-
Notifications
You must be signed in to change notification settings - Fork 5.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[train][docs] fix doc search issues, examples gallery & filter #31635
Conversation
@matthewdeng can I get a review from you?! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is great!
My main confusion is that the "hover" color is almost the same as the "active" color which makes it hard to see if I am activating/deactivating a filter
@krfricke thanks, great comments. will address both later |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is really nice! I feel like I learned some neat things here - is this in itself something we can codify as best practices?
which is why Ray Train provides ``Backend`` implementations (e.g. ``TorchBackend``) | ||
that can be used to run the training process using a `BackendExecutor`. | ||
|
||
Here's a visual overview of the architecture components of Ray Train: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we rename this guide to be deep learning specific? The boosted tree trainers actually don't have any of these backend/executor/worker group abstractions.
Also, this diagram is severely outdated 😅
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I know... man, this particular doc is painful to read. That's why I touched it up, but it's a bit of a rabbit hole!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'll add a TODO, but we need to commit to fixing this. I think having a general Architecture Guide is useful to have.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @maxpumperla, this looks great! Left some comments
Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com> Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
@matthewdeng yes, I think we can codify this. I've summarised the approach here: https://docs.google.com/document/d/1cXb1FELZme9zxEYhXXxrGvtLB2xs1auHA46v-1uMYf0/edit#heading=h.8h9qzb62266g The gallery improvements of this PR are not part of that, but having consistent, self-explaining headers (esp. for examples) is. If we want to boost certain concepts and workloads, ideally we have documents with titles corresponding to these searches. One example: it's difficult to find good search results (on-site or search engines) on "distributed training with PyTorch/TensorFlow (on Ray)". We cover these workloads in the examples touched in this PR. But if we're literally never mentioning what we're doing anywhere, no paid doc search service can salvage that. |
Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Happy with the content, just two minor comments for the tags
Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com> Signed-off-by: Max Pumperla <max.pumperla@googlemail.com>
…roject#31635) Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com> Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com> Signed-off-by: Edward Oakes <ed.nmi.oakes@gmail.com>
docs preview