Merge pull request #1 from TheAlgorithms/master

Getting New Changes

.travis.yml

@@ -1,14 +0,0 @@
-language: python
-python:
-  - "3.2"
-  - "3.3"
-  - "3.4"
-  - "3.5"
-  - "3.6"
-  - "3.6-dev"
-
-install: 
-  - if [ "$TRAVIS_PYTHON_VERSION" == "3.2" ]; then travis_retry pip install coverage==3.7.1; fi
-  - if [ "$TRAVIS_PYTHON_VERSION" != "3.2" ]; then travis_retry pip install coverage; fi
-  - "pip install pytest pytest-cov"
-script: py.test --doctest-modules --cov ./

Graphs/A*.py

@@ -0,0 +1,101 @@
+
+grid = [[0, 1, 0, 0, 0, 0],
+        [0, 1, 0, 0, 0, 0],#0 are free path whereas 1's are obstacles
+        [0, 1, 0, 0, 0, 0],
+        [0, 1, 0, 0, 1, 0],
+        [0, 0, 0, 0, 1, 0]]
+
+'''
+heuristic = [[9, 8, 7, 6, 5, 4],
+             [8, 7, 6, 5, 4, 3],
+             [7, 6, 5, 4, 3, 2],
+             [6, 5, 4, 3, 2, 1],
+             [5, 4, 3, 2, 1, 0]]'''
+
+init = [0, 0]
+goal = [len(grid)-1, len(grid[0])-1] #all coordinates are given in format [y,x] 
+cost = 1
+
+#the cost map which pushes the path closer to the goal
+heuristic = [[0 for row in range(len(grid[0]))] for col in range(len(grid))]
+for i in range(len(grid)):    
+    for j in range(len(grid[0])):            
+        heuristic[i][j] = abs(i - goal[0]) + abs(j - goal[1])
+        if grid[i][j] == 1:
+            heuristic[i][j] = 99 #added extra penalty in the heuristic map
+
+
+#the actions we can take
+delta = [[-1, 0 ], # go up
+         [ 0, -1], # go left
+         [ 1, 0 ], # go down
+         [ 0, 1 ]] # go right
+
+
+#function to search the path
+def search(grid,init,goal,cost,heuristic):
+
+    closed = [[0 for col in range(len(grid[0]))] for row in range(len(grid))]# the referrence grid
+    closed[init[0]][init[1]] = 1
+    action = [[0 for col in range(len(grid[0]))] for row in range(len(grid))]#the action grid
+
+    x = init[0]
+    y = init[1]
+    g = 0
+    f = g + heuristic[init[0]][init[0]]
+    cell = [[f, g, x, y]]
+
+    found = False  # flag that is set when search is complete
+    resign = False # flag set if we can't find expand
+
+    while not found and not resign:
+        if len(cell) == 0:
+            resign = True
+            return "FAIL"
+        else:
+            cell.sort()#to choose the least costliest action so as to move closer to the goal
+            cell.reverse()
+            next = cell.pop()
+            x = next[2]
+            y = next[3]
+            g = next[1]
+            f = next[0]
+
+            
+            if x == goal[0] and y == goal[1]:
+                found = True
+            else:
+                for i in range(len(delta)):#to try out different valid actions
+                    x2 = x + delta[i][0]
+                    y2 = y + delta[i][1]
+                    if x2 >= 0 and x2 < len(grid) and y2 >=0 and y2 < len(grid[0]):
+                        if closed[x2][y2] == 0 and grid[x2][y2] == 0:
+                            g2 = g + cost
+                            f2 = g2 + heuristic[x2][y2]
+                            cell.append([f2, g2, x2, y2])
+                            closed[x2][y2] = 1
+                            action[x2][y2] = i
+    invpath = []
+    x = goal[0]
+    y = goal[1]
+    invpath.append([x, y])#we get the reverse path from here
+    while x != init[0] or y != init[1]:
+    	x2 = x - delta[action[x][y]][0]
+        y2 = y - delta[action[x][y]][1]
+        x = x2
+        y = y2
+        invpath.append([x, y])
+
+    path = []
+    for i in range(len(invpath)):
+    	path.append(invpath[len(invpath) - 1 - i])
+    print "ACTION MAP"
+    for i in range(len(action)):
+        print action[i]
+                  
+    return path
+    
+a = search(grid,init,goal,cost,heuristic)
+for i in range(len(a)):
+	print a[i] 
+

Graphs/Breadth_First_Search.py

@@ -1,70 +0,0 @@
-class GRAPH:
-    """docstring for GRAPH"""
-    def __init__(self, nodes):
-        self.nodes=nodes
-        self.graph=[[0]*nodes for i in range (nodes)]
-        self.visited=[0]*nodes
-
-
-    def show(self):
-
-        for i in self.graph:
-            for j in i:
-                print(j, end=' ')
-            print(' ')
-    def bfs(self,v):
-
-        visited = [False]*self.vertex
-        visited[v - 1] = True
-        print('%d visited' % (v))
-
-        queue = [v - 1]
-        while len(queue) > 0:
-            v = queue[0]
-            for u in range(self.vertex):
-                if self.graph[v][u] == 1:
-                    if visited[u]== False:
-                        visited[u] = True
-                        queue.append(u)
-                        print('%d visited' % (u +1))
-            queue.pop(0)
-
-g = Graph(10)
-
-g.add_edge(1,2)
-g.add_edge(1,3)
-g.add_edge(1,4)
-g.add_edge(2,5)
-g.add_edge(3,6)
-g.add_edge(3,7)
-g.add_edge(4,8)
-g.add_edge(5,9)
-g.add_edge(6,10)
-g.bfs(4)
-=======
-        print self.graph
-
-    def add_edge(self, i, j):
-        self.graph[i][j]=1
-        self.graph[j][i]=1
-
-    def bfs(self,s):
-        queue=[s]
-        self.visited[s]=1
-        while len(queue)!=0:
-            x=queue.pop(0)
-            print(x)
-            for i in range(0,self.nodes):
-                if self.graph[x][i]==1 and self.visited[i]==0:
-                    queue.append(i)     
-                    self.visited[i]=1
-
-n=int(input("Enter the number of Nodes : "))
-g=GRAPH(n)
-e=int(input("Enter the no of edges : "))
-print("Enter the edges (u v)")
-for i in range(0,e):
-    u,v=map(int, raw_input().split())
-    g.add_edge(u,v)
-s=int(input("Enter the source node :"))
-g.bfs(s)

Graphs/Deep_First_Search.py

@@ -1,32 +0,0 @@
-class GRAPH:
-    """docstring for GRAPH"""
-    def __init__(self, nodes):
-        self.nodes=nodes
-        self.graph=[[0]*nodes for i in range (nodes)]
-        self.visited=[0]*nodes
-
-
-    def show(self):
-        print self.graph
-
-    def add_edge(self, i, j):
-        self.graph[i][j]=1
-        self.graph[j][i]=1
-
-    def dfs(self,s):
-        self.visited[s]=1
-        print(s)
-        for i in range(0,self.nodes):
-            if self.visited[i]==0 and self.graph[s][i]==1:
-                self.dfs(i)
-            
-
-n=int(input("Enter the number of Nodes : "))
-g=GRAPH(n)
-e=int(input("Enter the no of edges : "))
-print("Enter the edges (u v)")
-for i in range(0,e):
-    u,v=map(int, raw_input().split())
-    g.add_edge(u,v)
-s=int(input("Enter the source node :"))
-g.dfs(s)

Graphs/basic-graphs.py

@@ -0,0 +1,267 @@
+# Accept No. of Nodes and edges
+n, m = map(int, raw_input().split(" "))
+
+# Initialising Dictionary of edges
+g = {}
+for i in xrange(n):
+    g[i + 1] = []
+
+"""
+--------------------------------------------------------------------------------
+    Accepting edges of Unweighted Directed Graphs
+--------------------------------------------------------------------------------
+"""
+for _ in xrange(m):
+    x, y = map(int, raw_input().split(" "))
+    g[x].append(y)
+
+"""
+--------------------------------------------------------------------------------
+    Accepting edges of Unweighted Undirected Graphs
+--------------------------------------------------------------------------------
+"""
+for _ in xrange(m):
+    x, y = map(int, raw_input().split(" "))
+    g[x].append(y)
+    g[y].append(x)
+
+"""
+--------------------------------------------------------------------------------
+    Accepting edges of Weighted Undirected Graphs
+--------------------------------------------------------------------------------
+"""
+for _ in xrange(m):
+    x, y, r = map(int, raw_input().split(" "))
+    g[x].append([y, r])
+    g[y].append([x, r])
+
+"""
+--------------------------------------------------------------------------------
+    Depth First Search.
+        Args :  G - Dictionary of edges
+                s - Starting Node
+        Vars :  vis - Set of visited nodes
+                S - Traversal Stack
+--------------------------------------------------------------------------------
+"""
+
+
+def dfs(G, s):
+    vis, S = set([s]), [s]
+    print s
+    while S:
+        flag = 0
+        for i in G[S[-1]]:
+            if i not in vis:
+                S.append(i)
+                vis.add(i)
+                flag = 1
+                print i
+                break
+        if not flag:
+            S.pop()
+
+
+"""
+--------------------------------------------------------------------------------
+    Breadth First Search.
+        Args :  G - Dictionary of edges
+                s - Starting Node
+        Vars :  vis - Set of visited nodes
+                Q - Traveral Stack
+--------------------------------------------------------------------------------
+"""
+from collections import deque
+
+
+def bfs(G, s):
+    vis, Q = set([s]), deque([s])
+    print s
+    while Q:
+        u = Q.popleft()
+        for v in G[u]:
+            if v not in vis:
+                vis.add(v)
+                Q.append(v)
+                print v
+
+
+"""
+--------------------------------------------------------------------------------
+    Dijkstra's shortest path Algorithm
+        Args :  G - Dictionary of edges
+                s - Starting Node
+        Vars :  dist - Dictionary storing shortest distance from s to every other node
+                known - Set of knows nodes
+                path - Preceding node in path
+--------------------------------------------------------------------------------
+"""
+
+
+def dijk(G, s):
+    dist, known, path = {s: 0}, set(), {s: 0}
+    while True:
+        if len(known) == len(G) - 1:
+            break
+        mini = 100000
+        for i in dist:
+            if i not in known and dist[i] < mini:
+                mini = dist[i]
+                u = i
+        known.add(u)
+        for v in G[u]:
+            if v[0] not in known:
+                if dist[u] + v[1] < dist.get(v[0], 100000):
+                    dist[v[0]] = dist[u] + v[1]
+                    path[v[0]] = u
+    for i in dist:
+        if i != s:
+            print dist[i]
+
+
+"""
+--------------------------------------------------------------------------------
+    Topological Sort
+--------------------------------------------------------------------------------
+"""
+from collections import deque
+
+
+def topo(G, ind=None, Q=[1]):
+    if ind == None:
+        ind = [0] * (len(G) + 1) 		# SInce oth Index is ignored
+        for u in G:
+            for v in G[u]:
+                ind[v] += 1
+        Q = deque()
+        for i in G:
+            if ind[i] == 0:
+                Q.append(i)
+    if len(Q) == 0:
+        return
+    v = Q.popleft()
+    print v
+    for w in G[v]:
+        ind[w] -= 1
+        if ind[w] == 0:
+            Q.append(w)
+    topo(G, ind, Q)
+
+
+"""
+--------------------------------------------------------------------------------
+    Reading an Adjacency matrix
+--------------------------------------------------------------------------------
+"""
+
+
+def adjm():
+    n, a = input(), []
+    for i in xrange(n):
+        a.append(map(int, raw_input().split()))
+    return a, n
+
+
+"""
+--------------------------------------------------------------------------------
+    Floyd Warshall's algorithm
+        Args :  G - Dictionary of edges
+                s - Starting Node
+        Vars :  dist - Dictionary storing shortest distance from s to every other node
+                known - Set of knows nodes
+                path - Preceding node in path
+
+--------------------------------------------------------------------------------
+"""
+
+
+def floy((A, n)):
+    dist = list(A)
+    path = [[0] * n for i in xrange(n)]
+    for k in xrange(n):
+        for i in xrange(n):
+            for j in xrange(n):
+                if dist[i][j] > dist[i][k] + dist[k][j]:
+                    dist[i][j] = dist[i][k] + dist[k][j]
+                    path[i][k] = k
+    print dist
+
+
+"""
+--------------------------------------------------------------------------------
+    Prim's MST Algorithm
+        Args :  G - Dictionary of edges
+                s - Starting Node
+        Vars :  dist - Dictionary storing shortest distance from s to nearest node
+                known - Set of knows nodes
+                path - Preceding node in path
+--------------------------------------------------------------------------------
+"""
+
+
+def prim(G, s):
+    dist, known, path = {s: 0}, set(), {s: 0}
+    while True:
+        if len(known) == len(G) - 1:
+            break
+        mini = 100000
+        for i in dist:
+            if i not in known and dist[i] < mini:
+                mini = dist[i]
+                u = i
+        known.add(u)
+        for v in G[u]:
+            if v[0] not in known:
+                if v[1] < dist.get(v[0], 100000):
+                    dist[v[0]] = v[1]
+                    path[v[0]] = u
+
+
+"""
+--------------------------------------------------------------------------------
+    Accepting Edge list
+        Vars :  n - Number of nodes
+                m - Number of edges
+        Returns : l - Edge list
+                n - Number of Nodes
+--------------------------------------------------------------------------------
+"""
+
+
+def edglist():
+    n, m = map(int, raw_input().split(" "))
+    l = []
+    for i in xrange(m):
+        l.append(map(int, raw_input().split(' ')))
+    return l, n
+
+
+"""
+--------------------------------------------------------------------------------
+    Kruskal's MST Algorithm
+        Args :  E - Edge list
+                n - Number of Nodes
+        Vars :  s - Set of all nodes as unique disjoint sets (initially)
+--------------------------------------------------------------------------------
+"""
+
+
+def krusk((E, n)):
+    # Sort edges on the basis of distance
+    E.sort(reverse=True, key=lambda x: x[2])
+    s = [set([i]) for i in range(1, n + 1)]
+    while True:
+        if len(s) == 1:
+            break
+        print s
+        x = E.pop()
+        for i in xrange(len(s)):
+            if x[0] in s[i]:
+                break
+        for j in xrange(len(s)):
+            if x[1] in s[j]:
+                if i == j:
+                    break
+                s[j].update(s[i])
+                s.pop(i)
+                break

