Machine Learning

from keras import layers from keras import models seq_model= models.Sequential() seq_model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1))) seq_model.add(layers.MaxPooling2D((2, 2))) seq_model.add(layers.Conv2D(64, (3, 3), activation='relu')) seq_model.add(layers.MaxPooling2D((2, 2))) seq_model.add(layers.Conv2D(128, (3, 3), activation='relu'))There … [Read more...] about Convolutional Models Overview

Photo by Markus Spiske on UnsplashIn this article I will guide you through my thoughts on how to build a fuzzy search algorithm. A very practical use case of this algorithm is that we can use it to find alternative names for a brand saying ‘Amazon’ and we want it to return strings such as ‘AMZ’, ‘AMZN’ or ‘AMZN MKTP’.The article follows an outline as the following:Fuzzy search … [Read more...] about A Guide on How to Build a Fuzzy Search Algorithm with FuzzyWuzzy and HMNI

The theory was first published in the 2007 annals of Statistics by Gabor J. Szekely and others¹. It’s a measure of dependence between two paired random vectors of arbitrary, not necessarily equal, dimensions². It can also be used for K-sample testing with some design tweaks.What’s great about it? It’s nonparametric testing, meaning that there are no underlying assumptions about … [Read more...] about Hyperdimension Hypothesis Testing — Multiscale Graph Correlation and Distance Correlation

Deep learning neural network models used for predictive modeling may need to be updated. This may be because the data has changed since the model was developed and deployed, or it may be the case that additional labeled data has been made available since the model was developed and it is expected that the additional data will improve the performance of the model. It is … [Read more...] about How to Update Neural Network Models With More Data

Though we wish it were otherwise, the customer journey is rarely a straight line. Today, the average customer journey involves three or more distinct channels, with consumers moving seamlessly between your website, mobile app, email and more, creating a trail of valuable data points as they go. As consumers continue to devote an increasing amount of time and energy to engaging … [Read more...] about What is a Customer Data Platform?

Performance Monitoring of ML ModelsAs Machine Learning infrastructure has matured, the need for model monitoring has surged. Unfortunately this growing demand has not led to a foolproof playbook that explains to teams how to measure their model’s performance.Performance analysis of production models can be complex, and every situation comes with its own set of challenges. … [Read more...] about The Playbook to Monitor Your Model’s Performance in Production

Let’s create a new Python file for speech recognition.We’ll need a library to convert the Japanese characters (hiragana, katakana, and kanji) to Latin/Roman (rōmaji) characters. In my experience, Pykakasi is one of the best choices. If you know a better one, let me know in the comments.pip3 install pykakasiAdd the necessary imports:import requestsimport jsonimport pykakasifrom … [Read more...] about How To Build a Japanese Pronunciation Checker With Python and Wit.ai

How many times have you heard something like this? “AI will take over certain jobs… but it’s okay! New technology will create new jobs! That’s why the kids in school today should be learning concepts, not memorizing. As long as they get concepts, they can search for any information they need once they’re in the workforce.”So, AI will replace the work of lawyers (sharks in … [Read more...] about With AI, be careful what you wish for.

Medium doesn’t support math rendering. Check out my personal blog for proper inline math equations and a clickable table of contents!Bridging the Gap Between Basics and Modern ResearchTable of Contents1. Preamble2. Neural Network Generalization3. Back to Basics: The Bayesian Approach3.1 Frequentists3.2 Bayesianists3.3 Bayesian Inference and Marginalization4. How to Use a … [Read more...] about A Comprehensive Introduction to Bayesian Deep Learning

I chose accuracy as my metric because the majority class was between 50%-70%, at 51.7%.I didn’t need to use any data augmentation techniques since both binary classes are somewhat evenly balanced as seen above in the Target Occurrence visual.After cleaning and pre-processing the dataset and creating the feature matrix and target vector, I built three models:Logistic … [Read more...] about Which CrossFit Athletes Have Competitive Potential for the Sport of Weightlifting?