Implement model instantiators using dynamic execution of Python code (#…

…186) * Define deterministic models using dynamic execution of Python code * Implement all model instantiators using dynamic execution of Python code in torch * Implement all model instantiators using dynamic execution of Python code in jax * Add model instantiators test file in torch * Add model instantiators test file in jax * Add parameter to return the source class definition used to instantiate the model * Update annotation and doctring * Update CHANGELOG
Toni-SM · Aug 13, 2024 · 568313a · 568313a
1 parent 82ea390
commit 568313a
Show file tree

Hide file tree

Showing 5 changed files with 690 additions and 451 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -6,10 +6,12 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 ### Added
 - Distributed multi-GPU and multi-node learning (JAX implementation)
 - Utilities to start multiple processes from a single program invocation for distributed learning using JAX
+- Model instantiators `return_source` parameter to get the source class definition used to instantiate the models
 
 ### Changed
 - Move the KL reduction from the PyTorch `KLAdaptiveLR` class to each agent using it in distributed runs
 - Move the PyTorch distributed initialization from the agent base class to the ML framework configuration
+- Implement model instantiators using dynamic execution of Python code
 
 ### Fixed
 - Catch TensorBoard summary iterator exceptions in `TensorboardFileIterator` postprocessing utils