Neural_Network/convolution_neural_network.py

@@ -0,0 +1,305 @@
+#-*- coding: utf-8 -*-
+
+'''
+     - - - - - -- - - - - - - - - - - - - - - - - - - - - - -
+    Name - - CNN - Convolution Neural Network For Photo Recognizing
+    Goal - - Recognize Handing Writting Word Photo
+    Detail：Total 5 layers neural network
+            * Convolution layer
+            * Pooling layer
+            * Input layer layer of BP
+            * Hiden layer of BP
+            * Output layer of BP
+    Author: Stephen Lee
+    Github: 245885195@qq.com
+    Date: 2017.9.20
+    - - - - - -- - - - - - - - - - - - - - - - - - - - - - -
+          '''
+
+import numpy as np
+import matplotlib.pyplot as plt
+
+class CNN():
+
+    def __init__(self,conv1_get,size_p1,bp_num1,bp_num2,bp_num3,rate_w=0.2,rate_t=0.2):
+        '''
+        :param conv1_get: [a,c,d]，size, number, step of convolution kernel
+        :param size_p1: pooling size
+        :param bp_num1: units number of flatten layer
+        :param bp_num2: units number of hidden layer
+        :param bp_num3: units number of output layer
+        :param rate_w: rate of weight learning
+        :param rate_t: rate of threshold learning
+        '''
+        self.num_bp1 = bp_num1
+        self.num_bp2 = bp_num2
+        self.num_bp3 = bp_num3
+        self.conv1 = conv1_get[:2]
+        self.step_conv1 = conv1_get[2]
+        self.size_pooling1 = size_p1
+        self.rate_weight = rate_w
+        self.rate_thre = rate_t
+        self.w_conv1 = [np.mat(-1*np.random.rand(self.conv1[0],self.conv1[0])+0.5) for i in range(self.conv1[1])]
+        self.wkj = np.mat(-1 * np.random.rand(self.num_bp3, self.num_bp2) + 0.5)
+        self.vji = np.mat(-1*np.random.rand(self.num_bp2, self.num_bp1)+0.5)
+        self.thre_conv1 = -2*np.random.rand(self.conv1[1])+1
+        self.thre_bp2 = -2*np.random.rand(self.num_bp2)+1
+        self.thre_bp3 = -2*np.random.rand(self.num_bp3)+1
+
+
+    def save_model(self,save_path):
+        #save model dict with pickle
+        import pickle
+        model_dic = {'num_bp1':self.num_bp1,
+                    'num_bp2':self.num_bp2,
+                    'num_bp3':self.num_bp3,
+                    'conv1':self.conv1,
+                    'step_conv1':self.step_conv1,
+                    'size_pooling1':self.size_pooling1,
+                    'rate_weight':self.rate_weight,
+                    'rate_thre':self.rate_thre,
+                    'w_conv1':self.w_conv1,
+                    'wkj':self.wkj,
+                    'vji':self.vji,
+                    'thre_conv1':self.thre_conv1,
+                    'thre_bp2':self.thre_bp2,
+                     'thre_bp3':self.thre_bp3}
+        with open(save_path, 'wb') as f:
+            pickle.dump(model_dic, f)
+
+        print('Model saved： %s'% save_path)
+
+    @classmethod
+    def ReadModel(cls,model_path):
+        #read saved model
+        import pickle
+        with open(model_path, 'rb') as f:
+            model_dic = pickle.load(f)
+
+        conv_get= model_dic.get('conv1')
+        conv_get.append(model_dic.get('step_conv1'))
+        size_p1 = model_dic.get('size_pooling1')
+        bp1 = model_dic.get('num_bp1')
+        bp2 = model_dic.get('num_bp2')
+        bp3 = model_dic.get('num_bp3')
+        r_w = model_dic.get('rate_weight')
+        r_t = model_dic.get('rate_thre')
+        #create model instance
+        conv_ins = CNN(conv_get,size_p1,bp1,bp2,bp3,r_w,r_t)
+        #modify model parameter
+        conv_ins.w_conv1 = model_dic.get('w_conv1')
+        conv_ins.wkj = model_dic.get('wkj')
+        conv_ins.vji = model_dic.get('vji')
+        conv_ins.thre_conv1 = model_dic.get('thre_conv1')
+        conv_ins.thre_bp2 = model_dic.get('thre_bp2')
+        conv_ins.thre_bp3 = model_dic.get('thre_bp3')
+        return conv_ins
+
+
+    def sig(self,x):
+        return 1 / (1 + np.exp(-1*x))
+
+    def do_round(self,x):
+        return round(x, 3)
+
+    def convolute(self,data,convs,w_convs,thre_convs,conv_step):
+        #convolution process
+        size_conv = convs[0]
+        num_conv =convs[1]
+        size_data = np.shape(data)[0]
+        #get the data slice of original image data, data_focus
+        data_focus = []
+        for i_focus in range(0, size_data - size_conv + 1, conv_step):
+            for j_focus in range(0, size_data - size_conv + 1, conv_step):
+                focus = data[i_focus:i_focus + size_conv, j_focus:j_focus + size_conv]
+                data_focus.append(focus)
+        #caculate the feature map of every single kernel, and saved as list of matrix
+        data_featuremap = []
+        Size_FeatureMap = int((size_data - size_conv) / conv_step + 1)
+        for i_map in range(num_conv):
+            featuremap = []
+            for i_focus in range(len(data_focus)):
+                net_focus = np.sum(np.multiply(data_focus[i_focus], w_convs[i_map])) - thre_convs[i_map]
+                featuremap.append(self.sig(net_focus))
+            featuremap = np.asmatrix(featuremap).reshape(Size_FeatureMap, Size_FeatureMap)
+            data_featuremap.append(featuremap)
+
+        #expanding the data slice to One dimenssion
+        focus1_list = []
+        for each_focus in data_focus:
+            focus1_list.extend(self.Expand_Mat(each_focus))
+        focus_list = np.asarray(focus1_list)
+        return focus_list,data_featuremap
+
+    def pooling(self,featuremaps,size_pooling,type='average_pool'):
+        #pooling process
+        size_map = len(featuremaps[0])
+        size_pooled = int(size_map/size_pooling)
+        featuremap_pooled = []
+        for i_map in range(len(featuremaps)):
+            map = featuremaps[i_map]
+            map_pooled = []
+            for i_focus in range(0,size_map,size_pooling):
+                for j_focus in range(0, size_map, size_pooling):
+                    focus = map[i_focus:i_focus + size_pooling, j_focus:j_focus + size_pooling]
+                    if type == 'average_pool':
+                        #average pooling
+                        map_pooled.append(np.average(focus))
+                    elif type == 'max_pooling':
+                        #max pooling
+                        map_pooled.append(np.max(focus))
+            map_pooled = np.asmatrix(map_pooled).reshape(size_pooled,size_pooled)
+            featuremap_pooled.append(map_pooled)
+        return featuremap_pooled
+
+    def _expand(self,datas):
+        #expanding three dimension data to one dimension list
+        data_expanded = []
+        for i in range(len(datas)):
+            shapes = np.shape(datas[i])
+            data_listed = datas[i].reshape(1,shapes[0]*shapes[1])
+            data_listed = data_listed.getA().tolist()[0]
+            data_expanded.extend(data_listed)
+        data_expanded = np.asarray(data_expanded)
+        return data_expanded
+
+    def _expand_mat(self,data_mat):
+        #expanding matrix to one dimension list
+        data_mat = np.asarray(data_mat)
+        shapes = np.shape(data_mat)
+        data_expanded = data_mat.reshape(1,shapes[0]*shapes[1])
+        return data_expanded
+
+    def _calculate_gradient_from_pool(self,out_map,pd_pool,num_map,size_map,size_pooling):
+        '''
+        calcluate the gradient from the data slice of pool layer
+        pd_pool: list of matrix
+        out_map: the shape of data slice(size_map*size_map)
+        return: pd_all: list of matrix, [num, size_map, size_map]
+        '''
+        pd_all = []
+        i_pool = 0
+        for i_map in range(num_map):
+            pd_conv1 = np.ones((size_map, size_map))
+            for i in range(0, size_map, size_pooling):
+                for j in range(0, size_map, size_pooling):
+                    pd_conv1[i:i + size_pooling, j:j + size_pooling] = pd_pool[i_pool]
+                    i_pool = i_pool + 1
+            pd_conv2 = np.multiply(pd_conv1,np.multiply(out_map[i_map],(1-out_map[i_map])))
+            pd_all.append(pd_conv2)
+        return pd_all
+
+    def trian(self,patterns,datas_train, datas_teach, n_repeat, error_accuracy,draw_e = bool):
+        #model traning
+        print('----------------------Start Training-------------------------')
+        print(' - - Shape: Train_Data  ',np.shape(datas_train))
+        print(' - - Shape: Teach_Data  ',np.shape(datas_teach))
+        rp = 0
+        all_mse = []
+        mse  = 10000
+        while rp < n_repeat and mse >= error_accuracy:
+            alle = 0
+            print('-------------Learning Time %d--------------'%rp)
+            for p in range(len(datas_train)):
+                #print('------------Learning Image: %d--------------'%p)
+                data_train = np.asmatrix(datas_train[p])
+                data_teach = np.asarray(datas_teach[p])
+                data_focus1,data_conved1 = self.convolute(data_train,self.conv1,self.w_conv1,
+                                           self.thre_conv1,conv_step=self.step_conv1)
+                data_pooled1 = self.pooling(data_conved1,self.size_pooling1)
+                shape_featuremap1 = np.shape(data_conved1)
+                '''
+                print('  -----original shape   ', np.shape(data_train))
+                print('  ---- after convolution  ',np.shape(data_conv1))
+                print('  -----after pooling  ',np.shape(data_pooled1))
+               '''
+                data_bp_input = self._expand(data_pooled1)
+                bp_out1 = data_bp_input
+
+                bp_net_j = np.dot(bp_out1,self.vji.T) - self.thre_bp2
+                bp_out2 = self.sig(bp_net_j)
+                bp_net_k = np.dot(bp_out2 ,self.wkj.T) - self.thre_bp3
+                bp_out3 = self.sig(bp_net_k)
+
+                #--------------Model Leaning ------------------------
+                # calcluate error and gradient---------------
+                pd_k_all = np.multiply((data_teach - bp_out3), np.multiply(bp_out3, (1 - bp_out3)))
+                pd_j_all = np.multiply(np.dot(pd_k_all,self.wkj), np.multiply(bp_out2, (1 - bp_out2)))
+                pd_i_all = np.dot(pd_j_all,self.vji)
+
+                pd_conv1_pooled = pd_i_all / (self.size_pooling1*self.size_pooling1)
+                pd_conv1_pooled = pd_conv1_pooled.T.getA().tolist()
+                pd_conv1_all = self._calculate_gradient_from_pool(data_conved1,pd_conv1_pooled,shape_featuremap1[0],
+                                                    shape_featuremap1[1],self.size_pooling1)
+                #weight and threshold learning process---------
+                #convolution layer
+                for k_conv in range(self.conv1[1]):
+                    pd_conv_list = self._expand_mat(pd_conv1_all[k_conv])
+                    delta_w = self.rate_weight * np.dot(pd_conv_list,data_focus1)
+
+                    self.w_conv1[k_conv] = self.w_conv1[k_conv] + delta_w.reshape((self.conv1[0],self.conv1[0]))
+
+                    self.thre_conv1[k_conv] = self.thre_conv1[k_conv] - np.sum(pd_conv1_all[k_conv]) * self.rate_thre
+                #all connected layer
+                self.wkj = self.wkj + pd_k_all.T * bp_out2 * self.rate_weight
+                self.vji = self.vji + pd_j_all.T * bp_out1 * self.rate_weight
+                self.thre_bp3 = self.thre_bp3 - pd_k_all * self.rate_thre
+                self.thre_bp2 = self.thre_bp2 - pd_j_all * self.rate_thre
+                # calculate the sum error of all single image
+                errors = np.sum(abs((data_teach - bp_out3)))
+                alle = alle + errors
+                #print('   ----Teach      ',data_teach)
+                #print('   ----BP_output  ',bp_out3)
+            rp = rp + 1
+            mse = alle/patterns
+            all_mse.append(mse)
+        def draw_error():
+            yplot = [error_accuracy for i in range(int(n_repeat * 1.2))]
+            plt.plot(all_mse, '+-')
+            plt.plot(yplot, 'r--')
+            plt.xlabel('Learning Times')
+            plt.ylabel('All_mse')
+            plt.grid(True, alpha=0.5)
+            plt.show()
+        print('------------------Training Complished---------------------')
+        print(' - - Training epoch: ', rp, '     - - Mse: %.6f' % mse)
+        if draw_e:
+            draw_error()
+        return mse
+
+    def predict(self,datas_test):
+        #model predict
+        produce_out = []
+        print('-------------------Start Testing-------------------------')
+        print(' - - Shape: Test_Data  ',np.shape(datas_test))
+        for p in range(len(datas_test)):
+            data_test = np.asmatrix(datas_test[p])
+            data_focus1, data_conved1 = self.convolute(data_test, self.conv1, self.w_conv1,
+                                                     self.thre_conv1, conv_step=self.step_conv1)
+            data_pooled1 = self.pooling(data_conved1, self.size_pooling1)
+            data_bp_input = self._expand(data_pooled1)
+
+            bp_out1 = data_bp_input
+            bp_net_j = bp_out1 * self.vji.T - self.thre_bp2
+            bp_out2 = self.sig(bp_net_j)
+            bp_net_k = bp_out2 * self.wkj.T - self.thre_bp3
+            bp_out3 = self.sig(bp_net_k)
+            produce_out.extend(bp_out3.getA().tolist())
+        res = [list(map(self.do_round,each)) for each in produce_out]
+        return np.asarray(res)
+
+    def convolution(self,data):
+        #return the data of image after convoluting process so we can check it out
+        data_test = np.asmatrix(data)
+        data_focus1, data_conved1 = self.convolute(data_test, self.conv1, self.w_conv1,
+                                                     self.thre_conv1, conv_step=self.step_conv1)
+        data_pooled1 = self.pooling(data_conved1, self.size_pooling1)
+
+        return data_conved1,data_pooled1
+
+
+if __name__ == '__main__':
+    pass
+    '''
+    I will put the example on other file
+    '''

Neural_Network/neuralnetwork_bp3.py

@@ -0,0 +1,152 @@
+#-*- coding:utf-8 -*-
+'''
+Author: Stephen Lee
+Date: 2017.9.21
+
+BP neural network with three layers
+'''
+
+import numpy as np
+import matplotlib.pyplot as plt
+
+class Bpnn():
+
+    def __init__(self,n_layer1,n_layer2,n_layer3,rate_w=0.3,rate_t=0.3):
+        '''
+        :param n_layer1: number of input layer
+        :param n_layer2: number of hiden layer
+        :param n_layer3: number of output layer
+        :param rate_w: rate of weight learning
+        :param rate_t: rate of threshold learning
+        '''
+        self.num1 = n_layer1
+        self.num2 = n_layer2
+        self.num3 = n_layer3
+        self.rate_weight = rate_w
+        self.rate_thre = rate_t
+        self.thre2 = -2*np.random.rand(self.num2)+1
+        self.thre3 = -2*np.random.rand(self.num3)+1
+        self.vji = np.mat(-2*np.random.rand(self.num2, self.num1)+1)
+        self.wkj = np.mat(-2*np.random.rand(self.num3, self.num2)+1)
+
+    def sig(self,x):
+        return 1 / (1 + np.exp(-1*x))
+
+    def sig_plain(self,x):
+        return 1 / (1 + np.exp(-1*x))
+
+    def do_round(self,x):
+        return round(x, 3)
+
+    def trian(self,patterns,data_train, data_teach, n_repeat, error_accuracy, draw_e=False):
+        '''
+        :param patterns: the number of patterns
+        :param data_train: training data x; numpy.ndarray
+        :param data_teach: training data y; numpy.ndarray
+        :param n_repeat: echoes
+        :param error_accuracy: error accuracy
+        :return: None
+        '''
+        data_train = np.asarray(data_train)
+        data_teach = np.asarray(data_teach)
+        # print('-------------------Start Training-------------------------')
+        # print(' - - Shape: Train_Data  ',np.shape(data_train))
+        # print(' - - Shape: Teach_Data  ',np.shape(data_teach))
+        rp = 0
+        all_mse = []
+        mse  = 10000
+        while rp < n_repeat and mse >= error_accuracy:
+            alle = 0
+            final_out = []
+            for g in range(np.shape(data_train)[0]):
+                net_i = data_train[g]
+                out1 = net_i
+
+                net_j = out1 * self.vji.T  - self.thre2
+                out2=self.sig(net_j)
+
+                net_k = out2 * self.wkj.T  - self.thre3
+                out3 = self.sig(net_k)
+
+                # learning process
+                pd_k_all = np.multiply(np.multiply(out3,(1 - out3)),(data_teach[g]-out3))
+                pd_j_all = np.multiply(pd_k_all * self.wkj,np.multiply(out2,1-out2))
+                #upgrade weight
+                self.wkj = self.wkj + pd_k_all.T * out2 *self.rate_weight
+                self.vji = self.vji + pd_j_all.T * out1 * self.rate_weight
+                #upgrade threshold
+                self.thre3 = self.thre3 - pd_k_all * self.rate_thre
+                self.thre2 = self.thre2 - pd_j_all * self.rate_thre
+                #calculate sum of error
+                errors = np.sum(abs((data_teach[g] - out3)))
+
+                alle = alle + errors
+                final_out.extend(out3.getA().tolist())
+            final_out3 = [list(map(self.do_round,each)) for each in final_out]
+
+            rp = rp + 1
+            mse = alle/patterns
+            all_mse.append(mse)
+        def draw_error():
+            yplot = [error_accuracy for i in range(int(n_repeat * 1.2))]
+            plt.plot(all_mse, '+-')
+            plt.plot(yplot, 'r--')
+            plt.xlabel('Learning Times')
+            plt.ylabel('All_mse')
+            plt.grid(True,alpha = 0.7)
+            plt.show()
+        # print('------------------Training Complished---------------------')
+        # print(' - - Training epoch: ', rp, '     - - Mse: %.6f'%mse)
+        # print(' - - Last Output: ', final_out3)
+        if draw_e:
+            draw_error()
+
+    def predict(self,data_test):
+        '''
+        :param data_test: data test, numpy.ndarray
+        :return: predict output data
+        '''
+        data_test = np.asarray(data_test)
+        produce_out = []
+        # print('-------------------Start Testing-------------------------')
+        # print(' - - Shape: Test_Data  ',np.shape(data_test))
+        # print(np.shape(data_test))
+        for g in range(np.shape(data_test)[0]):
+
+            net_i = data_test[g]
+            out1 = net_i
+
+            net_j = out1 * self.vji.T - self.thre2
+            out2 = self.sig(net_j)
+
+            net_k = out2 * self.wkj.T - self.thre3
+            out3 = self.sig(net_k)
+            produce_out.extend(out3.getA().tolist())
+        res = [list(map(self.do_round,each)) for each in produce_out]
+        return np.asarray(res)
+
+
+def main():
+    #example data
+    data_x = [[1,2,3,4],
+              [5,6,7,8],
+              [2,2,3,4],
+              [7,7,8,8]]
+    data_y = [[1,0,0,0],
+              [0,1,0,0],
+              [0,0,1,0],
+              [0,0,0,1]]
+
+    test_x = [[1,2,3,4],
+              [3,2,3,4]]
+
+    #building network model
+    model = Bpnn(4,10,4)
+    #training the model
+    model.trian(patterns=4,data_train=data_x,data_teach=data_y,
+                n_repeat=100,error_accuracy=0.1,draw_e=True)
+    #predicting data
+    model.predict(test_x)
+
+if __name__ == '__main__':
+    main()

README.md

@@ -1,4 +1,4 @@
-# The Algorithms - Python [![Build Status](https://travis-ci.org/TheAlgorithms/Python.svg)](https://travis-ci.org/TheAlgorithms/Python)
+# The Algorithms - Python <!-- [![Build Status](https://travis-ci.org/TheAlgorithms/Python.svg)](https://travis-ci.org/TheAlgorithms/Python) -->
 
 ### All algorithms implemented in Python (for education)
 
@@ -128,6 +128,13 @@ The method is named after **Julius Caesar**, who used it in his private correspo
 The encryption step performed by a Caesar cipher is often incorporated as part of more complex schemes, such as the Vigenère cipher, and still has modern application in the ROT13 system. As with all single-alphabet substitution ciphers, the Caesar cipher is easily broken and in modern practice offers essentially no communication security.
 ###### Source: [Wikipedia](https://en.wikipedia.org/wiki/Caesar_cipher)
 
+### Vigenère
+The **Vigenère cipher** is a method of encrypting alphabetic text by using a series of **interwoven Caesar ciphers** based on the letters of a keyword. It is **a form of polyalphabetic substitution**.<br>
+The Vigenère cipher has been reinvented many times. The method was originally described by Giovan Battista Bellaso in his 1553 book La cifra del. Sig. Giovan Battista Bellaso; however, the scheme was later misattributed to Blaise de Vigenère in the 19th century, and is now widely known as the "Vigenère cipher".<br>
+Though the cipher is easy to understand and implement, for three centuries it resisted all attempts to break it; this earned it the description **le chiffre indéchiffrable**(French for 'the indecipherable cipher'). 
+Many people have tried to implement encryption schemes that are essentially Vigenère ciphers. Friedrich Kasiski was the first to publish a general method of deciphering a Vigenère cipher in 1863.
+###### Source: [Wikipedia](https://en.wikipedia.org/wiki/Vigen%C3%A8re_cipher)
+
 ### Transposition
 In cryptography, a **transposition cipher** is a method of encryption by which the positions held by units of plaintext (which are commonly characters or groups of characters) are shifted according to a regular system, so that the ciphertext constitutes a permutation of the plaintext. That is, the order of the units is changed (the plaintext is reordered).<br> 
 Mathematically a bijective function is used on the characters' positions to encrypt and an inverse function to decrypt.

data_structures/AVL/AVL.py

@@ -7,40 +7,42 @@ class Node:
 
     def __init__(self, label):
         self.label = label
-        self.left = None
-        self.rigt = None
-        self.parent = None
+        self._parent = None
+        self._left = None
+        self._right = None
         self.height = 0
 
-    def getLabel(self):
-        return self.label
+    @property
+    def right(self):
+        return self._right
 
-    def setLabel(self, label):
-        self.label = label
+    @right.setter
+    def right(self, node):
+        if node is not None:
+            node._parent = self
+            self._right = node
 
-    def getLeft(self):
-        return self.left
+    @property
+    def left(self):
+        return self._left
 
-    def setLeft(self, left):
-        self.left = left
+    @left.setter
+    def left(self, node):
+        if node is not None:
+            node._parent = self
+            self._left = node
 
-    def getRight(self):
-        return self.rigt
+    @property
+    def parent(self):
+        return self._parent
 
