App deployment for ML Level 1

Hi @Sara_EL-ATEIF @ckim,

I checked with Jillian. We can host the Flask app via our data science portal if it is dockerized.

Sara, is this something you can drive? Feel free to reach out to Jillian if you need help with any of the dockerization steps. I am tagging the ML pathway in case anyone else is interested in this discussion. If no one responds, you can work with Caleb at your convenience. Otherwise, please share meeting times with the people who indicate interest.

Caleb, take a quick look at the video below. The bioinformatics app is not yet online so sharing the video instead. If you can change the color to stemaway colors and add a little information about the project/ core contributors, we can start sharing externally. Sara can guide you about any other final steps needed (should be all minor).

@2021-ML-L-S1 @2021-ML-L-S2 @2021-ML-P-S1 @2021-ML-P-S2: Please like this post if you are interested to learn more about app dockerization and deployment.


Just an update. I watched the video. Also, I changed the colors to stemaway colors and wrote a little information about the project.



@ddas for sure, for the team please take time to check the last module about building the API and containerizing it (putting everything inside docker). If you have questions ping me and I will set a time so we can have a meeting about it.

To explain the process :


  1. Wrap the model inside an API using Flask for example
  2. Construct the app that will call the API (front-end app)
  3. Construct a docker file that will have instructions on libraries to install and commands to run the app (as in setup the environment of the app).
    PS: Try to only put only the API inside docker for now as you may need to perform some configs to the Flask App.
  4. You can either: a) push the docker to docker hub and then you will simply call/install that docker that has everything inside AWS or b) put the docker file and all necessary files inside a GitHub repo and clone the GitHub repo inside AWS then run dockerfile to set up the environment.
  5. After AWS was set up to have a public address you will use that IP address to access the app as you did with localhost. @ddas can you ask Jilian to help with AWS so we can deploy the app? Or for now @ckim you can try the free tier to test and then we can contact Jilian for further instructions.


Should I make an AWS account?

I will try and get you the account asap. Meanwhile you can work with the free tier if desired.

@ckim please check your email and check the account created by Parminder.

Thank you @ddas.

1 Like

@ckim are you available now so we can do the meeting and discuss?

Sorry for the late reply. But I am available now.

@Sara_EL-ATEIF I am tagging you just in case this might make you see it faster.

Let’s report it to another day. How is it going from your end have you been able to deploy it on AWS @ckim ?

It’s going ok, I’m just starting out. I’m using one of the articles you sent in this thread as a template.

1 Like

Alright! If you get stuck just ping me here and I will try to help :slightly_smiling_face:.

Sorry for the delay, I’m sorry but I don’t think I will be able to do the App deployment on AWS, because I am getting more busy with school as my classes have started. I thought that this was supposed to be an extra thing that I would do if I had time for, and I thought I could, but it ended up just not working out.

Again sorry for the delay, I thought maybe I could have done it but I think it’s just not working out.

Hi Caleb, no worries at all. Classes should be the priority now! Great work on the project, it was a pleasure having you with us in summer!