Open jcalebsmith opened 1 year ago
Oh dear. It looks like I've messed up the npm packaging for pandoc-svelte-components
. I will try to fix this this week.
But be aware, I'm not sure it's possible now to run this library directly. There are a few other issues that keep it from being served locally (including having the 20 million items served locally)
Oh dear. It looks like I've messed up the npm packaging for
pandoc-svelte-components
. I will try to fix this this week.But be aware, I'm not sure it's possible now to run this library directly. There are a few other issues that keep it from being served locally (including having the 20 million items served locally)
No worries at all! Deepscatter works just fine, which is the main thing.
What I was hoping to do was use this repo as a template/guide for my own project. In my project, we've clustered the pubmed corpus into ~30,000 nodes and have assigned x,y coordinates using our own algorithm. I'd like to use deepscatter for serving this visualization (and would love to figure out a hack to get it to incorporate edges!). My interest in this repo was primarily to use it as a UI and storytelling template. I love the scroller narrative you incorporated, and the UI is very clean and intuitive. I'm no web developer though, so I was hoping to plug in my own point data and then tweak the UI, scroller narrative, etc. as necessary rather than trying to recreate what you've already done.
I'd be grateful for any advice you have for incorporating UI and storytelling aspects similar to what you've done here top of Deepscatter!
tl;dr I love deepscatter in general and the UI you've put together here in particular. However, I need to substitute my own point data (~30,000 points). All other functionality remains the same.
Hi Caleb! I guess I know what dataset you are talking about ;-) (https://www.nature.com/articles/s41597-020-00749-y) Would indeed be great if there were an explorable visualisation of it available, I'm looking forward! Dmitry
Hi Caleb! I guess I know what dataset you are talking about ;-) (https://www.nature.com/articles/s41597-020-00749-y) Would indeed be great if there were an explorable visualisation of it available, I'm looking forward! Dmitry
Hi Dmitry! I've been found out! Congratulations on this paper (and thank you for posting your scripts on github)! It's certainly creating a buzz and the visualizations are beautiful.
We've just finished a new version of our own model and I would like to get something up so people can interact with it. I would love to see what the overlap is between your embedding-based pubmed clusters and our citation-based clusters. I have the beginnings of an idea for how the two might be combined...I'll reach out over email.
I've installed everything per the readme and have verified that I can get deepscatter working, however, once I run 'npm run dev' and attempt to go to http://localhost:2001/pubmed, I get an error stating
Below I've pasted the full output. I'm not sure what the issue is. I can run deepscatter just fine on its own. I've verified that pandoc is installed and is being found without issue.