-    def setRight(self, right):
-        self.rigt = right
-
-    def getParent(self):
-        return self.parent
-
-    def setParent(self, parent):
-        self.parent = parent
-
-    def setHeight(self, height):
-        self.height = height
-
-    def getHeight(self, height):
-        return self.height
+    @parent.setter
+    def parent(self, node):
+        if node is not None:
+            self._parent = node
+            self.height = self.parent.height + 1
+        else:
+            self.height = 0
 
 
 class AVL:
@@ -51,8 +53,10 @@ class AVL:
 
     def insert(self, value):
         node = Node(value)
+
         if self.root is None:
             self.root = node
+            self.root.height = 0
             self.size = 1
         else:
             # Same as Binary Tree
@@ -64,63 +68,77 @@ class AVL:
 
                     dad_node = curr_node
 
-                    if node.getLabel() < curr_node.getLabel():
-                        curr_node = curr_node.getLeft()
+                    if node.label < curr_node.label:
+                        curr_node = curr_node.left
                     else:
-                        curr_node = curr_node.getRight()
+                        curr_node = curr_node.right
                 else:
-                    if node.getLabel() < dad_node.getLabel():
-                        dad_node.setLeft(node)
-                        dad_node.setHeight(dad_node.getHeight() + 1)
-
-                        if (dad_node.getRight().getHeight() -
-                                dad_node.getLeft.getHeight() > 1):
-                            self.rebalance(dad_node)
-
+                    node.height = dad_node.height
+                    dad_node.height += 1
+                    if node.label < dad_node.label:
+                        dad_node.left = node
                     else:
-                        dad_node.setRight(node)
-                        dad_node.setHeight(dad_node.getHeight() + 1)
-
-                        if (dad_node.getRight().getHeight() -
-                                dad_node.getLeft.getHeight() > 1):
-                            self.rebalance(dad_node)
+                        dad_node.right = node
+                    self.rebalance(node)
+                    self.size += 1
                     break
 
     def rebalance(self, node):
-        if (node.getRight().getHeight() -
-                node.getLeft.getHeight() > 1):
-            if (node.getRight().getHeight() >
-                    node.getLeft.getHeight()):
-                pass
-            else:
-                pass
-            pass
-        elif (node.getRight().getHeight() -
-                node.getLeft.getHeight() > 2):
-            if (node.getRight().getHeight() >
-                    node.getLeft.getHeight()):
-                pass
-            else:
-                pass
-            pass
-        pass
+        n = node
+
+        while n is not None:
+            height_right = n.height
+            height_left = n.height
+
+            if n.right is not None:
+                height_right = n.right.height
+
+            if n.left is not None:
+                height_left = n.left.height
+
+            if abs(height_left - height_right) > 1:
+                if height_left > height_right:
+                    left_child = n.left
+                    if left_child is not None:
+                        h_right = (right_child.right.height
+                                    if (right_child.right is not None) else 0)
+                        h_left = (right_child.left.height
+                                    if (right_child.left is not None) else 0)
+                    if (h_left > h_right):
+                        self.rotate_left(n)
+                        break
+                    else:
+                        self.double_rotate_right(n)
+                        break
+                else:
+                    right_child = n.right
+                    if right_child is not None:
+                        h_right = (right_child.right.height
+                            if (right_child.right is not None) else 0)
+                        h_left = (right_child.left.height
+                            if (right_child.left is not None) else 0)
+                    if (h_left > h_right):
+                        self.double_rotate_left(n)
+                        break
+                    else:
+                        self.rotate_right(n)
+                        break
+            n = n.parent
 
     def rotate_left(self, node):
-        # TODO: is this pythonic enought?
-        aux = node.getLabel()
-        node = aux.getRight()
-        node.setHeight(node.getHeight() - 1)
-        node.setLeft(Node(aux))
-        node.getLeft().setHeight(node.getHeight() + 1)
-        node.getRight().setHeight(node.getRight().getHeight() - 1)
+        aux = node.parent.label
+        node.parent.label = node.label
+        node.parent.right = Node(aux)
+        node.parent.right.height = node.parent.height + 1
+        node.parent.left = node.right
+
 
     def rotate_right(self, node):
-        aux = node.getLabel()
-        node = aux.getLeft()
-        node.setHeight(node.getHeight() - 1)
-        node.setRight(Node(aux))
-        node.getLeft().setHeight(node.getHeight() + 1)
-        node.getLeft().setHeight(node.getLeft().getHeight() - 1)
+        aux = node.parent.label
+        node.parent.label = node.label
+        node.parent.left = Node(aux)
+        node.parent.left.height = node.parent.height + 1
+        node.parent.right = node.right
 
     def double_rotate_left(self, node):
         self.rotate_right(node.getRight().getRight())
@@ -129,3 +147,34 @@ class AVL:
     def double_rotate_right(self, node):
         self.rotate_left(node.getLeft().getLeft())
         self.rotate_right(node)
+
+    def empty(self):
+        if self.root is None:
+            return True
+        return False
+
+    def preShow(self, curr_node):
+        if curr_node is not None:
+            self.preShow(curr_node.left)
+            print(curr_node.label, end=" ")
+            self.preShow(curr_node.right)
+
+    def preorder(self, curr_node):
+        if curr_node is not None:
+            self.preShow(curr_node.left)
+            self.preShow(curr_node.right)
+            print(curr_node.label, end=" ")
+
+    def getRoot(self):
+        return self.root
+
+t = AVL()
+t.insert(1)
+t.insert(2)
+t.insert(3)
+# t.preShow(t.root)
+# print("\n")
+# t.insert(4)
+# t.insert(5)
+# t.preShow(t.root)
+# t.preorden(t.root)

data_structures/Arrays

@@ -0,0 +1 @@
+Arrays implimentation using python programming.

data_structures/Binary Tree/FenwickTree.py

@@ -0,0 +1,28 @@
+class FenwickTree:
+
+    def __init__(self, SIZE): # create fenwick tree with size SIZE
+        self.Size = SIZE
+        self.ft = [0 for i in range (0,SIZE)]
+
+    def update(self, i, val): # update data (adding) in index i in O(lg N)
+        while (i < self.Size):
+            self.ft[i] += val
+            i += i & (-i)
+
+    def query(self, i): # query cumulative data from index 0 to i in O(lg N)
+        ret = 0
+        while (i > 0):
+            ret += self.ft[i]
+            i -= i & (-i)
+        return ret
+            
+if __name__ == '__main__':
+    f = FenwickTree(100)
+    f.update(1,20)
+    f.update(4,4)
+    print (f.query(1))
+    print (f.query(3))
+    print (f.query(4))
+    f.update(2,-5)
+    print (f.query(1))
+    print (f.query(3))

data_structures/Binary Tree/LazySegmentTree.py

@@ -0,0 +1,90 @@
+import math
+
+class SegmentTree:
+    
+    def __init__(self, N):
+        self.N = N
+        self.st = [0 for i in range(0,4*N)] # approximate the overall size of segment tree with array N
+        self.lazy = [0 for i in range(0,4*N)] # create array to store lazy update
+        self.flag = [0 for i in range(0,4*N)] # flag for lazy update
+        
+    def left(self, idx):
+        return idx*2
+
+    def right(self, idx):
+        return idx*2 + 1
+
+    def build(self, idx, l, r, A):
+        if l==r:
+            self.st[idx] = A[l-1]
+        else :
+            mid = (l+r)//2
+            self.build(self.left(idx),l,mid, A)
+            self.build(self.right(idx),mid+1,r, A)
+            self.st[idx] = max(self.st[self.left(idx)] , self.st[self.right(idx)])
+
+    # update with O(lg N) (Normal segment tree without lazy update will take O(Nlg N) for each update)
+    def update(self, idx, l, r, a, b, val): # update(1, 1, N, a, b, v) for update val v to [a,b]
+        if self.flag[idx] == True:
+            self.st[idx] = self.lazy[idx]
+            self.flag[idx] = False
+            if l!=r:
+                self.lazy[self.left(idx)] = self.lazy[idx]
+                self.lazy[self.right(idx)] = self.lazy[idx]
+                self.flag[self.left(idx)] = True
+                self.flag[self.right(idx)] = True
+            
+        if r < a or l > b:
+            return True
+        if l >= a and r <= b :
+            self.st[idx] = val
+            if l!=r:
+                self.lazy[self.left(idx)] = val
+                self.lazy[self.right(idx)] = val
+                self.flag[self.left(idx)] = True
+                self.flag[self.right(idx)] = True
+            return True
+        mid = (l+r)//2
+        self.update(self.left(idx),l,mid,a,b,val)
+        self.update(self.right(idx),mid+1,r,a,b,val)
+        self.st[idx] = max(self.st[self.left(idx)] , self.st[self.right(idx)])
+        return True
+
+    # query with O(lg N)
+    def query(self, idx, l, r, a, b): #query(1, 1, N, a, b) for query max of [a,b]
+        if self.flag[idx] == True:
+            self.st[idx] = self.lazy[idx]
+            self.flag[idx] = False
+            if l != r:
+                self.lazy[self.left(idx)] = self.lazy[idx]
+                self.lazy[self.right(idx)] = self.lazy[idx]
+                self.flag[self.left(idx)] = True
+                self.flag[self.right(idx)] = True
+        if r < a or l > b:
+            return -math.inf
+        if l >= a and r <= b:
+            return self.st[idx]
+        mid = (l+r)//2
+        q1 = self.query(self.left(idx),l,mid,a,b)
+        q2 = self.query(self.right(idx),mid+1,r,a,b)
+        return max(q1,q2)
+
+    def showData(self):
+        showList = []
+        for i in range(1,N+1):
+            showList += [self.query(1, 1, self.N, i, i)]
+        print (showList)
+            
+
+if __name__ == '__main__':
+    A = [1,2,-4,7,3,-5,6,11,-20,9,14,15,5,2,-8]
+    N = 15
+    segt = SegmentTree(N)
+    segt.build(1,1,N,A)
+    print (segt.query(1,1,N,4,6))
+    print (segt.query(1,1,N,7,11))
+    print (segt.query(1,1,N,7,12))
+    segt.update(1,1,N,1,3,111)
+    print (segt.query(1,1,N,1,15))
+    segt.update(1,1,N,7,8,235)
+    segt.showData()

data_structures/Binary Tree/SegmentTree.py

@@ -0,0 +1,64 @@
+import math
+
+class SegmentTree:
+    
+    def __init__(self, N):
+        self.N = N
+        self.st = [0 for i in range(0,4*N)] # approximate the overall size of segment tree with array N
+        
+    def left(self, idx):
+        return idx*2
+
+    def right(self, idx):
+        return idx*2 + 1
+
+    def build(self, idx, l, r, A):
+        if l==r:
+            self.st[idx] = A[l-1]
+        else :
+            mid = (l+r)//2
+            self.build(self.left(idx),l,mid, A)
+            self.build(self.right(idx),mid+1,r, A)
+            self.st[idx] = max(self.st[self.left(idx)] , self.st[self.right(idx)])
+    
+    def update(self, idx, l, r, a, b, val): # update(1, 1, N, a, b, v) for update val v to [a,b]
+        if r < a or l > b:
+            return True
+        if l == r :
+            self.st[idx] = val
+            return True
+        mid = (l+r)//2
+        self.update(self.left(idx),l,mid,a,b,val)
+        self.update(self.right(idx),mid+1,r,a,b,val)
+        self.st[idx] = max(self.st[self.left(idx)] , self.st[self.right(idx)])
+        return True
+
+    def query(self, idx, l, r, a, b): #query(1, 1, N, a, b) for query max of [a,b]
+        if r < a or l > b:
+            return -math.inf
+        if l >= a and r <= b:
+            return self.st[idx]
+        mid = (l+r)//2
+        q1 = self.query(self.left(idx),l,mid,a,b)
+        q2 = self.query(self.right(idx),mid+1,r,a,b)
+        return max(q1,q2)
+
+    def showData(self):
+        showList = []
+        for i in range(1,N+1):
+            showList += [self.query(1, 1, self.N, i, i)]
+        print (showList)
+            
+
+if __name__ == '__main__':
+    A = [1,2,-4,7,3,-5,6,11,-20,9,14,15,5,2,-8]
+    N = 15
+    segt = SegmentTree(N)
+    segt.build(1,1,N,A)
+    print (segt.query(1,1,N,4,6))
+    print (segt.query(1,1,N,7,11))
+    print (segt.query(1,1,N,7,12))
+    segt.update(1,1,N,1,3,111)
+    print (segt.query(1,1,N,1,15))
+    segt.update(1,1,N,7,8,235)
+    segt.showData()

data_structures/Binary Tree/binary_seach_tree.py

@@ -8,7 +8,7 @@ class Node:
     def __init__(self, label):
         self.label = label
         self.left = None
-        self.rigt = None
+        self.right = None
 
     def getLabel(self):
         return self.label
@@ -23,10 +23,10 @@ class Node:
         self.left = left
 
     def getRight(self):
-        return self.rigt
+        return self.right
 
     def setRight(self, right):
-        self.rigt = right
+        self.right = right
 
 
 class BinarySearchTree:
@@ -68,7 +68,7 @@ class BinarySearchTree:
         return False
 
     def preShow(self, curr_node):
-        if curr_node is None:
+        if curr_node is not None:
             print(curr_node.getLabel(), end=" ")
 
             self.preShow(curr_node.getLeft())

