Sapie Space's mission is to connect businesses to social media influencers using machine learning.
# First, clone the repo to a desired folder
$ git clone https://github.com/sapie-space/sapie.git
# Now, we are going to install npm dependancies
$ cd ~.../sapie/frontend
$ npm install
# Next, we need to point search requires to our localhost
$ cd .../sapie/frontend/src/page
$ atom Search.js
# Around line 670, there is a line that says axios.post('https://ec2.../run_query')
# Change the URL ('https://ec2../run_query') to ('http://127.0.0.1:5000/run_query')
# Note: this is the default FLASK url and may differ if you've changed the default
# Now we are going to install the backend requirements
$ cd ~../sapie/backend
$ pip3 install -r requirements.txt
$ cd ~../sapie/backend
$ atom be_rest_api.py
# In this file, we will need to comment out this line : sys.path.insert(0, '/home/ec2-user/sapie/webcrawler/yougod/yougod/')
# Next, scroll to the bottom and in if __name__ == "__main__", comment out the try-catch run block and simply
# uncomment app.run(debug=True)
# Next, run the backend in cd .../sapie/backend
$ python3 be_rest_api.py
# Lastly, run the frontend. Open a new terminal window and navigate to /sapie/frontend
$ npm start
## Quick Start (with ENV and Docker)
Make sure you have latest Docker and Docker Compose installed. Download elasticsearch and run these commands from project root:
```console
# create a virtualenv in your parent directory if you need one to install requirements (Windows/MacOSx)
$ virtualenv your_environment_name
$ source /bin/activate/your_environment_name
# you will run echo only once to create local `.env` file indicating you are using dev environment.
$ echo COMPOSE_FILE=docker-compose.yml:docker-compose.dev.yml > .env
# run to build images every time dependencies get updated
$ docker-compose build
# This is one of many other ways to get Docker Compose running your selected environment.
$ docker-compose up
# Starting ES:
# elasticsearch is likely stored in your parent directory. Go back to that, enter
$ ./elasticsearch
# and then type:
$ ./bin/elasticsearch
# in root:
$ pip3 install -r new_requirements.txt
# in sapie/backend
$ python3 be_rest_api.py
# in sapie/fe:
$ npm install
$ npm startThen you just have to point your browser to http://localhost:3000/ to see it running. You can also access API endpoints directly like http://localhost:3000/api/influencer
To connect to the SapieSpace EC2 Instance on AWS:
change directories into the Sapie space project on your local machine
$cd /sapie-space
Your key must not be publicly viewable for SSH to work. Use this command if needed:
chmod 400 sapiekeypair.pem
Connect to your instance using its Public DNS
Here's our specific instance name: ec2-34-209-86-220.us-west-2.compute.amazonaws.com
Here's how we'll connect to the EC2 instance:
ssh -i "sapiekeypair.pem" ec2-user@ec2-34-209-86-220.us-west-2.compute.amazonaws.com
Please note that in most cases the username above will be correct, check AWS EC2 or ask Brody to see if the default AMI username changed.
### Known Issues
You might have trouble pulling ElasticSearch official images. You can try it manually just in case:
```console
$ docker pull docker.elastic.co/elasticsearch/elasticsearch:6.1.1If you're on an encrypted network (such as an institution's network), you may have trouble connecting to AWS. You'll have to create a VM in AWS. Speak with Brody more about that.
$ ssh -A -i sapie-dev.pem ubuntu@18.218.86.249The interface is a business-facing platform that allows businesses to connect with influencers. To see an investor's pitch that I did, you can go to this link (and request Google Drive access, if needed).
Jack Treseler and Brody are the co-founders of Sapie Space. Brody was working at Google and Jack was an employee at Square. Jack came up with the idea when he found it difficult to make connections with influencers, see how influential they actually are, as well as if they were a good fit for the company. Jack and Brody have been working to make Sapie Space a reaity since April 2017.
A decoupled front-end started being developed and can be found under
frontend. The idea is to merge it with a REST API back-end under
backend. We have also a docker compose environment defined that
should ease daily development and deployment.
We use Asana for task management, and We use Slack to communicate.
We are organizing the house, but if you are trying to run this, you might try to look into these first.
Make you run also have run the following commands in the correponding directory to install all python and js dependencies: