-
Notifications
You must be signed in to change notification settings - Fork 1.2k
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
Epic: exp init #6446
Comments
Maybe we can have a label for these summaries? |
From the @dberenbaum's comment on #6681 (comment).
One way may be to make Another would be to create a |
Although it adds another option, I like this mainly because it's more explicit. Now that we don't have templates, it's less transparent what the command and various options are doing. Having a separate option for this suggests that it's not just another path to add but actually changes the stage configuration. We can also further explain it in the help text for that option. On a similar note, at least interactive mode should probably have a confirmation so that users can see what will be added to |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
We should probably include docs work also. |
I noticed a couple of workflow issues when doing some QA of
cc @pmrowla |
New/untracked files aren't pulled into the from https://dvc.org/doc/user-guide/experiment-management/running-experiments#the-experiments-queue:
It sounds like we need:
|
This is only true for the data dependency, since the code and params dependencies probably should be git tracked. We might just start with a hint here. Doing a
👍 Can we also try to add (not force but at least do a regular git add) any other dependencies of the experiment? |
Discussed with @pmrowla and he noted that iterating over stages and git-adding everything during Would it be possible to mark the
This also seems like it might be useful generally as an option in |
Till now, we have considered dependency to be a git-tracked or dvc-tracked dependency, so we do not track them by default. But makes sense to do for Would it be confusing to create Regarding the |
@dberenbaum, should we also |
Ideally, we could specify in
👍
👍 |
@pmrowla, do we need to also stage EDIT: looks like we do need to stage that file ourselves. |
Just wanted to upvote:
At least for the And regardless of whether is a deep learning ( |
What about naming the types to match the use cases, like classical ml and deep learning? |
I'm sorry, I do not quite understand why |
Two reasons:
I would consider these to mostly be useful when doing deep learning, although maybe that's my personal bias. @karajan1001 do you see those being useful in other scenarios? However, dvclive may become more of a general-purpose logger, in which case it may be useful in any ml scenario. That's one reason to consider renaming |
My 2c:
may place nicely together then. |
@iesahin, |
@skshetry Can we go with |
After discussion, these are being de-prioritized for now since they are not simple changes and are not strictly necessary. |
|
Yeah, it's not like ML, but it's fairly common. Google "deep learning abbreviation." |
@dberenbaum, with |
What's the status @skshetry? |
I think |
Also, I noticed |
@jorgeorpinel See my response in iterative/dvc.org#3071 (comment). |
What about generating a dvc.yaml without |
related: #7130 |
Parent issue for lightweight command for initializing experiments.
Linked issues:
exp init: configure experiment stage templates #6443Small tasks that are left to do:
init
dvc repo if uninitialized (Epic: exp init #6446 (comment))train
as a default name? (exp init: iterate on ui, tree and show to-be-generated yaml content #6750 (comment))--no-scm
or if the project is notgit init
-ed.data
as a dvc-tracked directory (Epic: exp init #6446 (comment))The text was updated successfully, but these errors were encountered: