Improve quick start
falexwolf opened this issue · comments
Empty figures
I noticed that the figures in the quickstart look a bit empty:
Maybe this is how it's supposed to look! Could one add a description of what it's supposed to show?
Explanations
It'd be super nice to one sentence of explanation:
![image](https://private-user-images.githubusercontent.com/16916678/269911683-6893e834-79a4-4280-bd68-e9cd4e655a5d.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MjE2NDM0MzMsIm5iZiI6MTcyMTY0MzEzMywicGF0aCI6Ii8xNjkxNjY3OC8yNjk5MTE2ODMtNjg5M2U4MzQtNzlhNC00MjgwLWJkNjgtZTljZDRlNjU1YTVkLnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNDA3MjIlMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjQwNzIyVDEwMTIxM1omWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPTU5OTA1NTUwOTFiMTlmNzE1YzUxN2I1ZjcwMzViOTY4Zjk5MWQ3MTE4MzQyYjkwZmQzNjhhN2M2NjE1YzMwN2UmWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0JmFjdG9yX2lkPTAma2V5X2lkPTAmcmVwb19pZD0wIn0._AzfSnW08ZtkgsRdttqxdIC5q0RpsxPZd-ecOaH8LTY)
Warnings
It'd come across as higher code quality if there were no warnings - I'm not sure whether much of this is because of using a very toy dataset: #53
Hi Alex,
thanks for pointing out all of this. Some of the issues are indeed due to the choice of toy data. I would love to add a different example dataset, but I am hesitant to add one to the package. Do you have other suggestions to address this?
I'll have a look at the deprecation warning and add more explanations to the examples.
Hey Maren! Yes, I think @sunnyosun & I would happily host example datasets with reliable access.
In fact, we're currently thinking through which datasets would be pedagogically sensible to illustrate data management: https://lamin.ai/docs/facs
Right now, we thought we're taking some from the linked Alpert19 paper and hosting it stably on S3.
What are you favorite flow datasets? We'd be happy to curate and host them!
Hi Alex,
thank you so much with your support and for hosting a more suitable dataset. I have addressed your points in PR #57