graphsage - use_id embedding
Open
- Dominant language
- C++
- Stars
- 2.9k
- Forks
- 553
- PR merge metrics
- No merged PRs in 30d
Description
我发现用graphsage use_id后，会增加一个embedding layer. embedding 参数特别大。而且在训练的时候，没有用paritionedVariable造成PS特别慢。想了解一下为何这样？（我打算自己加上partitioned varaiable)
Contributor guide
No contributing guide indexed for this repository
Research direction
The issue names GraphSAGE's use_id path, the added embedding layer, and PS training with PartitionedVariable, but no file or test. Trace that path first, compare the embedding parameters and training behavior with and without partitioning, and treat an explanation plus a validated partitioned-variable change as done.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- distributed-systems, machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100