tensorflow / tensorflow/probability
training masked autoregressive flow on multiple GPUs
Open
Nobody has claimed this yet.
question
- Dominant language
- Jupyter Notebook
- Stars
- 4.4k
- Forks
- 1.1k
- PR merge metrics
- No merged PRs in 30d
Description
Hi, I have a MAF with a large number of dimensions and I am wondering if it's possible to train it with multiple GPUs. Basically, the model consists of a chain of MAF bijectors and the idea is to alternate between async forward pass and sync upon a entering the next bijector. Is this possible? Thanks in advance!
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files or tests are identified. Start by reviewing the TensorFlow Probability MAF bijector documentation and multi-GPU training guidance, then determine whether alternating asynchronous forward passes with synchronization between bijectors is supported and what concrete change would be required.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- tensorflow
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100