search

takemaru / graphillion

Fast, lightweight graphset operation library
Other
466 stars 40 forks source link

how to create a graph faslty from a BIG file with graphillion!!?? #31

Closed AsmaZbt closed 6 years ago

AsmaZbt commented 6 years ago

hello! I have a file of 1Gb , the format of each line be like this : word1 word2 distance i would like to creat an undirected graph of the related words quicly ,

i've using this function G = nx.read_edgelist("neighbours.txt", delimiter="\t", data=[("weight", float)]) it's work BUT 1h45mn to build the graph it's too much , i don't have too much time , i have lot of files to do the same process with them :(

can I do that faslty with graphillion??

takemaru commented 6 years ago

Thank you for using Graphillion. Since your question is about networkx, I recommend you to ask on the site of networkx.

Best regards,

ReblochonMasque commented 6 years ago

Maybe, Takemaru San, the question was: Can this be done with Graphillion efficiently, instead of using NetworkX?

takemaru commented 6 years ago

Sorry, I misread. Unfortunately, Graphillion isn’t faster than networkx for creating a single graph. Graphillion is designed to efficiently handle a large set of graphs (in other words, to handle complicated constraints such as s-t path).

ReblochonMasque commented 6 years ago

Thank you for this clarification!

AsmaZbt commented 6 years ago

Thank you so much , i'm sorry for the non clear question (my english is not very good) thanks for @ReblochonMasque to clarify my question , that was exactly my question.

Thanks Takemaru for the clear answer.