Test for automatic augmentations with EfficientNet example and CI #4695
Commits on Mar 7, 2023
-
Automatic augmentations gallery
Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Rename the selection operator Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Use random.uniform discrete mode Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Pass kwargs directly Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Assure uniform choice of manitudes in TA Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> AutoAugment: coallsce pollicies Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Avoid unnecessary casts in color operations expressed in terms of arithm ops Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> AutoContrast, one less float32 intermediate result Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Expose select operator Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Review rework, usage lessons learnt from the expiriments Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Adjust the docs Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Better handling of non-unique names, better custom kwargs handling, improved error messaging Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Rework pretty_select and utils Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Linter complaints Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Private modules Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Simplify docs, simplify code Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Move policy to separate module Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> More (and more intelegible) docs Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Type annotations Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Improved error reporting for as_param Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> More specific callable annotations Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Polish names Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Better error messaging, polishing Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Better error-reporting, warnings, default random_sign Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Make Aug private Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Wrapper tests Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> More tests Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> More tests Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Test select Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> More tests Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com> Shape passing fix Signed-off-by: Kamil Tokarski <ktokarski@nvidia.com>
-
Add EfficientNet subset of Classification/ConvNets from JoC
Signed-off-by: Krzysztof Lecki <klecki@nvidia.com> Delete unused files from the example Signed-off-by: Krzysztof Lecki <klecki@nvidia.com> Keep only one EfficientNet layout Signed-off-by: Krzysztof Lecki <klecki@nvidia.com> Propagate defaults for EfficientNet training from yml to main.py Signed-off-by: Krzysztof Lecki <klecki@nvidia.com> Add implementation for DALI AA and TA with readme Adjust some configuration options to accomodate it. Remove the obsolete pipeline Signed-off-by: Krzysztof Lecki <klecki@nvidia.com> Add L3 test with benchmarks The test run 1 less epoch than the readme to make it a bit shorter Signed-off-by: Krzysztof Lecki <klecki@nvidia.com>
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.