Consider re-introducing five-line approach in docs

[RaoOfPhysics]

[Pasting this from the Slack thread, per @edwardchalstrey1’s suggestion!]

Hey folks! There was some info about the five-lines approach to Scivision that I can’t seem to find now. Is there some place where this is referenced and explained in the docs or in the repo? Thanks!

---

On Slack, @ots22 provided these steps:

1. import everything
2. Assuming you have a model in mind, pass the path or URL to the model
3. Find datasources in the catalog that are compatible with the model
4. Based on the output of the above, pick a dataset, then load it
5. Try out the model on the dataset (selecting one image from the dataset to call predict on

which are based on https://scivision.readthedocs.io/en/latest/user_guide.html#getting-started

[edwardchalstrey1]

Hi @RaoOfPhysics thanks for this, do you think the user guide is unclear? How could we make this easier to understand, or was it a case of not easily finding it in the first place?

[RaoOfPhysics]

Hi Ed! No, I don’t think the user guide was unclear. I had come across references to this so-called “five-lines usage” (probably on Slack but it’s also on the Scivision section of the white paper, so possibly from my notes) and was curious if it was formally mentioned somewhere I could reference.