data_structures/Graph/dijkstra_algorithm.py

@@ -0,0 +1,211 @@
+# Title: Dijkstra's Algorithm for finding single source shortest path from scratch
+# Author: Shubham Malik
+# References: https://en.wikipedia.org/wiki/Dijkstra%27s_algorithm
+
+import math
+import sys
+# For storing the vertex set to retreive node with the lowest distance
+
+
+class PriorityQueue:
+    # Based on Min Heap
+    def __init__(self):
+        self.cur_size = 0
+        self.array = []
+        self.pos = {}   # To store the pos of node in array
+
+    def isEmpty(self):
+        return self.cur_size == 0
+
+    def min_heapify(self, idx):
+        lc = self.left(idx)
+        rc = self.right(idx)
+        if lc < self.cur_size and self.array(lc)[0] < self.array(idx)[0]:
+            smallest = lc
+        else:
+            smallest = idx
+        if rc < self.cur_size and self.array(rc)[0] < self.array(smallest)[0]:
+            smallest = rc
+        if smallest != idx:
+            self.swap(idx, smallest)
+            self.min_heapify(smallest)
+
+    def insert(self, tup):
+        # Inserts a node into the Priority Queue
+        self.pos[tup[1]] = self.cur_size
+        self.cur_size += 1
+        self.array.append((sys.maxsize, tup[1]))
+        self.decrease_key((sys.maxsize, tup[1]), tup[0])
+
+    def extract_min(self):
+        # Removes and returns the min element at top of priority queue
+        min_node = self.array[0][1]
+        self.array[0] = self.array[self.cur_size - 1]
+        self.cur_size -= 1
+        self.min_heapify(1)
+        del self.pos[min_node]
+        return min_node
+
+    def left(self, i):
+        # returns the index of left child
+        return 2 * i + 1
+
+    def right(self, i):
+        # returns the index of right child
+        return 2 * i + 2
+
+    def par(self, i):
+        # returns the index of parent
+        return math.floor(i / 2)
+
+    def swap(self, i, j):
+        # swaps array elements at indices i and j
+        # update the pos{}
+        self.pos[self.array[i][1]] = j
+        self.pos[self.array[j][1]] = i
+        temp = self.array[i]
+        self.array[i] = self.array[j]
+        self.array[j] = temp
+
+    def decrease_key(self, tup, new_d):
+        idx = self.pos[tup[1]]
+        # assuming the new_d is atmost old_d
+        self.array[idx] = (new_d, tup[1])
+        while idx > 0 and self.array[self.par(idx)][0] > self.array[idx][0]:
+            self.swap(idx, self.par(idx))
+            idx = self.par(idx)
+
+
+class Graph:
+    def __init__(self, num):
+        self.adjList = {}   # To store graph: u -> (v,w)
+        self.num_nodes = num    # Number of nodes in graph
+        # To store the distance from source vertex
+        self.dist = [0] * self.num_nodes
+        self.par = [-1] * self.num_nodes  # To store the path
+
+    def add_edge(self, u, v, w):
+        #  Edge going from node u to v and v to u with weight w
+        # u (w)-> v, v (w) -> u
+        # Check if u already in graph
+        if u in self.adjList.keys():
+            self.adjList[u].append((v, w))
+        else:
+            self.adjList[u] = [(v, w)]
+
+        # Assuming undirected graph
+        if v in self.adjList.keys():
+            self.adjList[v].append((u, w))
+        else:
+            self.adjList[v] = [(u, w)]
+
+    def show_graph(self):
+        # u -> v(w)
+        for u in self.adjList:
+            print(u, '->', ' -> '.join(str("{}({})".format(v, w))
+                                       for v, w in self.adjList[u]))
+
+    def dijkstra(self, src):
+        # Flush old junk values in par[]
+        self.par = [-1] * self.num_nodes
+        # src is the source node
+        self.dist[src] = 0
+        Q = PriorityQueue()
+        Q.insert((0, src))  # (dist from src, node)
+        for u in self.adjList.keys():
+            if u != src:
+                self.dist[u] = sys.maxsize  # Infinity
+                self.par[u] = -1
+
+        while not Q.isEmpty():
+            u = Q.extract_min()  # Returns node with the min dist from source
+            # Update the distance of all the neighbours of u and
+            # if their prev dist was INFINITY then push them in Q
+            for v, w in self.adjList[u]:
+                new_dist = self.dist[u] + w
+                if self.dist[v] > new_dist:
+                    if self.dist[v] == sys.maxsize:
+                        Q.insert((new_dist, v))
+                    else:
+                        Q.decrease_key((self.dist[v], v), new_dist)
+                    self.dist[v] = new_dist
+                    self.par[v] = u
+
+        # Show the shortest distances from src
+        self.show_distances(src)
+
+    def show_distances(self, src):
+        print("Distance from node: {}".format(src))
+        for u in range(self.num_nodes):
+            print('Node {} has distance: {}'.format(u, self.dist[u]))
+
+    def show_path(self, src, dest):
+        # To show the shortest path from src to dest
+        # WARNING: Use it *after* calling dijkstra
+        path = []
+        cost = 0
+        temp = dest
+        # Backtracking from dest to src
+        while self.par[temp] != -1:
+            path.append(temp)
+            if temp != src:
+                for v, w in self.adjList[temp]:
+                    if v == self.par[temp]:
+                        cost += w
+                        break
+            temp = self.par[temp]
+        path.append(src)
+        path.reverse()
+
+        print('----Path to reach {} from {}----'.format(dest, src))
+        for u in path:
+            print('{}'.format(u), end=' ')
+            if u != dest:
+                print('-> ', end='')
+
+        print('\nTotal cost of path: ', cost)
+
+
+if __name__ == '__main__':
+    graph = Graph(9)
+    graph.add_edge(0, 1, 4)
+    graph.add_edge(0, 7, 8)
+    graph.add_edge(1, 2, 8)
+    graph.add_edge(1, 7, 11)
+    graph.add_edge(2, 3, 7)
+    graph.add_edge(2, 8, 2)
+    graph.add_edge(2, 5, 4)
+    graph.add_edge(3, 4, 9)
+    graph.add_edge(3, 5, 14)
+    graph.add_edge(4, 5, 10)
+    graph.add_edge(5, 6, 2)
+    graph.add_edge(6, 7, 1)
+    graph.add_edge(6, 8, 6)
+    graph.add_edge(7, 8, 7)
+    graph.show_graph()
+    graph.dijkstra(0)
+    graph.show_path(0, 4)
+
+# OUTPUT
+# 0 -> 1(4) -> 7(8)
+# 1 -> 0(4) -> 2(8) -> 7(11)
+# 7 -> 0(8) -> 1(11) -> 6(1) -> 8(7)
+# 2 -> 1(8) -> 3(7) -> 8(2) -> 5(4)
+# 3 -> 2(7) -> 4(9) -> 5(14)
+# 8 -> 2(2) -> 6(6) -> 7(7)
+# 5 -> 2(4) -> 3(14) -> 4(10) -> 6(2)
+# 4 -> 3(9) -> 5(10)
+# 6 -> 5(2) -> 7(1) -> 8(6)
+# Distance from node: 0
+# Node 0 has distance: 0
+# Node 1 has distance: 4
+# Node 2 has distance: 12
+# Node 3 has distance: 19
+# Node 4 has distance: 21
+# Node 5 has distance: 11
+# Node 6 has distance: 9
+# Node 7 has distance: 8
+# Node 8 has distance: 14
+# ----Path to reach 4 from 0----
+# 0 -> 7 -> 6 -> 5 -> 4
+# Total cost of path:  21

data_structures/LinkedList/singly_LinkedList.py

@@ -3,22 +3,15 @@ class Node:#create a Node
         self.data=data#given data
         self.next=None#given next to None
 class Linked_List:
+    
     pass
-    def insert_tail(Head,data):#insert the data at tail
-        tamp=Head#create a tamp as a head
-        if(tamp==None):#if linkedlist is empty
-            newNod=Node()#create newNode Node type and given data and next
-            newNod.data=data
-            newNod.next=None
-            Head=newNod
+    
+    def insert_tail(Head,data):
+        if(Head.next is None):
+            Head.next = Node(data)
         else:
-            while tamp.next!=None:#find  the last Node
-                tamp=tamp.next
-            newNod = Node()#create a new node
-            newNod.data = data
-            newNod.next = None
-            tamp.next=newNod#put the newnode into last node
-        return Head#return first node  of  linked list  
+            insert_tail(Head.next, data)
+
     def insert_head(Head,data):
         tamp = Head
         if (tamp == None):
@@ -32,16 +25,18 @@ class Linked_List:
             newNod.next = Head#put the Head at NewNode Next
             Head=newNod#make a NewNode to Head
         return Head   
-    def Print(Head):#print every node data
-        tamp=Node()
+    
+    def printList(Head):#print every node data
         tamp=Head
         while tamp!=None:
             print(tamp.data)
             tamp=tamp.next
+    
     def delete_head(Head):#delete from head
         if Head!=None:
             Head=Head.next
         return Head#return new Head
+    
     def delete_tail(Head):#delete from tail
         if Head!=None:
             tamp = Node()
@@ -50,12 +45,6 @@ class Linked_List:
                 tamp = tamp.next
             tamp.next=None#delete the last element by give next None to 2nd last Element
         return Head
+
     def isEmpty(Head):
-        if(Head==None):#check Head is None  or Not
-            return True#return Ture if list is empty
-        else:
-            return False#check False if it's not empty
-
-
-
-
+        return Head is None #Return if Head is none

data_structures/Stacks/Balanced_Parentheses.py

@@ -1,27 +0,0 @@
-# Author: OMKAR PATHAK
-
-import Stack
-
-def parseParenthesis(string):
-    balanced = 1
-    index = 0
-    myStack = Stack.Stack(len(string))
-    while (index < len(string)) and (balanced == 1):
-        check = string[index]
-        if check == '(':
-            myStack.push(check)
-        else:
-            if myStack.isEmpty():
-                balanced = 0
-            else:
-                myStack.pop()
-        index += 1
-
-    if balanced == 1 and myStack.isEmpty():
-        return True
-    else:
-        return False
-
-if __name__ == '__main__':
-    print(parseParenthesis('((()))'))   # True
-    print(parseParenthesis('((())'))    # False

data_structures/Stacks/Infix_To_Postfix_Conversion.py

@@ -1,48 +0,0 @@
-# Author: OMKAR PATHAK
-
-import Stack
-
-def isOperand(char):
-    return (ord(char) >= ord('a') and ord(char) <= ord('z')) or (ord(char) >= ord('A') and ord(char) <= ord('Z'))
-
-def precedence(char):
-    if char == '+' or char == '-':
-        return 1
-    elif char == '*' or char  == '/':
-        return 2
-    elif char == '^':
-        return 3
-    else:
-        return -1
-
-def infixToPostfix(myExp, myStack):
-    postFix = []
-    for i in range(len(myExp)):
-        if (isOperand(myExp[i])):
-            postFix.append(myExp[i])
-        elif(myExp[i] == '('):
-            myStack.push(myExp[i])
-        elif(myExp[i] == ')'):
-            topOperator = myStack.pop()
-            while(not myStack.isEmpty() and topOperator != '('):
-                postFix.append(topOperator)
-                topOperator = myStack.pop()
-        else:
-            while (not myStack.isEmpty()) and (precedence(myExp[i]) <= precedence(myStack.peek())):
-                postFix.append(myStack.pop())
-            myStack.push(myExp[i])
-
-    while(not myStack.isEmpty()):
-        postFix.append(myStack.pop())
-    return ' '.join(postFix)
-
-if __name__ == '__main__':
-    myExp = 'a+b*(c^d-e)^(f+g*h)-i'
-    myExp = [i for i in myExp]
-    print('Infix:',' '.join(myExp))
-    myStack = Stack.Stack(len(myExp))
-    print('Postfix:',infixToPostfix(myExp, myStack))
-
-    # OUTPUT:
-    # Infix: a + b * ( c ^ d - e ) ^ ( f + g * h ) - i
-    # Postfix: a b c d ^ e - f g h * + ^ * + i -

data_structures/Stacks/Stack.py

@@ -1,50 +0,0 @@
-# Author: OMKAR PATHAK
-
-class Stack(object):
-    def __init__(self, limit = 10):
-        self.stack = []
-        self.limit = limit
-
-    # for printing the stack contents
-    def __str__(self):
-        return ' '.join([str(i) for i in self.stack])
-
-    # for pushing an element on to the stack
-    def push(self, data):
-        if len(self.stack) >= self.limit:
-            print('Stack Overflow')
-        else:
-            self.stack.append(data)
-
-    # for popping the uppermost element
-    def pop(self):
-        if len(self.stack) <= 0:
-            return -1
-        else:
-            return self.stack.pop()
-
-    # for peeking the top-most element of the stack
-    def peek(self):
-        if len(self.stack) <= 0:
-            return -1
-        else:
-            return self.stack[len(self.stack) - 1]
-
-    # to check if stack is empty
-    def isEmpty(self):
-        return self.stack == []
-
-    # for checking the size of stack
-    def size(self):
-        return len(self.stack)
-
-if __name__ == '__main__':
-    myStack = Stack()
-    for i in range(10):
-        myStack.push(i)
-    print(myStack)
-    myStack.pop()           # popping the top element
-    print(myStack)
-    myStack.peek()          # printing the top element
-    myStack.isEmpty()
-    myStack.size()

data_structures/Stacks/balanced_parentheses.py

@@ -0,0 +1,21 @@
+from Stack import Stack
+
+__author__ = 'Omkar Pathak'
+
+
+def balanced_parentheses(parentheses):
+    """ Use a stack to check if a string of parentheses are balanced."""
+    stack = Stack(len(parentheses))
+    for parenthesis in parentheses:
+        if parenthesis == '(':
+            stack.push(parenthesis)
+        elif parenthesis == ')':
+            stack.pop()
+    return not stack.is_empty()
+
+
+if __name__ == '__main__':
+    examples = ['((()))', '((())']
+    print('Balanced parentheses demonstration:\n')
+    for example in examples:
+        print(example + ': ' + str(balanced_parentheses(example)))

