sepandhaghighi / pyrgg Goto Github PK

Description

Support .tsv format

Description

Migrating from Travis/Appveyor to GitHub Actions

Description

max_edge is not working as expected.

Steps/Code to Reproduce

csv_maker('testfile',0,100,10,0,3,True,True,False,False);print(open('testfile.csv','r').read())

Expected Behavior

Give an output with at most 3 edges.

Actual Behavior

It has got 11 edges as below:

>>> csv_maker('testfile',0,100,10,0,3,True,True,False,False);print(open('testfile.csv','r').read())
11
1,7,-64
2,10,-19
2,9,-76
2,8,-94
4,6,-14
4,5,35
5,1,-66
5,3,25
6,8,-26
9,3,-26
9,10,-15

Operating System

MacOS big sur.

Python Version

Python 3.9.12

Pyrgg Version

Pyrgg 1.1

Description

Support floating-point weights

Description

Bug in generating of multigraphs

Steps/Code to Reproduce

Expected Behavior

Total Number of Edges ~ 10000

Actual Behavior

Total Number of Edges ~ 5

Operating System

Windows 10

Python Version

3.5.2

Pyrgg Version

0.8

Description

Graph-Line format

0 1:4 2:4 3:1 4:4 5:4 6:2 7:1 8:1 9:4

<head_1> <tail_1>:<weight_1> <tail_2>:<weight_2>  ... <tail_n>:<weight_n>
<head_2> <tail_1>:<weight_1> <tail_2>:<weight_2>  ... <tail_n>:<weight_n>
.
.
.
<head_n> <tail_1>:<weight_1> <tail_2>:<weight_2>  ... <tail_n>:<weight_n>

Hi,

I am trying to use pyrgg with me own python application, could you please provide an example usage of importing pyrgg as a library in python?

Description

Graph-Line format empty line should be removed

Steps/Code to Reproduce

>>> file=open('testfile.gl','r')
>>> print(file.read())

Expected Behavior

1 2:92
2 5:-155
3 10:148
4 7:-185
5 9:-128
6 1:93 9:-97
7 3:60
8 4:-44
10 10:114

Actual Behavior

1 2:92
2 5:-155
3 10:148
4 7:-185
5 9:-128
6 1:93 9:-97
7 3:60
8 4:-44
9
10 10:114

Operating System

Windows 10

Python Version

3.5.2

Pyrgg Version

0.8

Description

Python 3.12 support

Description

Change PyRGG structure to support different engines.

Description

It has been observed that some metadata, such as max-weight, min-weight, max-edge, min-edge, self-loop, and multigraph, may contain inaccuracies.

We have two options:

1. As some of the engines we are supporting lack controls on these parameters, we can remove the metadata.
2. Write some functions that can accurately calculate this data.

Description

Optimize logger function

Expected Behavior

2020-09-09 16:58:02
Filename : test2.gr
Vertices : 2
Total Edges : 4
Max Edge: x
Min Edge: x
Directed: True/False
Signed: True/False
Multigraph: True/False
Self Loop: True/False
Weighted: True/False
Max Weight: x
Min Weight: x
Elapsed Time : 00 days, 00 hour, 00 minutes, 20 seconds

Actual Behavior

2020-09-09 16:58:02.103878
Filename : test2.gr
Vertices : 2
Edges : 4
Elapsed Time : 00 days, 00 hour, 00 minutes, 20 seconds

Operating System

Windows 10

Python Version

3.5.2

Pyrgg Version

0.8

Description

Support Python 3.10

Description

Update references section

Description

DOT is a graph description language. DOT graphs are typically files with the filename extension .gv or .dot. The extension .gv is preferred, to avoid confusion with the extension dot used by versions of Microsoft Word before 2007. Various programs can process DOT files.

References:

1. https://graphviz.org/doc/info/lang.html
2. https://en.wikipedia.org/wiki/DOT_(graph_description_language)

Description

Save and load different configs in a config file

Description

When JSON files are generated using PyRGG, whether directed or not - there is no distinction in the JSON for it.

Expected Behavior

Add a new properties object to the root of the file which will contain details of the properties set in the generation such as:

• directed
• signed
• multigraph
• weighted
• self loops
• signed

All of those should be boolean variables.

Actual Behavior

Those properties can't be found currently.

Pyrgg Version

v1.0

Description

Anaconda workflow

Description

Some formats are too slow :

1. JSON
2. YAML
3. Pickle
4. ASP
5. TGF
6. UCINET

Operating System

Windows 10

Python Version

Python 3.5.2

Pyrgg Version

0.5

Description

Add Erdős–Rényi engine

References:

1. https://en.wikipedia.org/wiki/Erd%C5%91s%E2%80%93R%C3%A9nyi_model
2. https://www.geeksforgeeks.org/implementation-of-erdos-renyi-model-on-social-networks/

Description

I think pyrgg is now an executable script that takes the graph specifications and outputs a file containing a random graph.
What if someone wants to generate a random graph in their own code and have a graph object?
My suggestion is to divide the current program into two parts.

1. The functional core of Random Graph Generator in which graph specifications are passed to core functions and a Graph object is returned by functions.
2. The executable wrapper of the functional core which is same as the current interface of pyrgg, uses the functional core and stores graphs in the files.

Steps/Code to Reproduce

Maybe these steps:

1. Extract core functions and separate them for functional core
2. Design Graph data model and class
3. Connect the executable part to the core part

Expected Behavior

• If someone needs the random graph generator as a library in which its functions return randomly generated graphs, they can use the functional core of pyrgg.
• If someone just needs to execute the code and have a random graph in a file, they can use the executable part.

