Major upgrade to TF2 (v2.8) #137
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ResNets and singleCNNs are upgraded, but not the stereo models yet.
|
I've marked this PR as a draft because there are some things missing. @nietootein regarding the installation I was wondering if we can give it a try to rather install CUDA from the Nvidia conda account in the environment than systemwide? It'd be useful for non-root users. |
enhanced the predict mode data loading with mode
Reshape image shape to concat telescopes (mainly used in MAGIC) Added telescope(s) selection from command line. Makes config files per telescope types redundant. Solved bad file descriptor OSError at the end of training.
Users should use the installed version of the models by default. If they would like to customize the models they can provide the path to their models.
This commit adds several options to the command line in order to keep the config file as general as possible. As a result we can ship generic config files for each model with the installation of CTLearn. Also the user can select the model from the command line without providing an own config file. Developers can still use all the features as before.
Energy labels had shape (batch_size, ), now it is (batch_size, label_shape), as it is for particle type labels and direction labels.
Set energy label shape to (batch_size, label_shape)
|
Ready from my side! |
Some are deprecated due to new functionality. They will keep in the repo for now in case users wants to get inspired by them and write their own scripts.
nietootein
approved these changes
Mar 31, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This is a PR draft for the CTLearn v0.6.0 using Keras with the latest TensorFlow v.2.8.
ToDos:
input_fn()should be replaced with a Keras batch generator.tf.distribute.MirroredStrategy