data_structures/Stacks/infix_to_postfix_conversion.py

@@ -0,0 +1,62 @@
+import string
+
+from Stack import Stack
+
+__author__ = 'Omkar Pathak'
+
+
+def is_operand(char):
+    return char in string.ascii_letters or char in string.digits
+
+
+def precedence(char):
+    """ Return integer value representing an operator's precedence, or
+    order of operation.
+
+    https://en.wikipedia.org/wiki/Order_of_operations
+    """
+    dictionary = {'+': 1, '-': 1,
+                  '*': 2, '/': 2,
+                  '^': 3}
+    return dictionary.get(char, -1)
+
+
+def infix_to_postfix(expression):
+    """ Convert infix notation to postfix notation using the Shunting-yard
+    algorithm.
+
+    https://en.wikipedia.org/wiki/Shunting-yard_algorithm
+    https://en.wikipedia.org/wiki/Infix_notation
+    https://en.wikipedia.org/wiki/Reverse_Polish_notation
+    """
+    stack = Stack(len(expression))
+    postfix = []
+    for char in expression:
+        if is_operand(char):
+            postfix.append(char)
+        elif char not in {'(', ')'}:
+            while (not stack.is_empty()
+                    and precedence(char) <= precedence(stack.peek())):
+                postfix.append(stack.pop())
+            stack.push(char)
+        elif char == '(':
+            stack.push(char)
+        elif char == ')':
+            while not stack.is_empty() and stack.peek() != '(':
+                postfix.append(stack.pop())
+            # Pop '(' from stack. If there is no '(', there is a mismatched
+            # parentheses.
+            if stack.peek() != '(':
+                raise ValueError('Mismatched parentheses')
+            stack.pop()
+    while not stack.is_empty():
+        postfix.append(stack.pop())
+    return ' '.join(postfix)
+
+
+if __name__ == '__main__':
+    expression = 'a+b*(c^d-e)^(f+g*h)-i'
+
+    print('Infix to Postfix Notation demonstration:\n')
+    print('Infix notation: ' + expression)
+    print('Postfix notation: ' + infix_to_postfix(expression))

data_structures/Stacks/next.py

@@ -0,0 +1,16 @@
+# Function to print element and NGE pair for all elements of list
+def printNGE(arr):
+ 
+    for i in range(0, len(arr), 1):
+ 
+        next = -1
+        for j in range(i+1, len(arr), 1):
+            if arr[i] < arr[j]:
+                next = arr[j]
+                break
+             
+        print(str(arr[i]) + " -- " + str(next))
+ 
+# Driver program to test above function
+arr = [11,13,21,3]
+printNGE(arr)

data_structures/Stacks/stack.py

@@ -0,0 +1,68 @@
+__author__ = 'Omkar Pathak'
+
+
+class Stack(object):
+    """ A stack is an abstract data type that serves as a collection of
+    elements with two principal operations: push() and pop(). push() adds an
+    element to the top of the stack, and pop() removes an element from the top
+    of a stack. The order in which elements come off of a stack are
+    Last In, First Out (LIFO).
+
+    https://en.wikipedia.org/wiki/Stack_(abstract_data_type)
+    """
+
+    def __init__(self, limit=10):
+        self.stack = []
+        self.limit = limit
+
+    def __bool__(self):
+        return not bool(self.stack)
+
+    def __str__(self):
+        return str(self.stack)
+
+    def push(self, data):
+        """ Push an element to the top of the stack."""
+        if len(self.stack) >= self.limit:
+            raise StackOverflowError
+        self.stack.append(data)
+
+    def pop(self):
+        """ Pop an element off of the top of the stack."""
+        if self.stack:
+            return self.stack.pop()
+        else:
+            raise IndexError('pop from an empty stack')
+
+    def peek(self):
+        """ Peek at the top-most element of the stack."""
+        if self.stack:
+            return self.stack[-1]
+
+    def is_empty(self):
+        """ Check if a stack is empty."""
+        return not bool(self.stack)
+
+    def size(self):
+        """ Return the size of the stack."""
+        return len(self.stack)
+
+
+class StackOverflowError(BaseException):
+    pass
+
+
+if __name__ == '__main__':
+    stack = Stack()
+    for i in range(10):
+        stack.push(i)
+
+    print('Stack demonstration:\n')
+    print('Initial stack: ' + str(stack))
+    print('pop(): ' + str(stack.pop()))
+    print('After pop(), the stack is now: ' + str(stack))
+    print('peek(): ' + str(stack.peek()))
+    stack.push(100)
+    print('After push(100), the stack is now: ' + str(stack))
+    print('is_empty(): ' + str(stack.is_empty()))
+    print('size(): ' + str(stack.size()))

dynamic_programming/FloydWarshall.py

@@ -0,0 +1,37 @@
+import math
+
+class Graph:
+    
+    def __init__(self, N = 0): # a graph with Node 0,1,...,N-1
+        self.N = N
+        self.W = [[math.inf for j in range(0,N)] for i in range(0,N)] # adjacency matrix for weight
+        self.dp = [[math.inf for j in range(0,N)] for i in range(0,N)] # dp[i][j] stores minimum distance from i to j
+
+    def addEdge(self, u, v, w):
+        self.dp[u][v] = w;
+
+    def floyd_warshall(self):
+        for k in range(0,self.N):
+            for i in range(0,self.N):
+                for j in range(0,self.N):
+                    self.dp[i][j] = min(self.dp[i][j], self.dp[i][k] + self.dp[k][j])
+
+    def showMin(self, u, v):
+        return self.dp[u][v]
+    
+if __name__ == '__main__':
+    graph = Graph(5)
+    graph.addEdge(0,2,9)
+    graph.addEdge(0,4,10)
+    graph.addEdge(1,3,5)
+    graph.addEdge(2,3,7)
+    graph.addEdge(3,0,10)
+    graph.addEdge(3,1,2)
+    graph.addEdge(3,2,1)
+    graph.addEdge(3,4,6)
+    graph.addEdge(4,1,3)
+    graph.addEdge(4,2,4)
+    graph.addEdge(4,3,9)
+    graph.floyd_warshall()
+    graph.showMin(1,4)
+    graph.showMin(0,3)

dynamic_programming/fastfibonacci.py

@@ -0,0 +1,42 @@
+"""
+This program calculates the nth Fibonacci number in O(log(n)).
+It's possible to calculate F(1000000) in less than a second.
+"""
+import sys
+
+
+# returns F(n)
+def fibonacci(n: int):
+    if n < 0:
+        raise ValueError("Negative arguments are not supported")
+    return _fib(n)[0]
+
+
+# returns (F(n), F(n-1))
+def _fib(n: int):
+    if n == 0:
+        # (F(0), F(1))
+        return (0, 1)
+    else:
+        # F(2n) = F(n)[2F(n+1) − F(n)]
+        # F(2n+1) = F(n+1)^2+F(n)^2
+        a, b = _fib(n // 2)
+        c = a * (b * 2 - a)
+        d = a * a + b * b
+        if n % 2 == 0:
+            return (c, d)
+        else:
+            return (d, c + d)
+
+
+if __name__ == "__main__":
+    args = sys.argv[1:]
+    if len(args) != 1:
+        print("Too few or too much parameters given.")
+        exit(1)
+    try:
+        n = int(args[0])
+    except ValueError:
+        print("Could not convert data to an integer.")
+        exit(1)
+    print("F(%d) = %d" % (n, fibonacci(n)))

dynamic_programming/fibonacci.py

@@ -30,7 +30,7 @@ if __name__ == '__main__':
     import sys
 
     print("\n********* Fibonacci Series Using Dynamic Programming ************\n")
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

dynamic_programming/longest_increasing_subsequence_O(nlogn).py

@@ -0,0 +1,40 @@
+#############################
+# Author: Aravind Kashyap
+# File: lis.py
+# comments: This programme outputs the Longest Strictly Increasing Subsequence in O(NLogN)
+#           Where N is the Number of elements in the list 
+#############################
+def CeilIndex(v,l,r,key):
+	while r-l > 1:
+		m = (l + r)/2
+		if v[m] >= key:
+			r = m
+		else:
+			l = m
+	
+	return r
+	
+
+def LongestIncreasingSubsequenceLength(v):
+	if(len(v) == 0):
+		return 0	
+	
+	tail = [0]*len(v)
+	length = 1
+	
+	tail[0] = v[0]
+	
+	for i in range(1,len(v)):
+		if v[i] < tail[0]:
+			tail[0] = v[i]
+		elif v[i] > tail[length-1]:
+			tail[length] = v[i]
+			length += 1			
+		else:
+			tail[CeilIndex(tail,-1,length-1,v[i])] = v[i]
+	
+	return length
+	
+
+v = [2, 5, 3, 7, 11, 8, 10, 13, 6]
+print LongestIncreasingSubsequenceLength(v)

machine_learning/decision_tree.py

@@ -0,0 +1,139 @@
+"""
+Implementation of a basic regression decision tree.
+Input data set: The input data set must be 1-dimensional with continuous labels.
+Output: The decision tree maps a real number input to a real number output. 
+"""
+
+import numpy as np
+
+class Decision_Tree:
+    def __init__(self, depth = 5, min_leaf_size = 5):
+        self.depth = depth
+        self.decision_boundary = 0
+        self.left = None
+        self.right = None
+        self.min_leaf_size = min_leaf_size
+        self.prediction = None
+
+    def mean_squared_error(self, labels, prediction):
+        """
+        mean_squared_error:
+        @param labels: a one dimensional numpy array 
+        @param prediction: a floating point value
+        return value: mean_squared_error calculates the error if prediction is used to estimate the labels
+        """
+        if labels.ndim != 1:
+            print("Error: Input labels must be one dimensional")
+
+        return np.mean((labels - prediction) ** 2)
+
+    def train(self, X, y):
+        """
+        train:
+        @param X: a one dimensional numpy array
+        @param y: a one dimensional numpy array. 
+        The contents of y are the labels for the corresponding X values
+
+        train does not have a return value
+        """
+
+        """
+        this section is to check that the inputs conform to our dimensionality constraints
+        """
+        if X.ndim != 1:
+            print("Error: Input data set must be one dimensional")
+            return
+        if len(X) != len(y):
+            print("Error: X and y have different lengths")
+            return
+        if y.ndim != 1:
+            print("Error: Data set labels must be one dimensional")
+            return
+
+        if len(X) < 2 * self.min_leaf_size:
+            self.prediction = np.mean(y)
+            return
+
+        if self.depth == 1:
+            self.prediction = np.mean(y)
+            return
+
+        best_split = 0
+        min_error = self.mean_squared_error(X,np.mean(y)) * 2
+
+
+        """
+        loop over all possible splits for the decision tree. find the best split.
+        if no split exists that is less than 2 * error for the entire array
+        then the data set is not split and the average for the entire array is used as the predictor
+        """
+        for i in range(len(X)):
+            if len(X[:i]) < self.min_leaf_size:
+                continue
+            elif len(X[i:]) < self.min_leaf_size:
+                continue
+            else:
+                error_left = self.mean_squared_error(X[:i], np.mean(y[:i]))
+                error_right = self.mean_squared_error(X[i:], np.mean(y[i:]))
+                error = error_left + error_right
+                if error < min_error:
+                    best_split = i
+                    min_error = error
+
+        if best_split != 0:
+            left_X = X[:best_split]
+            left_y = y[:best_split]
+            right_X = X[best_split:]
+            right_y = y[best_split:]
+
+            self.decision_boundary = X[best_split]
+            self.left = Decision_Tree(depth = self.depth - 1, min_leaf_size = self.min_leaf_size)
+            self.right = Decision_Tree(depth = self.depth - 1, min_leaf_size = self.min_leaf_size)
+            self.left.train(left_X, left_y)
+            self.right.train(right_X, right_y)
+        else:
+            self.prediction = np.mean(y)
+
+        return
+
+    def predict(self, x):
+        """
+        predict:
+        @param x: a floating point value to predict the label of
+        the prediction function works by recursively calling the predict function
+        of the appropriate subtrees based on the tree's decision boundary
+        """
+        if self.prediction is not None:
+            return self.prediction
+        elif self.left or self.right is not None:
+            if x >= self.decision_boundary:
+                return self.right.predict(x)
+            else:
+                return self.left.predict(x)
+        else:
+            print("Error: Decision tree not yet trained")
+            return None
+
+def main():
+    """
+    In this demonstration we're generating a sample data set from the sin function in numpy.
+    We then train a decision tree on the data set and use the decision tree to predict the
+    label of 10 different test values. Then the mean squared error over this test is displayed.
+    """
+    X = np.arange(-1., 1., 0.005)
+    y = np.sin(X)
+
+    tree = Decision_Tree(depth = 10, min_leaf_size = 10)
+    tree.train(X,y)
+
+    test_cases = (np.random.rand(10) * 2) - 1
+    predictions = np.array([tree.predict(x) for x in test_cases])
+    avg_error = np.mean((predictions - test_cases) ** 2)
+
+    print("Test values: " + str(test_cases))
+    print("Predictions: " + str(predictions))
+    print("Average error: " + str(avg_error))
+
+            
+if __name__ == '__main__':
+    main()