We need to enable support for directed graphs in this engine.

Originally posted by @sepandhaghighi in #146 (comment)

Description

GML format

Reference : https://gephi.org/users/supported-graph-formats/gml-format/

graph
[
  multigraph 0
  directed  0
  node
  [
   id A
  ]
  node
  [
   id B
  ]
  node
  [
   id C
  ]
   edge
  [
   source B
   target A
   value W1
  ]
  edge
  [
   source C
   target A
   value W2
  ]
]

Hi, are you going to support formats like MatrixMarket and SNAP?

Description

Print number of edges at the end of generation

Description

Update snapshots and screen record

Description

Currently when a new JSON graph is being generated, PyRGG generates a graph object which is redundant.

Expected Behavior

Have nodes and edges in the root of the JSON file:

{
    "nodes" : [
      {"id": "1"},
      {...}
    ],
    "edges" : [
      {    
        "source": "1",
        "target": "4",
        "weight": "8"
      },
      {...}
    ]
  }

If it is done because you are laying the ground for some future work regarding graph properties - you could maybe add a new properties objects at the JSON root.

Actual Behavior

{
  "graph" : {
    "nodes" : [
      {"id": "1"},
      {...}
    ],
    "edges" : [
      {    
        "source": "1",
        "target": "4",
        "weight": "8"
      },
      {...}
    ]
  }
}

Pyrgg Version

v1.0

• Missing statement of need: the software generates random graphs, but the user can only choose the number of vertices, min and max weight for the edges, and a range for a random number of edges. I would expect a useful random graph generator to provide some more options, like connectivity, min, max and avg degree.

• Example usage: the README.md states that the output can be used as input to other programs, but have the files generated by this program ever been used as an input to some other algorithm? how?

• Tests: no tests provided

• Other:

  • there is a "gpu" tag in paper.md, why?
  • ideally, large binary files as the ones in the datasets dir should not belong in a git repository. See: http://blog.deveo.com/storing-large-binary-files-in-git-repositories/

Description

Sometimes it can be useful if we make a visualized output which helps the user making some sort of illustration about recently made graph.

It can be shown after each time of generation in pyrgg menu.
networkX has something like this, you can see it in the link bellow:
https://www.python-course.eu/networkx.php

Description

GEXF Format

<?xml version="1.0" encoding="UTF-8"?>
<gexf xmlns="http://www.gexf.net/1.2draft" version="1.2">
    <meta lastmodifieddate="2009-03-20">
        <creator>PyRGG</creator>
        <description>File Name</description>
    </meta>
    <graph defaultedgetype="directed">
        <nodes>
            <node id="1" label="Node 1" />
            <node id="2" label="Node 2" />
        ...
        </nodes>
        <edges>
            <edge id="1" source="1" target="2" weight="400" />
         ...
        </edges>
    </graph>
</gexf>

Reference : https://github.com/gephi/gexf/wiki/Basic-Concepts#network-topology

Description

Add a github action to publish package

Rename max_range and min_range parameters to max_weight and min_weight

Description

This line is time-consuming (about 60% of the total time!!).
We need it only for tests, so I think we can bypass it by a flag in the normal run.

Description

Currently the nodes array in the JSON file is full of objects for each nodes - each containing only the id field.
I think that it would be much better to just get rid of those objects and replace them all with the IDs themselves.

Description

Speedup edge_gen function

Description

Edge number is incorrect

Steps/Code to Reproduce

Expected Behavior

Total Number of Edges : 2

Actual Behavior

Total Number of Edges : 4

Operating System

Windows 10

Python Version

3.5.2

Pyrgg Version

0.8

Description

Currently when a new JSON graph is created all numeric values are written as strings even though they can be integers.

Expected Behavior

Have a JSON file with IDs, (including "source" and "target") as integers.

Actual Behavior

All those fields are strings

Pyrgg Version

v1.0

Description

Get files number from user

Description

Add console_scripts entry point to setup.py to run PyRGG directly from the terminal.

pyrgg

instead of

python -m pyrgg

Description

In main menu each flag assigned by 1 or 2, but it seems more friendly if it changes to 0 or 1.
In second situation debugging may be more easy.
This issue may include these flags:

• weighted
• direct
• self_loop
• multigraph (which can be changed to singlegraph to seems more ordered)

Description

The generated graph in yaml format is not a DAG graph, however, during configuration phase directed option is chosen. Please let me know if there is a solution to this issue. Please have a look at the generated file using the link below :

https://drive.google.com/file/d/1o2WODs34h2pwAeOgItcmTgjxT6mA0zjD/view?usp=sharing

You can find the loop in these nodes: [[4, 7, 10], [1, 5, 9]]

Expected Behavior

Directed acyclic graph

Actual Behavior

Looped graph

Operating System

Linux

Python Version

PYthon 3.7.5

Pyrgg Version

V1.0

Description

The documentation (or at least the asciinema clip) needs to be updated to version v1.1

Description

GDF format

Reference :

1. https://github.com/GabrielTrettel/GDF_Formatter
2. https://gephi.org/users/supported-graph-formats/gdf-format/

nodedef>name VARCHAR,label VARCHAR
s1,Site number 1
s2,Site number 2
s3,Site number 3
edgedef>node1 VARCHAR,node2 VARCHAR, weight DOUBLE
s1,s2,122
s2,s3,45
s3,s2,12
s3,s1,5