Bounding Box
A bounding box, also known as a bounding rectangle, is a rectangular box that surrounds an object in an image or video and provides information
Category: Lexicon
Brute Force Search
An exhaustive search that looks across all the given inputs and does not limit itself to clustering or approximations. It’s usually more expensive and time-consuming,
COCO
COCO is a large-scale segmentation, captioning, and object detection dataset. This dataset compiles ninety objects such as sports balls, dogs, cats, horses, persons, cars, etc.
Computer Vision
Computer vision is a branch of Artificial Intelligence used to develop techniques that enable computers to process visual input from JPEG files or camera videos
Convolutional Neural Networks (CNN)
Convolutional Neural Networks, or CNNs, extract information from images with the help of sequential pooling and convolutional layers. This deep-learning network feeds the extracted data
Data Augmentation
To ensure your Machine Learning and Artificial Intelligence projects thrive, you need two key ingredients: Unstructured and Structured Data. Unstructured Data refers to raw, unprocessed
Data Cleaning
When combining data from many sources, there is a high risk of duplication and incorrect labeling. The algorithms could provide wildly different outcomes even when
Data Decomposition
Decomposition is a statistical task that includes dissecting Time Series data into its constituent parts or extracting trends and seasonality from a given set of
Data granularity
Granularity is a term that is hard to pin down due to its several meanings; nonetheless, in software and marketing, it refers to the accuracy
Deep Learning
Deep Learning is a specialized form of Machine Learning and a crucial part of artificial intelligence. It uses artificial neural networks with multiple layers, hence
Eager Execution
Eager execution as an environment evaluates operations immediately, and the operations return values rather than computational graphs to run later. Similarly, TensorFlow calculates tensor values
Early Stopping
Early stopping is a technique that is commonly used to avoid overtraining a model. In practice, model training data is split into a “training set”
