Mix It Up!!!!

1. 12 Matrix Operations You Should Know While Starting your Deep Learning Journey

So, In this article, we will discuss important linear algebra matrix operations that are used in the description of deep learning methods. The topics which we will be discussing in this article are as follows: Matrices are rectangular arrays consisting of numbers and can be seen as 2nd-order tensors. Sometimes, instead of describing the full matrix components, we use the following abbreviation of a matrix: In this example, with the help of the NumPy library, we will create a matrix. If the shape of the matrices is not the same it throws an error saying, that addition or subtraction is not possible.

Categories: Matrix Operations

Level: Advanced

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/07/12-matrix-operations-you-should-know-while-starting-your-deep-learning-journey/

2. Artificial Neural Networks- 25 Questions to Test Your Skills on ANN

The loss function is used as a measure of accuracy to identify whether our neural network has learned the patterns accurately or not with the help of the training data. In Deep Learning, a good-performing neural network will have a low value of the loss function at all times when training happens. While selecting the learning rate to train the neural network, we have to choose the value very carefully due to the following reasons: If the learning rate is set too low, training of the model will continue very slowly as we are making very small changes to the weights since our step size that is governed by the equation of gradient descent is small. If the derivatives are large e.g, If using a ReLU-like activation function then the value of the gradient will increase exponentially as we propagate down the model until they eventually explode, and this is what we call the problem of Exploding gradient. So, for the same level of accuracy, deeper networks can be much more powerful and efficient in terms of both computation and the number of parameters to learn.

Categories: Artificial Neural Networks,?Data Science Interview Questions?

Level: Beginner

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/05/artificial-neural-networks-25-questions-to-test-your-skills-on-ann/

3. How do Neural Networks really work?

We can illustrate that in the example here: Now, what do all these numbers represent? Let’s take a look at it using a handy function that pandas provide us with: As shown above, the main way that computers interpret images is through the form of pixels, which are the smallest building blocks of any computer display. Before even calculating the predictions we have to ensure that the data is structured in the same way for the program to process all the different images. It can be thought of as the emphasis that is given to each data point for the program to work.

Categories: ?Artificial Intelligence,?Artificial Neural Network,?Machine Learning?

Level: Advanced

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/12/how-do-neural-networks-really-work/

4. Face Detection and Recognition capable of beating humans using FaceNet

Creating face recognition is considered to be a very easy task in the field of computer vision, but it is extremely tough to have a pipeline that can predict faces with complex backgrounds when you have multiple faces, different lighting conditions, and different scales of images. We need a technique that can generalize well irrespective of the number of samples it needs and how different the train and test data are. In our experiments, we found that dlib produces better results than HAAR, though we noticed some improvements could still be made. Why SVM, you may ask? With a lot of experience, I have come to know that SVM + DL-based features can outperform any other method even Deep learning methods when the amount of data is small. Once the SVM is trained, it’s time to do some testing, but our test data has multiple faces in a list.

Categories: FaceNet, Computer Vision, Deep Learning

Level: Advanced

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/06/face-detection-and-recognition-capable-of-beating-humans-using-facenet/

5. Part 4: Step by Step Guide to Master NLP – Text Cleaning Techniques

This article is part of an ongoing blog series on Natural Language Processing (NLP). In the previous part of this blog series, we complete the initial steps involved in text cleaning and preprocessing that are related to NLP. Now, in continuation of that part, in this article, we will cover the next techniques involved in the NLP pipeline of Text preprocessing. In this article, we will first discuss some more text cleaning techniques which might be useful in some NLP tasks and then we start our journey towards the normalization techniques, Stemming, and Lemmatization which are very crucial techniques that you must know while you are working with on an NLP based project. This is part-4 of the blog series on the Step by Step Guide to Natural Language Processing.

