ComE
ComE copied to clipboard
andompesta
Implementation of ComE algorithm
I clone the project from github, the runtime environment are python3.6.1, cpython0.25. The error is described below: File "E:/ComE/ADSCModel/context_embeddings.py", line 15, in from utils.training_sdg_inner import train_o2, loss_o2, FAST_VERSION ModuleNotFoundError: No...
I have the trouble "name '__current_graph' is not defined" in function "_write_walks_to_disk(agrs)"
when i use the function "_write_walks_to_disk(args)", the terminal will hint "name '__current_graph' is not defined". I wonder if global variables cannot be invoked in the process.
## Update 1. Fixed `object of type 'dict_keyiterator' has no len()` error. 2. Fixed `name '__current_graph' is not defined` error.
On p.381 in algorithm 1 line 3 of the ComE paper, it states that multiple iterations are run: updating, in turn, community detection/embedding with EM and node embedding with SGD....
"As of conda 4.7, the free channel was removed from conda’s default channels." https://docs.anaconda.com/anaconda/user-guide/tasks/using-repositories/#free-channel If you're using conda >=4.7, you need to add the `free` channel to /env.yml for creating...
Nice work to follow. However, I didn't see the comparison against traditional community detection algorithms, such as louvain, LPA, infomap etc. Any opinions about the pros and cons between the...