machine_learning/k_means_clust.py

@@ -0,0 +1,172 @@
+'''README, Author - Anurag Kumar(mailto:anuragkumarak95@gmail.com)
+
+Requirements:
+  - sklearn
+  - numpy
+  - matplotlib
+
+Python:
+  - 3.5
+
+Inputs:
+  - X , a 2D numpy array of features.
+  - k , number of clusters to create.
+  - initial_centroids , initial centroid values generated by utility function(mentioned in usage).
+  - maxiter , maximum number of iterations to process.
+  - heterogeneity , empty list that will be filled with hetrogeneity values if passed to kmeans func.
+
+Usage:
+  1. define 'k' value, 'X' features array and 'hetrogeneity' empty list
+  
+  2. create initial_centroids,
+        initial_centroids = get_initial_centroids(
+            X, 
+            k, 
+            seed=0 # seed value for initial centroid generation, None for randomness(default=None)
+            )
+
+  3. find centroids and clusters using kmeans function.
+  
+        centroids, cluster_assignment = kmeans(
+            X, 
+            k, 
+            initial_centroids, 
+            maxiter=400,
+            record_heterogeneity=heterogeneity, 
+            verbose=True # whether to print logs in console or not.(default=False)
+            )
+  
+  
+  4. Plot the loss function, hetrogeneity values for every iteration saved in hetrogeneity list.
+        plot_heterogeneity(
+            heterogeneity, 
+            k
+        )
+  
+  5. Have fun..
+  
+'''
+from sklearn.metrics import pairwise_distances
+import numpy as np
+
+TAG = 'K-MEANS-CLUST/ '
+
+def get_initial_centroids(data, k, seed=None):
+    '''Randomly choose k data points as initial centroids'''
+    if seed is not None: # useful for obtaining consistent results
+        np.random.seed(seed)
+    n = data.shape[0] # number of data points
+        
+    # Pick K indices from range [0, N).
+    rand_indices = np.random.randint(0, n, k)
+    
+    # Keep centroids as dense format, as many entries will be nonzero due to averaging.
+    # As long as at least one document in a cluster contains a word,
+    # it will carry a nonzero weight in the TF-IDF vector of the centroid.
+    centroids = data[rand_indices,:]
+    
+    return centroids
+
+def centroid_pairwise_dist(X,centroids):
+    return pairwise_distances(X,centroids,metric='euclidean')
+
+def assign_clusters(data, centroids):
+    
+    # Compute distances between each data point and the set of centroids:
+    # Fill in the blank (RHS only)
+    distances_from_centroids = centroid_pairwise_dist(data,centroids)
+    
+    # Compute cluster assignments for each data point:
+    # Fill in the blank (RHS only)
+    cluster_assignment = np.argmin(distances_from_centroids,axis=1)
+    
+    return cluster_assignment
+
+def revise_centroids(data, k, cluster_assignment):
+    new_centroids = []
+    for i in range(k):
+        # Select all data points that belong to cluster i. Fill in the blank (RHS only)
+        member_data_points = data[cluster_assignment==i]
+        # Compute the mean of the data points. Fill in the blank (RHS only)
+        centroid = member_data_points.mean(axis=0)
+        new_centroids.append(centroid)
+    new_centroids = np.array(new_centroids)
+    
+    return new_centroids
+
+def compute_heterogeneity(data, k, centroids, cluster_assignment):
+    
+    heterogeneity = 0.0
+    for i in range(k):
+        
+        # Select all data points that belong to cluster i. Fill in the blank (RHS only)
+        member_data_points = data[cluster_assignment==i, :]
+        
+        if member_data_points.shape[0] > 0: # check if i-th cluster is non-empty
+            # Compute distances from centroid to data points (RHS only)
+            distances = pairwise_distances(member_data_points, [centroids[i]], metric='euclidean')
+            squared_distances = distances**2
+            heterogeneity += np.sum(squared_distances)
+        
+    return heterogeneity
+
+from matplotlib import pyplot as plt
+def plot_heterogeneity(heterogeneity, k):
+    plt.figure(figsize=(7,4))
+    plt.plot(heterogeneity, linewidth=4)
+    plt.xlabel('# Iterations')
+    plt.ylabel('Heterogeneity')
+    plt.title('Heterogeneity of clustering over time, K={0:d}'.format(k))
+    plt.rcParams.update({'font.size': 16})
+    plt.show()
+
+def kmeans(data, k, initial_centroids, maxiter=500, record_heterogeneity=None, verbose=False):
+    '''This function runs k-means on given data and initial set of centroids.
+       maxiter: maximum number of iterations to run.(default=500)
+       record_heterogeneity: (optional) a list, to store the history of heterogeneity as function of iterations
+                             if None, do not store the history.
+       verbose: if True, print how many data points changed their cluster labels in each iteration'''
+    centroids = initial_centroids[:]
+    prev_cluster_assignment = None
+    
+    for itr in range(maxiter):        
+        if verbose:
+            print(itr, end='')
+        
+        # 1. Make cluster assignments using nearest centroids
+        cluster_assignment = assign_clusters(data,centroids)
+            
+        # 2. Compute a new centroid for each of the k clusters, averaging all data points assigned to that cluster.
+        centroids = revise_centroids(data,k, cluster_assignment)
+            
+        # Check for convergence: if none of the assignments changed, stop
+        if prev_cluster_assignment is not None and \
+          (prev_cluster_assignment==cluster_assignment).all():
+            break
+        
+        # Print number of new assignments 
+        if prev_cluster_assignment is not None:
+            num_changed = np.sum(prev_cluster_assignment!=cluster_assignment)
+            if verbose:
+                print('    {0:5d} elements changed their cluster assignment.'.format(num_changed))   
+        
+        # Record heterogeneity convergence metric
+        if record_heterogeneity is not None:
+            # YOUR CODE HERE
+            score = compute_heterogeneity(data,k,centroids,cluster_assignment)
+            record_heterogeneity.append(score)
+        
+        prev_cluster_assignment = cluster_assignment[:]
+        
+    return centroids, cluster_assignment
+
+# Mock test below
+if False: # change to true to run this test case.
+    import sklearn.datasets as ds
+    dataset = ds.load_iris()
+    k = 3
+    heterogeneity = []
+    initial_centroids = get_initial_centroids(dataset['data'], k, seed=0)
+    centroids, cluster_assignment = kmeans(dataset['data'], k, initial_centroids, maxiter=400,
+                                        record_heterogeneity=heterogeneity, verbose=True)
+    plot_heterogeneity(heterogeneity, k)

other/LinearCongruentialGenerator.py

@@ -0,0 +1,34 @@
+__author__ = "Tobias Carryer"
+
+from time import time
+
+class LinearCongruentialGenerator(object):
+    """
+    A pseudorandom number generator.
+    """
+    
+    def __init__( self, multiplier, increment, modulo, seed=int(time()) ):
+        """
+        These parameters are saved and used when nextNumber() is called.
+     
+        modulo is the largest number that can be generated (exclusive). The most
+        efficent values are powers of 2. 2^32 is a common value.
+        """
+        self.multiplier = multiplier
+        self.increment = increment
+        self.modulo = modulo
+        self.seed = seed
+        
+    def next_number( self ):
+        """
+        The smallest number that can be generated is zero.
+        The largest number that can be generated is modulo-1. modulo is set in the constructor.
+        """
+        self.seed = (self.multiplier * self.seed + self.increment) % self.modulo
+        return self.seed
+
+if __name__ == "__main__":
+    # Show the LCG in action.
+    lcg = LinearCongruentialGenerator(1664525, 1013904223, 2<<31)
+    while True :
+        print lcg.next_number()

other/binary_exponentiation.py

@@ -0,0 +1,49 @@
+"""
+* Binary Exponentiation for Powers
+* This is a method to find a^b in a time complexity of O(log b)
+* This is one of the most commonly used methods of finding powers.
+* Also useful in cases where solution to (a^b)%c is required,
+* where a,b,c can be numbers over the computers calculation limits.
+* Done using iteration, can also be done using recursion
+
+* @author chinmoy159
+* @version 1.0 dated 10/08/2017
+"""
+
+
+def b_expo(a, b):
+    res = 1
+    while b > 0:
+        if b&1:
+            res *= a
+
+        a *= a
+        b >>= 1
+
+    return res
+
+
+def b_expo_mod(a, b, c):
+    res = 1
+    while b > 0:
+        if b&1:
+            res = ((res%c) * (a%c)) % c
+
+        a *= a
+        b >>= 1
+
+    return res
+
+"""
+* Wondering how this method works !
+* It's pretty simple.
+* Let's say you need to calculate a ^ b
+* RULE 1 : a ^ b = (a*a) ^ (b/2) ---- example : 4 ^ 4 = (4*4) ^ (4/2) = 16 ^ 2
+* RULE 2 : IF b is ODD, then ---- a ^ b = a * (a ^ (b - 1)) :: where (b - 1) is even.
+* Once b is even, repeat the process to get a ^ b
+* Repeat the process till b = 1 OR b = 0, because a^1 = a AND a^0 = 1
+*
+* As far as the modulo is concerned,
+* the fact : (a*b) % c = ((a%c) * (b%c)) % c
+* Now apply RULE 1 OR 2 whichever is required.
+"""

other/binary_exponentiation_2.py

@@ -0,0 +1,50 @@
+"""
+* Binary Exponentiation with Multiplication
+* This is a method to find a*b in a time complexity of O(log b)
+* This is one of the most commonly used methods of finding result of multiplication.
+* Also useful in cases where solution to (a*b)%c is required,
+* where a,b,c can be numbers over the computers calculation limits.
+* Done using iteration, can also be done using recursion
+
+* @author chinmoy159
+* @version 1.0 dated 10/08/2017
+"""
+
+
+def b_expo(a, b):
+    res = 0
+    while b > 0:
+        if b&1:
+            res += a
+
+        a += a
+        b >>= 1
+
+    return res
+
+
+def b_expo_mod(a, b, c):
+    res = 0
+    while b > 0:
+        if b&1:
+            res = ((res%c) + (a%c)) % c
+
+        a += a
+        b >>= 1
+
+    return res
+
+
+"""
+* Wondering how this method works !
+* It's pretty simple.
+* Let's say you need to calculate a ^ b
+* RULE 1 : a * b = (a+a) * (b/2) ---- example : 4 * 4 = (4+4) * (4/2) = 8 * 2
+* RULE 2 : IF b is ODD, then ---- a * b = a + (a * (b - 1)) :: where (b - 1) is even.
+* Once b is even, repeat the process to get a * b
+* Repeat the process till b = 1 OR b = 0, because a*1 = a AND a*0 = 0
+*
+* As far as the modulo is concerned,
+* the fact : (a+b) % c = ((a%c) + (b%c)) % c
+* Now apply RULE 1 OR 2, whichever is required.
+"""

other/euclidean_gcd.py

@@ -0,0 +1,18 @@
+# https://en.wikipedia.org/wiki/Euclidean_algorithm
+
+def euclidean_gcd(a, b):
+    while b:
+        t = b
+        b = a % b
+        a = t
+    return a
+
+def main():
+    print("GCD(3, 5) = " + str(euclidean_gcd(3, 5)))
+    print("GCD(5, 3) = " + str(euclidean_gcd(5, 3)))
+    print("GCD(1, 3) = " + str(euclidean_gcd(1, 3)))
+    print("GCD(3, 6) = " + str(euclidean_gcd(3, 6)))
+    print("GCD(6, 3) = " + str(euclidean_gcd(6, 3)))
+
+if __name__ == '__main__':
+    main()

searches/binary_search.py

@@ -110,10 +110,10 @@ def binary_search_by_recursion(sorted_collection, item, left, right):
     if sorted_collection[midpoint] == item:
         return midpoint
     elif sorted_collection[midpoint] > item:
-        return binary_search_by_recursion(sorted_collection, item, left, right-1)
+        return binary_search_by_recursion(sorted_collection, item, left, midpoint-1)
     else:
-        return binary_search_by_recursion(sorted_collection, item, left+1, right)
-
+        return binary_search_by_recursion(sorted_collection, item, midpoint+1, right)
+      
 def __assert_sorted(collection):
     """Check if collection is sorted, if not - raises :py:class:`ValueError`
 
@@ -137,14 +137,14 @@ def __assert_sorted(collection):
 
 if __name__ == '__main__':
     import sys
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input
     else:
         input_function = input
 
