[Doc/Readme] Steps to convert ParMetis partition to dgl graph
- Dominant language
- Python
- Stars
- 14.3k
- Forks
- 3.1k
- PR merge metrics
- No merged PRs in 30d
Description
## 🐛 Bug
## To Reproduce
Steps to reproduce the behavior:
I'm following the instructions to convert the ParMetis format to dgl graph but I fail on the third steps: https://github.com/dmlc/dgl/tree/master/examples/pytorch/rgcn/experimental#step-3-convert-the-parmetis-partitions-into-dglgraph
1. dgl/tools/convert_partition.py This file doesn't exist. dgl/tools/distpartitioning/convert_partition.py is a module. Is this the right file to call: https://github.com/dmlc/dgl/blob/master/tools/distpartitioning/parmetis_postprocess.py?
2. I also notice that the parmetis_wrapper calls pm_dglpart3 but all the other documentation calls pm_dglpart https://github.com/dmlc/dgl/blob/f635e2a11eb4d74618594fc99aadbc3b3d3f4111/tools/distpartitioning/parmetis_wrapper.py#L68 Which one should we call?
I also open a discussion here: https://discuss.dgl.ai/t/convert-parmetis-partition-to-dgl-graph/3646 Not sure whether it's an issue or a general question. But I think that Readme should be update since that file doesn't exist right now.
## Expected behavior
## Environment
- DGL Version (e.g., 1.0): dgl master
- Backend Library & Version (e.g., PyTorch 0.4.1, MXNet/Gluon 1.3): Pytorch
- OS (e.g., Linux): Linux
- How you installed DGL (`conda`, `pip`, source): conda
- Build command you used (if compiling from source):
- Python version: 3.9
- CUDA/cuDNN version (if applicable):
- GPU models and configuration (e.g. V100):
- Any other relevant information:
I'm using the parallel metis to partition the graph.
## Additional context
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.