quiver-team/torch-quiver

Issues

• Does the feature caching module support different GPU cache redundant data？

  #167 opened 8 months ago by Baizx98
  0

• Does Quiver support heterogeneous graphs?

  #166 opened 9 months ago by jasperzhong
  1

• There is an error when quiver's features are applied to the ogb-papers00M dataset on DGL

  #160 opened a year ago by lzc17
  2

• CPU only sample will hang

  #165 opened a year ago by LukeLIN-web
  1

• os.sched_setaffinity(0, [2*(cpu_offset+rank)]) OSError: [Errno 22] Invalid argument

  #162 opened a year ago by LukeLIN-web
  2

• What is the difference between test_request_from_local_preparation and test_request_from_local

  #163 opened a year ago by LukeLIN-web
  4

• Test Serving Fail device = rank%len(device_list) ZeroDivisionError: integer division or modulo by zero

  #161 opened a year ago by LukeLIN-web
  2

• Benchmark mag240m, preprocessing issues

  #157 opened a year ago by Juelin-Liu
  1

• Slow installation

  #156 opened a year ago by LukeLIN-web
  0

• Need pypi version

  #155 opened a year ago by LukeLIN-web
  0

• How to develop quiver

  #139 opened a year ago by LukeLIN-web
  1

• quiver sampler has no rand?

  #140 opened 2 years ago by ZonghaoWang
  1

• 是否有与DGL采样的对比，性能如何？

  #141 opened 2 years ago by longmingg
  0

• Provide a newer dockerfile

  #136 opened 2 years ago by LukeLIN-web
  1

• something about loading `sort_feature.pt` in Paper100MDataset

  #138 opened 2 years ago by zhaoyin214
  1

• benchmark ogbn-papers100M, CUDA free failed

  #137 opened 2 years ago by zhaoyin214
  0

• Quiver multi-node ogbn-mag240m benchmark

  #134 opened 2 years ago by prenc
  0

• module 'torch_quiver' has no attribute 'device_quiver_from_csr_array'

  #131 opened 2 years ago by xcwanAndy
  4

• AttributeError: module 'torch_quiver' has no attribute 'device_quiver_from_csr_array'

  #133 opened 2 years ago by LukeLIN-web
  5

• A strange phenomenon about quiver sampler

  #122 opened 3 years ago by DesmonDay
  6

• How to run quiver on server with complex GPU topology?

  #135 opened 2 years ago by JIESUN233
  0

• Dockerfile should put in same directory with setup.py

  #132 opened 2 years ago by LukeLIN-web
  0

• Does Quiver support multi-node training now?

  #130 opened 2 years ago by xcwanAndy
  2

• support FP16 in ShardTensor

  #128 opened 2 years ago by eedalong
  0

• Installation Problem: which GNU version does torch-quiver require?

  #126 opened 2 years ago by initzhang
  2

• Partitions hit rate imbalance

  #123 opened 3 years ago by AustinCheang
  0

• An error occurred when installing torch-quiver.

  #121 opened 3 years ago by l1nkr
  3

• Support distribute feature collection

  #120 opened 3 years ago by eedalong
  0

• Generate partition results folder

  #112 opened 3 years ago by eedalong
  0

• 有个疑问，Quiver采样的时候在batch_size或者采样的节点比较小的时候，性能加速不明显，如何进行加速了？

  #111 opened 3 years ago by SoupFree
  1

• Use memory pool for memory allocation when sampling && feature collecting

  #115 opened 3 years ago by eedalong
  3

• Add docs for quiver's partition algorithm

  #119 opened 3 years ago by eedalong
  0

• 能否在 Readme 提供一下C++编译流程脚本

  #117 opened 3 years ago by hadoopmore
  1

• Support write operation for ShardTensor

  #116 opened 3 years ago by eedalong
  0

• TODO for 0.2.0 Release

  #65 opened 3 years ago by ZenoTan
  0

• Update Reddit/Products Example

  #107 opened 3 years ago by eedalong
  0

• Failed to build torch-quiver

  #110 opened 3 years ago by decoherencer
  7

• CUDA

  #104 opened 3 years ago by yingtongxiong
  2

• Cuda Error, When run the quiver example with the dataset of ogbn-papers100M in the benchmark

  #102 opened 3 years ago by SoupFree
  5

• Typo in Introduction_cn.md

  #105 opened 3 years ago by huwan
  1

• poor scalability when using multiple gpus

  #99 opened 3 years ago by Joeyzhouqihui
  4

• The strategy of hot/cold feature allocation in the different device

  #84 opened 3 years ago by hadoopmore
  7

• Why cuda stream?

  #101 opened 3 years ago by hadoopmore
  2

• Error when cache size is larger than the size of feature

  #97 opened 3 years ago by Joeyzhouqihui
  6

• AttributeError: 'Feature' object has no attribute 'dim'

  #81 opened 3 years ago by lausannel
  1

• support random_walk in pyg

  #92 opened 3 years ago by eedalong
  0

• Memory leak

  #89 opened 3 years ago by Ningsir
  3

• Performance evaluation about dgl, pyg and torch-quiver

  #86 opened 3 years ago by Ningsir
  17

• feature collection and feature aggregation

  #83 opened 3 years ago by Ningsir
  4

• support sample_num = -1 for sampling all neighbors from seeds

  #64 opened 3 years ago by eedalong
  1