Categories: Data Cleaning,?NLP,?Python, Text Mining

Level: Beginner

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/06/part-4-step-by-step-guide-to-master-natural-language-processing-in-python/

6. Guide to Build Better Predictive Models using Segmentation

In addition, it is the common business intuition (which may not always have a sound statistical rationale), to develop separate models if the difference in response rates between adjacent nodes is at least 30% (e.g. if the response rate in a particular node is 0.7% and the same for the adjacent node is 0.5% then the difference in response rate is ~30%) The commonly adopted approach would suggest that one should build separate models for each of the terminal (or end) nodes, which have been depicted in green in Fig-1. For instance, one can use the following dummies (it should be noted that due to the degree of freedom constraint there will be one less than all possible number of dummies) The predictive power of the model will be even better if one uses dummies to replicate the segmentation tree These dummies would provide the same differentiation in response rate as that of the five individual segments. Fig-5: Variables across the 4 child models Fig-6: Predictive Pattern of the Variable “Number of purchases in the last 24 months” Across the Five Segments, In this case, one can observe that the predictive pattern of a particular variable is significantly different across segments. One can relate this philosophy (at a broad level) with the idea behind the creation of segments for developing models, wherein the objective of the segmentation is not to achieve a closer fit with the target but to identify interaction effects.

Categories: Linear Regression,?Logistic Regression for Segmentation,?Machine Learning for Market Segmentation,?Market Segmentation,?Marketing Analytics,?Regression for Segmentation,?

Level: Intermediate

Link to the entire article: https://www.analyticsvidhya.com/blog/2016/02/guide-build-predictive-models-segmentation/

7. Implementing Convolution As An Image Filter Using OpenCV

Image filtering is changing the pixel value of a specific image to blur, sharpen, emboss, or make edges more clear. Image filtering is application-specific, as sometimes we need to blur the image and sometimes we wish to have sharpness in the image. It is very easy to use and see the filters on our image. The box blur is a straightforward blur in which each pixel is set to the average of the pixels surrounding it.

Categories: Convolution,?OpenCV,?Computer Vision, Python?

Level: Beginner

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/08/implementing-convolution-as-an-image-filter-using-opencv/

8. Learn how to Build your own Speech-to-Text Model (using Python)

“Hey Google. What’s the weather like today?”

This will sound familiar to anyone who has owned a smartphone in the last decade. I can’t remember the last time I took the time to type out the entire query on Google Search. I simply ask the question – and Google lays out the entire weather pattern for me.

It saves me a ton of time and I can quickly glance at my screen and get back to work. A win-win for everyone! But how does Google understand what I’m saying? And how does Google’s system convert my query into text on my phone’s screen?

Categories: Convert Speech to Text,?NLP,?Speech Recognition,?Speech Recognition Model,?Speech To Text,?Speech To Text Model?

Level: Intermediate

Link to the entire article: https://www.analyticsvidhya.com/blog/2019/07/learn-build-first-speech-to-text-model-python/

9. The Ultimate Guide To Setting Up An ETL (Extract, Transform, and Load) Process Pipeline

ETL is a process that extracts data from multiple source systems, changes it (through calculations, concatenations, and so on), and then puts it into the Data Warehouse system. This section of sql_queries.py is the place where we are going to store all of our SQL queries for extracting from source databases and importing them into our target database (data warehouse). Here are some of the most famous examples: MarkLogic is a data warehousing system that uses an array of business capabilities to make data integration easier and faster. Creating an ETL pipeline from scratch for such data is a hard procedure since organizations will have to use a large number of resources in order to create this pipeline and then ensure that it can keep up with the high data volume and Schema changes.

Categories: Extract Transform Load, ETL Pipeline

Level: Beginner

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/11/the-ultimate-guide-to-setting-up-an-etl-extract-transform-and-load-process-pipeline/

10. Building an Interactive Dashboard using Bokeh and Pandas

A huge amount of data is being generated every instant due to business activities in globalization. Companies are extracting useful information from such generated data to make important business decisions. Exploratory Data analysis can help them visualize the current market situation and forecast the likely future trends, understand what their customers say and expect from the product, improve the product by taking suitable measures, and more.

Categories: Bokeh,?Dashboard,?Pandas, Data Visualisation

Level: Beginner

Link to the entire article: https://www.analyticsvidhya.com/blog/2021/09/building-an-interactive-dashboard-using-bokeh-and-pandas/