ArchivesPortalEuropeFoundation / Topic-Detection

Using machine learning approaches for automatic topic detection in a multilingual environment
6 stars 0 forks source link

Setup Docker for new VM #55

Open fedenanni opened 2 years ago

fedenanni commented 2 years ago

@kstamatis @kerstarno I am preparing here an overview of the repo. The Docker container would need to:

fedenanni commented 2 years ago

I re-adapted the repo (in dev for the moment) and the web-app is now deployable with a few commands, which we can automatise in docker. With Python 3.7, this is the way to set it up:

git clone https://github.com/ArchivesPortalEuropeFoundation/Topic-Detection.git
cd Topic-Detection
pip install -r requirements.txt
cd web-app
python start_api.py
fedenanni commented 2 years ago

Current update, I have a very initial version of a Docker image where I can start the webapp. I now need to work on:

a) how to make it lighter (in terms of both OS and resources) b) how to make sure it works when scaling up to the real resource and not the demo dataset

fedenanni commented 2 years ago

@kstamatis @kerstarno Ok - I finally have a version of a Docker container which runs, opens the correct port (5000), receives the query and gets back the results (https://github.com/ArchivesPortalEuropeFoundation/Topic-Detection/commit/a0f80b10c43b16d594e968910a38e5d35c745b14). It currently works in test mode and requires very few resources. I'll test it now from scratch to make sure it works in another environment and if that's the case I just need to work on how to attach the external large-scale resources and then I am done