I am trying to creating an image detection and image replacement app
I have the code working in Google Colab. There are two notebooks in google colab, each with their own environment, and they are working fine.
Now, I want to use the webapp for public use, which means it has to be live all the time. But, with Google Colab, I will need to make it run all the time on my laptop which is not possible.
Is there a way to migrate the code to Anvil or host it using some other mechanism in order to have it running all the time
You could try copying all the code doing the heavy lifting to Anvil’s server module. Then, build a front end with Anvil’s client-side modules.
It will then be public if you want it to be.
You are likely going to want to go through some introductory Anvil tutorials to see how it works. Also, you can check out the docs on Uplink, as that might be relevant. You may also search the forum and tutorials for Jupyter and Colab for background.
So, basically, what you’re asking is possible but you need to copy your code to Anvil in a way that fits with its paradigm.
Give it a shot and let us know, with as many details as possible, how it goes !
My code in Google Colab works on different Tensorflow versions. For example, both work on Tensorflow v1.0.
So, how do I get a server module to work withing Tensorflow v1.0. Similarly, I have many different versions of packages in each code. So, how can I create a separate virtual environment within server modules?
I understand that we need to write to support@anvil to get a new package installed. But, how do I separate packages for each server module?
We have a tutorial which documents how to migrate your code from a Colab notebook and deploy it on one of your own machines with Docker: Deploy a Google Colab Notebook with Docker
!