Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 2 years ago.
Improve this question
In machine learning, we are constantly updating the models. But we still want a way to use our old trained model checkpoints. These checkpoints will not load if the model class has been changed with different keyword arguments etc.
What's the best way to manage "model creep" or models gradually changing?
One way is, we can associated each pytorch checkpoint with a github commit. Then load that commit to run our model again.
Another way is, every time we change the model, we implement backwards compatibility (could get messy).
Related
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 2 years ago.
Improve this question
Do I need to put my Firestore listeners in my View Controller or can I put them in a model class? Based on my research, best practices is apparently to keep the database code within separate models.
However, I need to update my UI based on database changes.
That means I should at least attach the listeners within my View Controllers and not in the model, correct?
Or am I violating best practices here?
Blessings.
Tai
Yes, you need to put your listener on your viewController or you can create another class like singleton contains listener methods and with it's methods update the Ui.This also reduce the rewriting of the same code.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 5 years ago.
Improve this question
I've seen plenty of posts on the pros and cons of the EF but finding quite hard to find anything on the relationship between EF and project management methodologies such as Waterfall SDLC and Agile?
Would anyone have any links or information on the above? Thanks.
Frameworks and methodologies are typically orthogonal.
The only (indirect) relationship is that Agile methods would have you build your application to be decoupled from the persistence mechanism altogether. In fact, you would delay even using a database until you have a user story that requires that you have one. This would force you to keep the two decoupled, leaving your options open.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
This question stems from curiosity more than anything else.
If one was to sit down and write a Perl module with the sole objective of getting that module to eventually become a part of the CORE load, what factors should they pay attention to? Here are a few that spring to mind, though I'm sure there are exceptions to every one of them:
Usefulness
Maturity
Stability
Core-modules-only dependency
Problem area being solved
Cross-platform compatibility
CPAN testers review
Reviews
Also, am I right in assuming that the Perl 5 Porters make the final call on whether to include something in the core load or not?
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 8 years ago.
Improve this question
I would like to store one large file which is updated daily.
I am interested in storing only last 'N' versions of this file.
Which version control tool is best suited for this task ?
If not version control tool, is there any other free/open source tool available for this task?
It depends whether you mind rewriting the repository periodically. If you don't, you can use Mercurial, with, for example hg convert to periodically rewrite the repos to discard older revisions.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 3 years ago.
Improve this question
Instead of using UIImageWirteToSavedAlbum, is it safe to write your image to /private/var/mobile/Media/DCIM/100APPLE/ directly?
Without consulting the actual developer contracts at all, I feel safe in saying "Count on it, you will get rejected".
(Besides, is this even possible, given the sandboxing?)
Edit: The iPhone Application Programming Guide says that "for security purposes, your application is placed in its own directory and is limited to reading and writing files in that directory only." If that isn't actually the case (i.e. you can write to /private/var/... anyway, I'm pretty sure it's still "forbidden").