Skip to content
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

Fix JIT compilation with TensorFlow >= 2.14.0 #197

Merged
merged 1 commit into from
Dec 29, 2023

Commits on Dec 25, 2023

  1. Fix JIT compilation with TensorFlow >= 2.14.0

    The previous check if code is currently being compiled no longer works with new TensorFlow versions because the `Tensor` type is now called `SymbolicTensor`.
    
    This change adds a helper function to check if code is being compiled for JAX, TensorFlow or PyTorch.
    If tf.is_symbolic_tensor() is available, i.e. if the TensorFlow version is high enough,
    we use this function to check if code is being compiled.
    
    To avoid inconsistencies between backends,
    the check for integration domain values is disabled if code is being compiled with PyTorch even if the check works with PyTorch.
    FHof committed Dec 25, 2023
    Configuration menu
    Copy the full SHA
    ce7cf11 View commit details
    Browse the repository at this point in the history