google-deepmind / google-deepmind/deepmind-research
Gated Linear Network example throws RuntimeError
Open
- Dominant language
- Jupyter Notebook
- Stars
- 15.2k
- Forks
- 2.9k
- PR merge metrics
- No merged PRs in 30d
Description
When I am trying to run example in CPU, I get next error (all `*_test.py` are passed.):
>RuntimeError: Resource exhausted: Failed to allocate request for 30.12GiB (32341200016B) on device ordinal 0
Does it mean that your GLN with 200 neurons uses >30GiB RAM with MNIST dataset?
Contributor guide
Assessment
This issue has not been assessed yet.