, type make all

4. Run each function seperately

4.1. Set up environment

With virtualenv

pip install virtualenv

virtualenv Kickstarter

source Kickstarter/bin/activate

pip install -r requirements.txt

With conda

conda create -n Kickstarter python=3.7.3
conda activate Kickstarter
pip install -r requirements.txt

With Make

make venv

4.2. Download the data

Original source: https://www.kaggle.com/kemical/kickstarter-projects The data has also been loaded on a public AWS bucket called: bossbucket

Local

Local: Using run file

Run the following command in bash:

python run.py loadS3

Running this code will download the raw data from the s3 bucket and will put it in /Data/raw/

Local: With Make

make load_data

AWS

AWS: Using run file

Run the following command in bash:

python run.py loadS3 --where=AWS --bucket=<destination_bucket_name>

Running this code will download the raw data from the s3 bucket and will put it in <destination_bucket_name>/raw/

AWS: With Make

make load_data

4.3. Initialize the database

Local

Run the following command in bash:

python run.py create_db

Running this code will create a sqlite database to log the app usage at: /Database/msia423.db The table name is Userinput

With Make

make create_db

AWS

Make sure the environment varia Run the following command in command line:

python run.py create_db --where=AWS

With Make

make create_db

Running this code will create the database specified in the given RDS instance

4.4. Clean the data

Local

Run the following command in command line:

python run.py clean_data

With Make

make clean_data

Running this code will clean the raw data and create a cleaned dataset at: /Data/clean/

AWS

Run the following command in command line:

python run.py clean_data --where=AWS --bucket=<destination_bucket_name>

With Make

make clean_data

Running this code will clean the raw data from the S3 bucket and will put it in <destination_bucket_name>/clean/

4.5. Train Model

Local

Run the following command in command line:

python run.py Model_fitting

With Make

make train_model

Running this code will train the prediction model and will dump it at: /models/

AWS

Run the following command in command line:

python run.py Model_fitting --where=AWS --bucket=<destination_bucket_name>

With Make

make train_model

Running this code will clean the raw data from the S3 bucket and will put it in <destination_bucket_name>/models/

4.6. Launch app

Local or AWS

Run the following command in command line:

python run.py run_app

With Make

make app

Running this code will start the flask app at the IP specified in config/config.py

Testing

In order to unit test the functions, follow the following steps:

1. Navigate to the tests folder
2. Run the following command on the command line:

pytest test_all_functions.py

Logging

All logs are saved at logs/logfile.log