Add High-Level API for Building and Training Neural Networks in Pharo #17
Conversation
This file contains hidden or 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
rename classes to remove the underscore in the names fix references to old class names. move deprecated protocol to new package Update TensorFlow to 2.3.1 Reify the tensor shape into a first order object make TFDataTypeEnum more portable Rename packages and separate tests from model Pull down MNISTFile class variables Add missing messages in order to tests run green delete file Delete MNISTFile duplicated class variables Add missing messages in MNISTFile
+ the new model of optimizers, trainer and neural network models Add a simple visualizer of a training summary Add an experimental dataset provider Add TensorFlowEnvironmentModel package and VAST-Compatibility-Model Add TensorFlowEnvironmentModelTests Add VAST compatibility messages
Add some missing classes and messages Update DatasetProvider and add MLTrainingLaboratoryModel
and delete vast specific objects
Move classes in MLNeuralNetworkModel and tests to LibTensorFlowExamplesApp
Initialize the MNIST classes on load Ensure dataset folder existance Improve error handling on MNIST dataset downloading Change the source of the MNIST dataset as is constantly unavailable
make integer scalar reciprocal test deterministic
update deprecated CI configuration
+ support attribute of list of integers
because in Pharo 9 is not the default callout API class and it doesn't work
move test assertion messages to core apps
since order of sentences may change
|
Thanks this is great! |
Member
|
Thank you, this is great. I will have a look ASAP. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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 PR introduces a Keras-style high-level API for defining, training, and evaluating neural network models in Pharo, built on top of the existing TensorFlow C++ bindings.
Main Additions:
Summary:
This contribution provides a more idiomatic and concise way to work with neural networks in Pharo, enabling rapid model experimentation with a clean, expressive API as in Keras.