Add Docling conversion tutorials #21
Conversation
iamemilio
left a comment
iamemilio
left a comment
There was a problem hiding this comment.
Aside from a nit pick, this is great!
fabianofranz
force-pushed
the
docling-conversion-tutorials
branch
from
b02e195 to
7938ce2
Compare
|
@JustinXHale Let me know if you have any comments about the folder structure, or anything else. Thank you! |
JustinXHale
left a comment
JustinXHale
left a comment
There was a problem hiding this comment.
the folder structure under docs/docling-conversion/ looks clean and intuitive. I like that each .py script corresponds directly to a documented conversion technique. Maybe think about better naming structure for "mostly default settings", maybe something like baseline, default, or quickstart, so that the naming isnt as vague.
fabianofranz
force-pushed
the
docling-conversion-tutorials
branch
5 times, most recently
from
3b42046 to
d84b0f1
Compare
fabianofranz
force-pushed
the
docling-conversion-tutorials
branch
from
d84b0f1 to
a934487
Compare
|
@JustinXHale I renamed it to "standard settings". Thank you for the review! |
alimaredia
left a comment
alimaredia
left a comment
There was a problem hiding this comment.
Great job getting this off the ground, I love how concise this is.
The only thing I could think about adding are examples of using the wrong document conversion pipeline on a document, what they look like and how users should adjust and see better results. Something like this could always be added as a follow up
Signed-off-by: Fabiano Franz <[email protected]>
Signed-off-by: Fabiano Franz <[email protected]>
Signed-off-by: Fabiano Franz <[email protected]>
fabianofranz
force-pushed
the
docling-conversion-tutorials
branch
from
a934487 to
5dcd845
Compare
4 participants
RHELAI-4141
Adds a tutorial exposing a set of flags and options that are effective fixes for some of the most common issues faced in document parsing with Docling.