Activation functions are any functions that defines the output of a neuron. The activation function associated with each neurons in a neural network determines whether it should be activated or not, based on the output of that function. There are three types of activation functions- Binary, Linear and Non-Linear activation function. In this article we will focus mainly on non-linear activation functions. Input to the neural network is usually linear transformation (i.e. input*weight + bias), but most of the real world data are non-linear. So, to make that input non-linear, non-linear activatio ..read more

Following is the code to implement KNN algorithm from scratch in python import pandas as pd import numpy as np import operator # loading data file into the program. give the location of your csv file dataset = pd.read_csv("E:/input/iris.csv") print(dataset.head()) # prints first five tuples of your data. # making function for calculating euclidean distance def E_Distance(x1, x2, length):     distance = 0     for x in range(length):         distance += np.square(x1[x] - x2[x])     return np.sqrt(distance) # making function for defining K-NN model def knn(trainingSet, testInstance, k):     d ..read more

In computing, Data transformation is the method of transforming data from one format or structure to another format or structure. It is a fundamental aspect of most data integration and data management tasks such as data wrangling, data storage, data integration. In this process, we try to change the nature of data using some strategies, so that we can extract important information from it.   Some of the techniques used for data transformation are:   i.  Aggregation: In this technique the summation or aggregation operation is applied over the data. E.g. the daily sales data may be aggregated s ..read more

Decimal Normalization is a normalization technique in which we normalize the given value by moving the decimal points of that value. The number of decimal points to move is defined by the maximum absolute value of the given data set. If  Vi  value of attribute A, then normalized value Ui is given as,    Decimal Normalization Where, j is the smallest integer such that max|Ui|<1.        Let's clarify it with an example: Suppose we have data set in which the value ranges from -9900 to 9877.  In this case, the maximum absolute value is 9900. So to perform decimal normalization, we divi ..read more

Min-Max normalization performs the linear transformation on original data. Let (X1, X2) be a min and max boundary of an attribute and (Y1, Y2) be the new scale at which we are normalizing, then for Vi  value of the attribute, the normalized value Ui is given as, Min-Max Normalization  Special thing about min-max normalization is that preserves the relationship between the original data values. If in future the input values come to be beyond the limit of normalization, then it will encounter an error known as “out-of-bound error.”   Let’s Understand it with an example: Suppose the minim ..read more

Gradient descent is one of the most common method of training a neural network. It is an optimization algorithm used to optimize the parameters( eg weights and bias) in neural network. The way this works is you define a loss(cost) function  that tells how well your parameters(weights and bias) fits your training data. Greater the loss, poor is the fitting. Then we gradually reduce loss using the gradient descent algorithm. The reduction of loss includes the updating of weights and bias in the direction in which the loss descrease. The direction is determined by obtaining the gradients(derivati ..read more

Normalization or standardization is defined as the process of rescaling original data without changing its original behavior or nature. It is a technique often applied as part of data pre-processing in Machine Learning. The main aim of normalization is to change the value of data in dataset to a common scale, without distirting the differences in the ranges of value.We often define new boundary (most common is (0,1),(-1,1)) and convert data accordingly. This technique is useful in classification algorithms involving neural network or distance based algorithm (e.g. KNN, K-means).  In Z score n ..read more

An operator is a symbol that tells the compiler to perform some specific Logical or mathematical or calculations. C++ contains a lot of in built-in operators that carries operations like Arithmetic, Logical, relational etcThe meaning of an operator is always same for variable of basic types like: int, float, double etc. For example: To add two integers, + operator is used. However, for user-defined types (like: objects), you can redefine the way operator works.   For example: If there are two objects of a class that contains string as its data members. You can redefine the meaning of + ope ..read more

Machine Learning is one of the widely talked topic of present time. It has gain popularity because of its wide scope and applications. Almost every field whether it is education, health, research or business there is huge application of machine Learning. Machine Learning is a sub-set of artificial intelligence that uses different algorithms to learn from data and information autonomously. Computers (or any machines) need not be explicitly programmed in machine learning, but can alter and enhance their algorithms on their own. Nowadays, machine learning algorithms allow machines to interact wit ..read more

Linear Regression Algorithm is one of the simplest and easy Regression algorithms in Machine Learning. It is easy to code and implements. Python supports libraries like scikit learn which allows implementing Linear regression in a few lines of code. However, if we want to implement the algorithm from scratch we need to be a bit smart. In this article, we will see how to implement Linear regression from scratch in python using numpy only. You can directly download the code from here. !!! Strongly Recomended- Introduction to Linear Regression Algorithm The full code is as follows: import numpy a ..read more