-
Notifications
You must be signed in to change notification settings - Fork 4
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
perf: improve performance of model & forward layer #616
perf: improve performance of model & forward layer #616
Conversation
…olumns => 0 rows
Co-authored-by: WinPlay02 <[email protected]>
🦙 MegaLinter status: ✅ SUCCESS
See detailed report in MegaLinter reports |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #616 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 61 62 +1
Lines 4609 4634 +25
=========================================
+ Hits 4609 4634 +25 ☔ View full report in Codecov by Sentry. |
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.
Just a friendly reminder (as I do not know about the planing in the Library group): Please add __hash__
, __eq__
and __sizeof__
implementations where possible (as they are really valuable for running the library with the memoization infrastructure provided by the runner)
E.g. for FNNLayer
, input_size
and output_size
could be used to construct a hash
Nothing of this needs to happen in this PR, this is just a reminder
I just wanna start a discussion here, so mby we should start an extra discssion thread. Should there be an abstract Layer class, which contains all signatures of the Layer class, so the User gets better Feedback on the function calls. Aswell I question myself, do we want the user to change the activation function by himself?, or do we keep only internal handle. |
…able_dataset' into 610-improve-fnn-layer-and-model-performance-and-remove-some-bugs # Conflicts: # src/safeds/data/tabular/containers/_tagged_table.py # src/safeds/ml/nn/_model.py
…performance-and-remove-some-bugs' into 610-improve-fnn-layer-and-model-performance-and-remove-some-bugs
…performance-and-remove-some-bugs' into 610-improve-fnn-layer-and-model-performance-and-remove-some-bugs # Conflicts: # src/safeds/ml/nn/_model.py
…performance-and-remove-some-bugs' into 610-improve-fnn-layer-and-model-performance-and-remove-some-bugs # Conflicts: # src/safeds/ml/nn/_model.py
…and-remove-some-bugs
## [0.21.0](v0.20.0...v0.21.0) (2024-04-17) ### Features * add ARIMA model ([#577](#577)) ([8b9c7a9](8b9c7a9)), closes [#570](#570) * Add ImageList class ([#534](#534)) ([3cb74a2](3cb74a2)), closes [#528](#528) [#599](#599) [#600](#600) * more hash, sizeof and eq implementations ([#609](#609)) ([2bc0b0a](2bc0b0a)) ### Performance Improvements * Add special case to `Table.add_rows` to increase performance ([#608](#608)) ([ffb8304](ffb8304)), closes [#606](#606) * improve performance of model & forward layer ([#616](#616)) ([e856cd5](e856cd5)), closes [#610](#610) * lazily import our modules and external libraries ([#624](#624)) ([20fc313](20fc313)) * treat Tables specially when calling add_rows ([#606](#606)) ([e555b85](e555b85)), closes [#575](#575)
🎉 This PR is included in version 0.21.0 🎉 The release is available on:
Your semantic-release bot 📦🚀 |
@Safe-DS/library For future PRs, please separate renamings of public API elements from internal changes. One is a minor version bump (later even a major version bump once we have a 1.x.x release), the other is a patch version bump. |
Closes #610
Summary of Changes
Fixed some bugs and improved the performance of some methods, there are still some changes to be made but it is helpful to merge this now as @Marsmaennchen221 and @Gerhardsa0 partly depend on it