Aditya Chatterjee

machine learning

Understand basic TensorFlow programming concepts

We understand basic TensorFlow programming concepts using a code to add two numbers in TensorFlow. We created placeholders like tf.placeholder, operations like tf.constant and tf.add, sessions using tf.session and configuration using tf.configProto

machine learning

Training vs Inference

Training and Inference are two major processes of Machine Learning and is deeply connected. Training is the process by which we generate various parameters such as weights and biases which are used in a particular model. Inference is the process of using the trained model to do a particular task

machine learning

Deep Learning on 2-Dimensional Images

Applying deep learning concepts on images has proved to be one of the important work which has resulted in early detection of diseases resulting in saving millions of life to monitoring activities on the entire Earth We take a look at medical images, Satellite Images and the various Python libraries

machine learning

Deep Learning for Medical Imaging and Diagnosis

One of the major medical challenges that we face today is the early detection of diseases so that the proper threatment can be applied. This can be solved by applying machine learning to analyse MRI scan, CT Scan, Xray Scans, InfraRed Images, Arthoscopy, UV radiation, Scintigraphy and Ultrasound

machine learning

Floating point operations per second (FLOPS) of Machine Learning models

In this article, we take a look at the FLOPs values of various machine learning models like VGG19, VGG16, GoogleNet, ResNet18, ResNet34, ResNet50, ResNet152 and others. The FLOPS range from 19.6 billion to 0.72 billion.

machine learning

Evolution of CNN Architectures: LeNet, AlexNet, ZFNet, GoogleNet, VGG and ResNet

It all started with LeNet in 1998 and eventually, after nearly 15 years, lead to ground breaking models winning the ImageNet Large Scale Visual Recognition Challenge which includes AlexNet in 2012, ZFNet in 2013, GoogleNet in 2014, VGG in 2014, ResNet in 2015 to ensemble of previous models in 2016.

machine learning

BLAS vs BLIS

BLAS (Basic Linear Algebra Subprograms) and BLIS (BLAS Like Interface Software) are libraries that have revolutionized scientific computing by accelerating execution of mathematical operations on a vast range of platforms. In short, BLIS is the new generation alternative to BLAS

machine learning

Types of Neural Network optimizations

The types of neural network optimizations are weight pruning, structured pruning, convolution, fully-connected, structured group, structure ranking with activations like Lp norm, block pruning, model thinning, compression schedule, regularization, group lasso, group variance, quantization and others

similarity measurement

Minkowski distance

Minkowski distance is a distance/ similarity measurement between two points in the normed vector space (N dimensional real space) and is a generalization of the Euclidean distance and the Manhattan distance. See the applications of Minkowshi distance and its visualization using an unit circle.

similarity measurement

Damerau Levenshtein distance

Damerau Levenshtein distance is a variant of Levenshtein distance which is a type of Edit distance. Damerau stated that the four operations in Damerau Levenshtein distance correspond to more than 80% of all human misspellings. It adds an extra operation named transposition to its set of operations

similarity measurement

Levenshtein distance

evenshtein distance is a type of Edit distance which is a large class of distance metric of measuring the dissimilarity between two strings by computing a minimum number of operations (from a set of operations) used to convert one string to another string. It is a way of pairwise string alignment.

machine learning

Build / Install Eigen Library from source

In this article, we explore the way to build and install Eigen library from source using cmake. Eigen is an efficient open-source C++ library for linear algebra, matrix and vector operations, geometrical transformations, numerical solvers and related algorithms.

similarity measurement

Edit distance

Edit distance is a large class of distance metric of measuring the dissimilarity between two strings by computing a minimum number of operations (from a set of operations) used to convert one string to another string. It can be seen as a way of pairwise string alignment.

similarity measurement

Euclidean vs Manhattan vs Chebyshev Distance

Euclidean distance, Manhattan distance and Chebyshev distance are all distance metrics which compute a number based on two data points. All the three metrics are useful in various use cases and differ in some important aspects such as computation and real life usage.

similarity measurement

Chebyshev distance

Chebyshev distance is a distance metric which is the maximum absolute distance in one dimension of two N dimensional points. It has real world applications in Chess, Warehouse logistics and many other fields. It is known as Tchebychev distance, maximum metric, chessboard distance and L∞ metric.

similarity measurement

Euclidean distance (L2 norm)

Euclidean distance is the shortest distance between two points in an N dimensional space also known as Euclidean space. It is used as a common metric to measure the similarity between two data points and used in various fields such as geometry, data mining, deep learning and others.

similarity measurement

Manhattan distance (L1 norm)

Manhattan distance (L1 norm) is a distance metric between two points in a N dimensional vector space. It is the sum of the lengths of the projections of the line segment between the points onto the coordinate axes. It was introduced by Hermann Minkowski. It is used in regression analysis

tpu

What is Neural processing unit (NPU)?

A neural processing unit (NPU) is a microprocessor that specializes in the acceleration of machine learning algorithms. Examples include TPU by Google, NVDLA by Nvidia, EyeQ by Intel, Inferentia by Amazon, Ali-NPU by Alibaba, Kunlun by Baidu, Sophon by Bitmain, MLU by Cambricon, IPU by Graphcore

software engineering

Laws similar to Moore's Law

Laws similar to Moore's Law are Eroom's law, Engelbart's law, Haitz's law, Koomey's law, Carlson curve, Swanson's law, Pollack's rule, Wirth's law / Page's law / Gates's law / May's law / The great Moore's law compensator (TGMLC), Keck's law, Nielsen's Law and Dennard scaling.

software engineering

Moore's second law or Rock's law

Moore's second law or Rock's law, states that the cost of a semiconductor chip fabrication plant doubles every four years. Research and Development, manufacturing and testing have increased steadily with each new generation of chips Rising manufacturing costs are an important consideration

computational model

Stack Machine: A computational model

A Stack Machine is a computational model that uses a last-in, first-out stack to hold short-lived temporary values. Most of its instructions assume that operands will be from the stack, and results placed in the stack. It has its own advantages such as it does not need any addressing

software engineering

RISC vs CISC architecture

RISC and CISC differ based on the following factors: approach to improve computing performance, hardware and Software focus and hardware specifications. In general, both are equally useful. CISC is commonly used in automation devices whereas RISC is used in video and image processing applications.

software engineering

Moore's law

Moore's law is the observation that the number of transistors in a dense integrated circuit doubles about every two years. It states that computing would dramatically increase in power, and decrease in relative cost, at an exponential rate. It is named after Gordon Earle Moore, founder of Intel

tensorflow

How TensorFlow uses Graph data structure concepts?

In this article, we explain various concepts in TensorFlow such as tensors, dataflow graphs and several optimizations such as decision tree pruning and demonstrate the use of graph data structure and algorithm concepts in TensorFlow