A Quick Start to AllenNLP for Natural Language Processing If you are using AllenNLP for the first time, looking at the quick start guide can be overwhelming and there isn’t a guide for the multitude of NLP tasks out there. There are different tools out there to train a Natural Language Processing (NLP) model, with the two main ones being HuggingFace and AllenNLP. In this quick guide I will be covering how to start on an AllenNLP project using a Constituency Parser. At the point of writing this guide, AllenNLP is at version 2.9.3 [Read More]

Introduction Many companies around the world use SAP for various tasks, such as quality management, laboratory sample data, preventive maintenance scheduling, supply chain scheduling. However most companies do not give access to SAP Advanced Business Application Programming (ABAP) environment (Transaction code: SE38). Most users use SAP on a daily basis repeating the same tasks over and over again, such as data extraction and updating information in SAP which is repetitive and time consuming. [Read More]

What is a Machine Learning Pipeline? A Machine Learning Pipeline is a way to automate the workflow it takes to produce a machine learning model. It consists a sequential steps chained together that to perform various tasks such as data ingestion, data cleaning, feature engineering, model training and deployment. [Read More]

Predicting Stock Prices using Machine learning The world of finance contains LOTS of data. This makes finance a very suitable subject to be used in Machine Learning. In this case I’ve decided to investigate further if it is possible to predict the stock market data as the data is readily available. In this case I’ve selected a couple of stocks / Exchange Traded Funds (ETF) to find out if the selected Machine Learning Models are suitable for the broader based index (which tends to be more stable) or for stocks (higher volatility). [Read More]

What is Web Scraping? Web scraping involves the action of fetching data from a web page and extracting it. This can be done by extracting information from the html elements of the webpage. The information extracted can then be formatted, searched and subsequently parsed into a database or spreadsheet for further analysis. Although this can be done manually, scraping generally involves the use of a script or bot to automate the process over hundreds to thousands of webpages. [Read More]