-
Notifications
You must be signed in to change notification settings - Fork 226
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
Fatal Python error: Aborted when running mnist.py example #37
Comments
This error is indicating that CUDA didn't initialize properly. The JAX equivalent of the above TF configuration can be found here: One other thing that might be worth trying is encasing all the TFDS logic in a I'm fairly confident that this is a configuration issue; the folks in google/jax may be more able to pinpoint the problem than me. |
Following the configuration instructions in https://jax.readthedocs.io/en/latest/gpu_memory_allocation.html worked. |
This is the solution for the case when using JAX with tensorflow datasets that use tensorflow for the preprocessing as XLA and TF both fight for the memory. Using the documentation highlighted in this issue above points out using |
Hi there! I've been trying to get familiar with the library by running some examples in the
examples/
folder. My environment was set up according to the instructions onhttps://github.com/google/jax#installation
andhttps://github.com/deepmind/dm-haiku#installation
.On running the
mnist.py
example with TensorFlow 2.1.0, aFatal Python error: Aborted
occurs. The full error message is as below:One solution I've found to this is a more commonplace solution when using TensorFlow, by inserting the code:
However, this kind of defeats the purpose if one is simply trying to use JAX/NumPy instead of TensorFlow. Not sure what else I can provide to help, please do let me know!
The text was updated successfully, but these errors were encountered: