diff --git a/Graph_Sampling.egg-info/PKG-INFO b/Graph_Sampling.egg-info/PKG-INFO
new file mode 100644
index 0000000..da8a381
--- /dev/null
+++ b/Graph_Sampling.egg-info/PKG-INFO
@@ -0,0 +1,14 @@
+Metadata-Version: 2.1
+Name: Graph-Sampling
+Version: 0.0.1
+Summary: Graph Sampling Package
+Home-page: UNKNOWN
+Authors: [Kirti Jain,Ashish Aggarwal]
+Authors-email: [kirtij.mcs16.du@gmail.com,ashish.mcs16.du@gmail.com]
+License: UNKNOWN
+Description: A Graph Sampling package containing various approaches which samples the original graph according to different sample sizes
+Platform: UNKNOWN
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Description-Content-Type: text/markdown
diff --git a/Graph_Sampling.egg-info/SOURCES.txt b/Graph_Sampling.egg-info/SOURCES.txt
new file mode 100644
index 0000000..2c8ee93
--- /dev/null
+++ b/Graph_Sampling.egg-info/SOURCES.txt
@@ -0,0 +1,12 @@
+README.md
+setup.py
+Graph_Sampling/ForestFire.py
+Graph_Sampling/MHRW.py
+Graph_Sampling/SRW_RWF_ISRW.py
+Graph_Sampling/Snowball.py
+Graph_Sampling/TIES.py
+Graph_Sampling/__init__.py
+Graph_Sampling.egg-info/PKG-INFO
+Graph_Sampling.egg-info/SOURCES.txt
+Graph_Sampling.egg-info/dependency_links.txt
+Graph_Sampling.egg-info/top_level.txt
\ No newline at end of file
diff --git a/sampling.egg-info/dependency_links.txt b/Graph_Sampling.egg-info/dependency_links.txt
similarity index 100%
rename from sampling.egg-info/dependency_links.txt
rename to Graph_Sampling.egg-info/dependency_links.txt
diff --git a/Graph_Sampling.egg-info/top_level.txt b/Graph_Sampling.egg-info/top_level.txt
new file mode 100644
index 0000000..da67eec
--- /dev/null
+++ b/Graph_Sampling.egg-info/top_level.txt
@@ -0,0 +1 @@
+Graph_Sampling
diff --git a/Graph_Sampling/ForestFire.py b/Graph_Sampling/ForestFire.py
new file mode 100644
index 0000000..8687f4b
--- /dev/null
+++ b/Graph_Sampling/ForestFire.py
@@ -0,0 +1,44 @@
+import random
+import networkx as nx
+# import matplotlib.pyplot as plt
+# import time
+
+
+# G : Original Graph
+# size : size of the sampled graph
+class ForestFire():
+ def __init__(self):
+ self.G1 = nx.Graph()
+
+ def forestfire(self, G, size):
+ list_nodes = list(G.nodes())
+ # print(len(G))
+ dictt = set()
+ random_node = random.sample(set(list_nodes), 1)[0]
+ # print(random_node)
+ q = set() # q = set contains the distinct values
+ q.add(random_node)
+ while(len(self.G1.nodes()) < size):
+ if(len(q) > 0):
+ initial_node = q.pop()
+ if(initial_node not in dictt):
+ # print(initial_node)
+ dictt.add(initial_node)
+ neighbours = list(G.neighbors(initial_node))
+ # print(list(G.neighbors(initial_node)))
+ np = random.randint(1, len(neighbours))
+ # print(np)
+ # print(neighbours[:np])
+ for x in neighbours[:np]:
+ if(len(self.G1.nodes()) < size):
+ self.G1.add_edge(initial_node, x)
+ q.add(x)
+ else:
+ break
+ else:
+ continue
+ else:
+ random_node = random.sample(set(list_nodes) and dictt, 1)[0]
+ q.add(random_node)
+ q.clear()
+ return self.G1
diff --git a/Graph_Sampling/MHRW.py b/Graph_Sampling/MHRW.py
new file mode 100644
index 0000000..6006860
--- /dev/null
+++ b/Graph_Sampling/MHRW.py
@@ -0,0 +1,52 @@
+import random
+# import time
+import networkx as nx
+# import matplotlib.pyplot as plt
+
+
+class MHRW():
+ def __init__(self):
+ self.G1 = nx.Graph()
+
+ def mhrw(self, G, node, size):
+ dictt = {}
+ node_list = set()
+ node_list.add(node)
+ parent_node = node_list.pop()
+ dictt[parent_node] = parent_node
+ degree_p = G.degree(parent_node)
+ related_list = list(G.neighbors(parent_node))
+ node_list.update(related_list)
+
+ while(len(self.G1.nodes()) < size):
+ if(len(node_list) > 0):
+ child_node = node_list.pop()
+ p = round(random.uniform(0, 1), 4)
+ if(child_node not in dictt):
+ related_listt = list(G.neighbors(child_node))
+ degree_c = G.degree(child_node)
+ dictt[child_node] = child_node
+ if(p <= min(1, degree_p / degree_c) and child_node in list(G.neighbors(parent_node))):
+ self.G1.add_edge(parent_node, child_node)
+ parent_node = child_node
+ degree_p = degree_c
+ node_list.clear()
+ node_list.update(related_listt)
+ else:
+ del dictt[child_node]
+
+ # node_list set becomes empty or size is not reached
+ # insert some random nodes into the set for next processing
+ else:
+ node_list.update(random.sample(set(G.nodes()) - set(self.G1.nodes()), 3))
+ parent_node = node_list.pop()
+ G.add_node(parent_node)
+ related_list = list(G.neighbors(parent_node))
+ node_list.clear()
+ node_list.update(related_list)
+ return self.G1
+
+ def induced_mhrw(self, G, size, node):
+ sampled_graph = mhrw(self.G1, G, size, node)
+ induced_graph = G.subgraph(sampled_graph.nodes())
+ return induced_graph
diff --git a/Graph_Sampling/SRW_RWF_ISRW.py b/Graph_Sampling/SRW_RWF_ISRW.py
new file mode 100644
index 0000000..2e6ce3c
--- /dev/null
+++ b/Graph_Sampling/SRW_RWF_ISRW.py
@@ -0,0 +1,118 @@
+import random
+# import time
+# import datetime
+# import io
+# import array, re, itertools
+import numpy as np
+import networkx as nx
+# import matplotlib.pyplot as plt
+# from itertools import groupby
+
+
+class SRW_RWF_ISRW:
+
+ def __init__(self):
+ self.growth_size = 2
+ self.T = 100 # number of iterations
+ # with a probability (1-fly_back_prob) select a neighbor node
+ # with a probability fly_back_prob go back to the initial vertex
+ self.fly_back_prob = 0.15
+
+ def random_walk_sampling_simple(self, complete_graph, nodes_to_sample):
+ complete_graph = nx.convert_node_labels_to_integers(complete_graph, 0, 'default', True)
+ # giving unique id to every node same as built-in function id
+ for n, data in complete_graph.nodes(data=True):
+ complete_graph.nodes[n]['id'] = n
+
+ nr_nodes = len(complete_graph.nodes())
+ upper_bound_nr_nodes_to_sample = nodes_to_sample
+ index_of_first_random_node = random.randint(0, nr_nodes - 1)
+ sampled_graph = nx.Graph()
+
+ sampled_graph.add_node(complete_graph.nodes[index_of_first_random_node]['id'])
+
+ iteration = 1
+ edges_before_t_iter = 0
+ curr_node = index_of_first_random_node
+ while sampled_graph.number_of_nodes() != upper_bound_nr_nodes_to_sample:
+ edges = [n for n in complete_graph.neighbors(curr_node)]
+ index_of_edge = random.randint(0, len(edges) - 1)
+ chosen_node = edges[index_of_edge]
+ sampled_graph.add_node(chosen_node)
+ sampled_graph.add_edge(curr_node, chosen_node)
+ curr_node = chosen_node
+ iteration = iteration + 1
+
+ if iteration % self.T == 0:
+ if ((sampled_graph.number_of_edges() - edges_before_t_iter) < self.growth_size):
+ curr_node = random.randint(0, nr_nodes - 1)
+ edges_before_t_iter = sampled_graph.number_of_edges()
+ return sampled_graph
+
+ def random_walk_sampling_with_fly_back(self, complete_graph, nodes_to_sample, fly_back_prob):
+ complete_graph = nx.convert_node_labels_to_integers(complete_graph, 0, 'default', True)
+ # giving unique id to every node same as built-in function id
+ for n, data in complete_graph.nodes(data=True):
+ complete_graph.nodes[n]['id'] = n
+
+ nr_nodes = len(complete_graph.nodes())
+ upper_bound_nr_nodes_to_sample = nodes_to_sample
+
+ index_of_first_random_node = random.randint(0, nr_nodes - 1)
+ sampled_graph = nx.Graph()
+
+ sampled_graph.add_node(complete_graph.nodes[index_of_first_random_node]['id'])
+
+ iteration = 1
+ edges_before_t_iter = 0
+ curr_node = index_of_first_random_node
+ while sampled_graph.number_of_nodes() != upper_bound_nr_nodes_to_sample:
+ edges = [n for n in complete_graph.neighbors(curr_node)]
+ index_of_edge = random.randint(0, len(edges) - 1)
+ chosen_node = edges[index_of_edge]
+ sampled_graph.add_node(chosen_node)
+ sampled_graph.add_edge(curr_node, chosen_node)
+ choice = np.random.choice(['prev', 'neigh'], 1, p=[fly_back_prob, 1 - fly_back_prob])
+ if choice == 'neigh':
+ curr_node = chosen_node
+ iteration = iteration + 1
+
+ if iteration % self.T == 0:
+ if ((sampled_graph.number_of_edges() - edges_before_t_iter) < self.growth_size):
+ curr_node = random.randint(0, nr_nodes - 1)
+ print("Choosing another random node to continue random walk ")
+ edges_before_t_iter = sampled_graph.number_of_edges()
+
+ return sampled_graph
+
+ def random_walk_induced_graph_sampling(self, complete_graph, nodes_to_sample):
+ complete_graph = nx.convert_node_labels_to_integers(complete_graph, 0, 'default', True)
+ # giving unique id to every node same as built-in function id
+ for n, data in complete_graph.nodes(data=True):
+ complete_graph.nodes[n]['id'] = n
+
+ nr_nodes = len(complete_graph.nodes())
+ upper_bound_nr_nodes_to_sample = nodes_to_sample
+ index_of_first_random_node = random.randint(0, nr_nodes - 1)
+
+ Sampled_nodes = set([complete_graph.nodes[index_of_first_random_node]['id']])
+
+ iteration = 1
+ nodes_before_t_iter = 0
+ curr_node = index_of_first_random_node
+ while len(Sampled_nodes) != upper_bound_nr_nodes_to_sample:
+ edges = [n for n in complete_graph.neighbors(curr_node)]
+ index_of_edge = random.randint(0, len(edges) - 1)
+ chosen_node = edges[index_of_edge]
+ Sampled_nodes.add(complete_graph.nodes[chosen_node]['id'])
+ curr_node = chosen_node
+ iteration = iteration + 1
+
+ if iteration % self.T == 0:
+ if ((len(Sampled_nodes) - nodes_before_t_iter) < self.growth_size):
+ curr_node = random.randint(0, nr_nodes - 1)
+ nodes_before_t_iter = len(Sampled_nodes)
+
+ sampled_graph = complete_graph.subgraph(Sampled_nodes)
+
+ return sampled_graph
diff --git a/Graph_Sampling/Snowball.py b/Graph_Sampling/Snowball.py
new file mode 100644
index 0000000..63b5c09
--- /dev/null
+++ b/Graph_Sampling/Snowball.py
@@ -0,0 +1,78 @@
+# import json
+# import sys
+import random
+# import math
+# import time
+import networkx as nx
+# import matplotlib.pyplot as plt
+# from collections import defaultdict
+
+
+class Queue():
+ # Constructor creates a list
+ def __init__(self):
+ self.queue = list()
+
+ # Adding elements to queue
+ def enqueue(self, data):
+ # Checking to avoid duplicate entry (not mandatory)
+ if data not in self.queue:
+ self.queue.insert(0, data)
+ return True
+ return False
+
+ # Removing the last element from the queue
+ def dequeue(self):
+ if len(self.queue) > 0:
+ return self.queue.pop()
+ else:
+ # plt.show()
+ exit()
+
+ # Getting the size of the queue
+ def size(self):
+ return len(self.queue)
+
+ # printing the elements of the queue
+ def printQueue(self):
+ return self.queue
+
+
+class Snowball():
+
+ def __init__(self):
+ self.G1 = nx.Graph()
+
+ def snowball(self, G, size, k):
+ q = Queue()
+ list_nodes = list(G.nodes())
+ m = k
+ dictt = set()
+ while(m):
+ id = random.sample(list(G.nodes()), 1)[0]
+ q.enqueue(id)
+ m = m - 1
+ # print(q.printQueue())
+ while(len(self.G1.nodes()) <= size):
+ if(q.size() > 0):
+ id = q.dequeue()
+ self.G1.add_node(id)
+ if(id not in dictt):
+ dictt.add(id)
+ list_neighbors = list(G.neighbors(id))
+ if(len(list_neighbors) > k):
+ for x in list_neighbors[:k]:
+ q.enqueue(x)
+ self.G1.add_edge(id, x)
+ elif(len(list_neighbors) <= k and len(list_neighbors) > 0):
+ for x in list_neighbors:
+ q.enqueue(x)
+ self.G1.add_edge(id, x)
+ else:
+ continue
+ else:
+ initial_nodes = random.sample(list(G.nodes()) and list(dictt), k)
+ no_of_nodes = len(initial_nodes)
+ for id in initial_nodes:
+ q.enqueue(id)
+ return self.G1
diff --git a/Graph_Sampling/TIES.py b/Graph_Sampling/TIES.py
new file mode 100644
index 0000000..c1c5ae9
--- /dev/null
+++ b/Graph_Sampling/TIES.py
@@ -0,0 +1,40 @@
+import random
+import networkx as nx
+# import matplotlib.pyplot as plt
+import math
+# import time
+# import csv
+# from datetime import datetime
+
+
+class TIES():
+ def __init__(self):
+ self.G1 = nx.Graph()
+
+ def ties(self, G, size, phi):
+ V = G.nodes()
+ # Calculate number of nodes in Graph G
+ Vs = []
+ # Empty list Vs
+ phi = round((phi * 0.01), 2)
+ while (len(Vs)) <= math.floor(phi * len(V)):
+ # Loops run till sample size * length of V where V is number of nodes in graph as calculated above.
+ edges_sample = random.sample(G.edges(), 1)
+ # Randomly samples one edge from a graph at a time
+ for a1, a2 in edges_sample:
+ # Nodes corresponding to sample edge are retrieved and added in Graph G1
+ self.G1.add_edge(a1, a2)
+ if (a1 not in Vs):
+ Vs.append(a1)
+ if (a2 not in Vs):
+ Vs.append(a2)
+ # Statement written just to have a check of a program
+
+ for x in self.G1.nodes():
+ neigh = (set(self.G1.nodes()) & set(list(G.neighbors(x))))
+ # Check neighbours of sample node and if the nodes are their in sampled set then edge is included between them.
+ for y in neigh:
+ # Check for every node's neighbour in sample set of nodes
+ self.G1.add_edge(x, y)
+ # Add edge between the sampled nodes
+ return self.G1
diff --git a/Graph_Sampling/__init__.py b/Graph_Sampling/__init__.py
new file mode 100644
index 0000000..c0c6cfa
--- /dev/null
+++ b/Graph_Sampling/__init__.py
@@ -0,0 +1,5 @@
+from Graph_Sampling.SRW_RWF_ISRW import SRW_RWF_ISRW
+from Graph_Sampling.Snowball import Snowball, Queue
+from Graph_Sampling.ForestFire import ForestFire
+from Graph_Sampling.MHRW import MHRW
+from Graph_Sampling.TIES import TIES
\ No newline at end of file
diff --git a/Graph_Sampling/__pycache__/ForestFire.cpython-37.pyc b/Graph_Sampling/__pycache__/ForestFire.cpython-37.pyc
new file mode 100644
index 0000000..c84d04a
Binary files /dev/null and b/Graph_Sampling/__pycache__/ForestFire.cpython-37.pyc differ
diff --git a/Graph_Sampling/__pycache__/MHRW.cpython-37.pyc b/Graph_Sampling/__pycache__/MHRW.cpython-37.pyc
new file mode 100644
index 0000000..85bb014
Binary files /dev/null and b/Graph_Sampling/__pycache__/MHRW.cpython-37.pyc differ
diff --git a/Graph_Sampling/__pycache__/SRW_RWF_ISRW.cpython-37.pyc b/Graph_Sampling/__pycache__/SRW_RWF_ISRW.cpython-37.pyc
new file mode 100644
index 0000000..77836e2
Binary files /dev/null and b/Graph_Sampling/__pycache__/SRW_RWF_ISRW.cpython-37.pyc differ
diff --git a/Graph_Sampling/__pycache__/Snowball.cpython-37.pyc b/Graph_Sampling/__pycache__/Snowball.cpython-37.pyc
new file mode 100644
index 0000000..d5c38ac
Binary files /dev/null and b/Graph_Sampling/__pycache__/Snowball.cpython-37.pyc differ
diff --git a/Graph_Sampling/__pycache__/TIES.cpython-37.pyc b/Graph_Sampling/__pycache__/TIES.cpython-37.pyc
new file mode 100644
index 0000000..73f16c4
Binary files /dev/null and b/Graph_Sampling/__pycache__/TIES.cpython-37.pyc differ
diff --git a/Graph_Sampling/__pycache__/__init__.cpython-37.pyc b/Graph_Sampling/__pycache__/__init__.cpython-37.pyc
new file mode 100644
index 0000000..d41dcc6
Binary files /dev/null and b/Graph_Sampling/__pycache__/__init__.cpython-37.pyc differ
diff --git a/README.md b/README.md
index 1213122..d8c1a09 100644
--- a/README.md
+++ b/README.md
@@ -52,30 +52,31 @@ Edge sampling focuses on the selection of edges rather than nodes to populate th
### Pre-requisite
-The Graph Sampling package requires [Python](https://www.python.org/downloads/) 2.7, 3.4, 3.5, 3.6 or newer. If you don't have the pre-installed python in your system, please follow up the python link to download it. This package also requires [Networkx](https://networkx.github.io/documentation/latest/install.html) 2.1 or newer which helps to create the graphs and also perform manipulations on them.
+The Graph Sampling package requires [Python](https://www.python.org/downloads/) 3.X . If you don't have the pre-installed python in your system, please follow up the python link to download it. This package also requires [Networkx](https://networkx.github.io/documentation/latest/install.html) 2.1 or newer which helps to create the graphs and also perform manipulations on them.
### Installing the development version
If you have Git installed on your system, then it is also possible to install the development version of Graph Sampling package by running these commands on your terminal:
```sh
-$ git clone https://github.com/Ashish7129/Graph-Sampling.git
-$ cd Sampling
+$ git clone https://github.com/Ashish7129/Graph_Sampling.git
+$ cd Graph_Sampling
$ pip install -e .
```
Or you can install the current release of Graph Sampling package with pip. Please download the zip file and locate it into the current folder and then run the following command for installing the graph sampling package into your system:
```sh
-$ pip install Sampling
+$ python setup.py sdist bdist_wheel
+$ pip install dist/Graph_Sampling-0.0.1-py3-none-any.whl
```
### Usage
After installing the package, you can use the package by writing the following command:
```sh
->>> import Sampling
+>>> import Graph_Sampling
```
### Example
Check out the file test.py, which helps you to understand the procedure of executing different functions along with their type and number of arguments. For example, snowball sampling fuction is excecuted as follows:
```sh
->>> object = Sampling.Snowball()
+>>> object = Graph_Sampling.Snowball()
>>> sampled_subgraph = object.snowball(G,size,k)
```
*The object is the instance of the class Snowball. The class having the snowball function has 3 parameters as :*
diff --git a/sampling/ForestFire.py b/build/lib/Graph_Sampling/ForestFire.py
similarity index 100%
rename from sampling/ForestFire.py
rename to build/lib/Graph_Sampling/ForestFire.py
diff --git a/sampling/MHRW.py b/build/lib/Graph_Sampling/MHRW.py
similarity index 100%
rename from sampling/MHRW.py
rename to build/lib/Graph_Sampling/MHRW.py
diff --git a/sampling/SRW_RWF_ISRW.py b/build/lib/Graph_Sampling/SRW_RWF_ISRW.py
similarity index 97%
rename from sampling/SRW_RWF_ISRW.py
rename to build/lib/Graph_Sampling/SRW_RWF_ISRW.py
index bc1564a..08fa5db 100644
--- a/sampling/SRW_RWF_ISRW.py
+++ b/build/lib/Graph_Sampling/SRW_RWF_ISRW.py
@@ -8,7 +8,7 @@
import matplotlib.pyplot as plt
from itertools import groupby
-class SRW_RWF_ISRW():
+class SRW_RWF_ISRW:
def __init__(self):
self.growth_size = 2
@@ -79,7 +79,7 @@ def random_walk_sampling_with_fly_back(self,complete_graph, nodes_to_sample, fly
if iteration % self.T == 0:
if ((sampled_graph.number_of_edges() - edges_before_t_iter) < self.growth_size):
curr_node = random.randint(0, nr_nodes-1)
- print "Choosing another random node to continue random walk "
+ print ("Choosing another random node to continue random walk ")
edges_before_t_iter = sampled_graph.number_of_edges()
return sampled_graph
diff --git a/sampling/Snowball.py b/build/lib/Graph_Sampling/Snowball.py
similarity index 100%
rename from sampling/Snowball.py
rename to build/lib/Graph_Sampling/Snowball.py
diff --git a/sampling/TIES.py b/build/lib/Graph_Sampling/TIES.py
similarity index 92%
rename from sampling/TIES.py
rename to build/lib/Graph_Sampling/TIES.py
index 3b91cea..1c65d14 100644
--- a/sampling/TIES.py
+++ b/build/lib/Graph_Sampling/TIES.py
@@ -16,13 +16,10 @@ def ties(self,G,size,phi):
Vs = []
#Empty list Vs
phi = round((phi * 0.01), 2)
- print "here1"
while (len(Vs)) <