-    user_input = input_function('Enter numbers separated by coma:\n')
+    user_input = input_function('Enter numbers separated by comma:\n')
     collection = [int(item) for item in user_input.split(',')]
     try:
         __assert_sorted(collection)

searches/interpolation_search.py

@@ -0,0 +1,102 @@
+"""
+This is pure python implementation of interpolation search algorithm
+"""
+from __future__ import print_function
+import bisect
+
+
+def interpolation_search(sorted_collection, item):
+    """Pure implementation of interpolation search algorithm in Python
+    Be careful collection must be sorted, otherwise result will be
+    unpredictable
+    :param sorted_collection: some sorted collection with comparable items
+    :param item: item value to search
+    :return: index of found item or None if item is not found
+    """
+    left = 0
+    right = len(sorted_collection) - 1
+
+    while left <= right:
+        point = left + ((item - sorted_collection[left]) * (right - left)) // (sorted_collection[right] - sorted_collection[left])
+        
+        #out of range check
+        if point<0 or point>=len(sorted_collection):
+            return None
+
+        current_item = sorted_collection[point]
+        if current_item == item:
+            return point
+        else:
+            if item < current_item:
+                right = point - 1
+            else:
+                left = point + 1
+    return None
+
+
+def interpolation_search_by_recursion(sorted_collection, item, left, right):
+
+    """Pure implementation of interpolation search algorithm in Python by recursion
+    Be careful collection must be sorted, otherwise result will be
+    unpredictable
+    First recursion should be started with left=0 and right=(len(sorted_collection)-1)
+    :param sorted_collection: some sorted collection with comparable items
+    :param item: item value to search
+    :return: index of found item or None if item is not found
+    """
+    point = left + ((item - sorted_collection[left]) * (right - left)) // (sorted_collection[right] - sorted_collection[left])
+
+    #out of range check
+    if point<0 or point>=len(sorted_collection):
+        return None
+
+    if sorted_collection[point] == item:
+        return point
+    elif sorted_collection[point] > item:
+        return interpolation_search_by_recursion(sorted_collection, item, left, point-1)
+    else:
+        return interpolation_search_by_recursion(sorted_collection, item, point+1, right)
+      
+def __assert_sorted(collection):
+    """Check if collection is sorted, if not - raises :py:class:`ValueError`
+    :param collection: collection
+    :return: True if collection is sorted
+    :raise: :py:class:`ValueError` if collection is not sorted
+    Examples:
+    >>> __assert_sorted([0, 1, 2, 4])
+    True
+    >>> __assert_sorted([10, -1, 5])
+    Traceback (most recent call last):
+    ...
+    ValueError: Collection must be sorted
+    """
+    if collection != sorted(collection):
+        raise ValueError('Collection must be sorted')
+    return True
+
+
+if __name__ == '__main__':
+    import sys
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
+    # otherwise 2.x's input builtin function is too "smart"
+    if sys.version_info.major < 3:
+        input_function = raw_input
+    else:
+        input_function = input
+
+    user_input = input_function('Enter numbers separated by comma:\n')
+    collection = [int(item) for item in user_input.split(',')]
+    try:
+        __assert_sorted(collection)
+    except ValueError:
+        sys.exit('Sequence must be sorted to apply interpolation search')
+
+    target_input = input_function(
+        'Enter a single number to be found in the list:\n'
+    )
+    target = int(target_input)
+    result = interpolation_search(collection, target)
+    if result is not None:
+        print('{} found at positions: {}'.format(target, result))
+    else:
+        print('Not found')

searches/linear_search.py

@@ -41,7 +41,7 @@ def linear_search(sequence, target):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

searches/quick_select.py

@@ -0,0 +1,47 @@
+import collections
+import sys
+import random
+import time
+import math
+"""
+A python implementation of the quick select algorithm, which is efficient for calculating the value that would appear in the index of a list if it would be sorted, even if it is not already sorted
+https://en.wikipedia.org/wiki/Quickselect
+"""
+def _partition(data, pivot):
+    """
+    Three way partition the data into smaller, equal and greater lists,
+    in relationship to the pivot
+    :param data: The data to be sorted (a list)
+    :param pivot: The value to partition the data on
+    :return: Three list: smaller, equal and greater
+    """
+    less, equal, greater = [], [], []
+    for element in data:
+        if element.address < pivot.address:
+            less.append(element)
+        elif element.address > pivot.address:
+            greater.append(element)
+        else:
+            equal.append(element)
+    return less, equal, greater
+    
+    def quickSelect(list, k):
+    #k = len(list) // 2 when trying to find the median (index that value would be when list is sorted)
+      smaller = []
+      larger = []
+      pivot = random.randint(0, len(list) - 1)
+      pivot = list[pivot]
+      count = 0
+      smaller, equal, larger =_partition(list, pivot)
+      count = len(equal)
+      m = len(smaller)
+
+      #k is the pivot
+      if m <= k < m + count:
+        return pivot
+    # must be in smaller
+      elif m > k:
+        return quickSelect(smaller, k)
+    #must be in larger
+      else:
+        return quickSelect(larger, k - (m + count))

searches/ternary_search.py

@@ -0,0 +1,112 @@
+'''
+This is a type of divide and conquer algorithm which divides the search space into
+3 parts and finds the target value based on the property of the array or list 
+(usually monotonic property).
+
+Time Complexity  : O(log3 N)
+Space Complexity : O(1)
+'''
+
+import sys
+
+# This is the precision for this function which can be altered.
+# It is recommended for users to keep this number greater than or equal to 10.
+precision = 10
+
+# This is the linear search that will occur after the search space has become smaller.
+def lin_search(left, right, A, target):
+    for i in range(left, right+1):
+        if(A[i] == target):
+            return i
+
+# This is the iterative method of the ternary search algorithm.
+def ite_ternary_search(A, target):
+    left = 0
+    right = len(A) - 1;
+    while(True):
+        if(left<right):
+            
+            if(right-left < precision):
+                return lin_search(left,right,A,target)
+
+            oneThird = (left+right)/3+1;
+            twoThird = 2*(left+right)/3+1;
+            
+            if(A[oneThird] == target):
+                return oneThird
+            elif(A[twoThird] == target):
+                return twoThird
+            
+            elif(target < A[oneThird]):
+                right = oneThird-1
+            elif(A[twoThird] < target):
+                left = twoThird+1
+            
+            else:
+                left = oneThird+1
+                right = twoThird-1
+        else:
+            return None
+
+# This is the recursive method of the ternary search algorithm.
+def rec_ternary_search(left, right, A, target):
+    if(left<right):
+        
+        if(right-left < precision):
+            return lin_search(left,right,A,target)
+
+        oneThird = (left+right)/3+1;
+        twoThird = 2*(left+right)/3+1;
+
+        if(A[oneThird] == target):
+            return oneThird
+        elif(A[twoThird] == target):
+            return twoThird
+        
+        elif(target < A[oneThird]):
+            return rec_ternary_search(left, oneThird-1, A, target)
+        elif(A[twoThird] < target):
+            return rec_ternary_search(twoThird+1, right, A, target)
+        
+        else:
+            return rec_ternary_search(oneThird+1, twoThird-1, A, target)
+    else:
+        return None
+
+# This function is to check if the array is sorted.
+def __assert_sorted(collection):
+    if collection != sorted(collection):
+        raise ValueError('Collection must be sorted')
+    return True
+
+
+if __name__ == '__main__':
+    
+    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # otherwise 2.x's input builtin function is too "smart"
+    
+    if sys.version_info.major < 3:
+        input_function = raw_input
+    else:
+        input_function = input
+
+    user_input = input_function('Enter numbers separated by coma:\n')
+    collection = [int(item) for item in user_input.split(',')]
+
+    try:
+        __assert_sorted(collection)
+    except ValueError:
+        sys.exit('Sequence must be sorted to apply the ternary search')
+
+    target_input = input_function(
+        'Enter a single number to be found in the list:\n'
+    )
+    target = int(target_input)
+    result1 = ite_ternary_search(collection, target)
+    result2 = rec_ternary_search(0, len(collection)-1, collection, target)
+    
+    if result2 is not None:
+        print('Iterative search: {} found at positions: {}'.format(target, result1))
+        print('Recursive search: {} found at positions: {}'.format(target, result2))
+    else:
+        print('Not found')

sorts/bogosort.py

@@ -41,7 +41,7 @@ def bogosort(collection):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/bubble_sort.py

@@ -41,7 +41,7 @@ def bubble_sort(collection):
 
 if __name__ == '__main__':
     import sys
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/cocktail_shaker_sort.py

@@ -23,7 +23,7 @@ def cocktail_shaker_sort(unsorted):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/counting_sort.py

@@ -0,0 +1,72 @@
+"""
+This is pure python implementation of counting sort algorithm
+For doctests run following command:
+python -m doctest -v counting_sort.py
+or
+python3 -m doctest -v counting_sort.py
+For manual testing run:
+python counting_sort.py
+"""
+
+from __future__ import print_function
+
+
+def counting_sort(collection):
+    """Pure implementation of counting sort algorithm in Python
+    :param collection: some mutable ordered collection with heterogeneous
+    comparable items inside
+    :return: the same collection ordered by ascending
+    Examples:
+    >>> counting_sort([0, 5, 3, 2, 2])
+    [0, 2, 2, 3, 5]
+    >>> counting_sort([])
+    []
+    >>> counting_sort([-2, -5, -45])
+    [-45, -5, -2]
+    """
+    # if the collection is empty, returns empty
+    if collection == []:
+        return []
+
+    # get some information about the collection
+    coll_len = len(collection)
+    coll_max = max(collection)
+    coll_min = min(collection)
+
+    # create the counting array
+    counting_arr_length = coll_max + 1 - coll_min
+    counting_arr = [0] * counting_arr_length
+
+    # count how much a number appears in the collection
+    for number in collection:
+        counting_arr[number - coll_min] += 1
+
+    # sum each position with it's predecessors. now, counting_arr[i] tells
+    # us how many elements <= i has in the collection
+    for i in range(1, counting_arr_length):
+        counting_arr[i] = counting_arr[i] + counting_arr[i-1]
+
+    # create the output collection
+    ordered = [0] * coll_len
+
+    # place the elements in the output, respecting the original order (stable
+    # sort) from end to begin, updating counting_arr
+    for i in reversed(range(0, coll_len)):
+        ordered[counting_arr[collection[i] - coll_min]-1] = collection[i]
+        counting_arr[collection[i] - coll_min] -= 1
+
+    return ordered
+
+
+if __name__ == '__main__':
+    import sys
+    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # otherwise 2.x's input builtin function is too "smart"
+    if sys.version_info.major < 3:
+        input_function = raw_input
+    else:
+        input_function = input
+
+    user_input = input_function('Enter numbers separated by a comma:\n')
+    unsorted = [int(item) for item in user_input.split(',')]
+    print(counting_sort(unsorted))

sorts/gnome_sort.py

@@ -21,7 +21,7 @@ def gnome_sort(unsorted):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/insertion_sort.py

@@ -41,7 +41,7 @@ def insertion_sort(collection):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/merge_sort.py

@@ -64,7 +64,7 @@ def merge_sort(collection):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/quick_sort.py

@@ -42,7 +42,7 @@ def quick_sort(ARRAY):
 if __name__ == '__main__':
     import sys
 
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/radix_sort.py

@@ -2,19 +2,19 @@ def radixsort(lst):
   RADIX = 10
   maxLength = False
   tmp , placement = -1, 1
- 
+
   while not maxLength:
     maxLength = True
     # declare and initialize buckets
     buckets = [list() for _ in range( RADIX )]
- 
+
     # split lst between lists
     for  i in lst:
-      tmp = i / placement
+      tmp = i // placement
       buckets[tmp % RADIX].append( i )
       if maxLength and tmp > 0:
         maxLength = False
- 
+
     # empty lists into lst array
     a = 0
     for b in range( RADIX ):
@@ -22,6 +22,6 @@ def radixsort(lst):
       for i in buck:
         lst[a] = i
         a += 1
- 
+
     # move to next
     placement *= RADIX

sorts/selection_sort.py

@@ -44,7 +44,7 @@ def selection_sort(collection):
 
 if __name__ == '__main__':
     import sys
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/shell_sort.py

@@ -45,7 +45,7 @@ def shell_sort(collection):
 
 if __name__ == '__main__':
     import sys
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input

sorts/timsort.py

@@ -0,0 +1,81 @@
+def binary_search(lst, item, start, end):
+    if start == end:
+        if lst[start] > item:
+            return start
+        else:
+            return start + 1
+    if start > end:
+        return start
+
+    mid = (start + end) // 2
+    if lst[mid] < item:
+        return binary_search(lst, item, mid + 1, end)
+    elif lst[mid] > item:
+        return binary_search(lst, item, start, mid - 1)
+    else:
+        return mid
+
+
+def insertion_sort(lst):
+    length = len(lst)
+
+    for index in range(1, length):
+        value = lst[index]
+        pos = binary_search(lst, value, 0, index - 1)
+        lst = lst[:pos] + [value] + lst[pos:index] + lst[index+1:]
+
+    return lst
+
+
+def merge(left, right):
+    if not left:
+        return right
+
+    if not right:
+        return left
+
+    if left[0] < right[0]:
+        return [left[0]] + merge(left[1:], right)
+
+    return [right[0]] + merge(left, right[1:])
+
+
+def timsort(lst):
+    runs, sorted_runs = [], []
+    length = len(lst)
+    new_run = [lst[0]]
+    sorted_array = []
+
+    for i in range(1, length):
+        if i == length - 1:
+            new_run.append(lst[i])
+            runs.append(new_run)
+            break
+
+        if lst[i] < lst[i - 1]:
+            if not new_run:
+                runs.append([lst[i - 1]])
+                new_run.append(lst[i])
+            else:
+                runs.append(new_run)
+                new_run = []
+        else:
+            new_run.append(lst[i])
+
+    for run in runs:
+        sorted_runs.append(insertion_sort(run))
+
+    for run in sorted_runs:
+        sorted_array = merge(sorted_array, run)
+
+    return sorted_array
+
+
+def main():
+
+    lst = [5,9,10,3,-4,5,178,92,46,-18,0,7]
+    sorted_lst = timsort(lst)
+    print(sorted_lst)
+
+if __name__ == '__main__':
+    main()

traversals/binary_tree_traversals.py

@@ -84,7 +84,7 @@ if __name__ == '__main__':
     import sys
 
     print("\n********* Binary Tree Traversals ************\n")
-    # For python 2.x and 3.x compatibility: 3.x has not raw_input builtin
+    # For python 2.x and 3.x compatibility: 3.x has no raw_input builtin
     # otherwise 2.x's input builtin function is too "smart"
     if sys.version_info.major < 3:
         input_function = raw_input