Exploring the capabilities of ChatGPT, particularly GPT-4, I exposed a shortcoming regarding the model's ability to remember or store information it has 'thought' about during a dialogue sequence. Probing deeper, I developed an experimental system named ChatGPT Memory to input detailed information into the system like 'dreams', 'goals', 'inner dialogue' and more. While this method doesn't make the AI truly sentient, it definitely pushes the envelope and leads to interesting outputs. Although there are limitations, especially when handling more complex tasks, the enhancements present an exciting prospect for future iterations of the model.
In this enlightening blog post, I dive into the tantalizing world of ChatGPT and Large Language Models. Clarifying its operation, I unlock this enigma by comparing its mechanisms to a simple language model. However, Challenges arise due to the explosion of possible token combinations, leading to an inherent 'lossy' compression of our world's vast information. Surprisingly, even with such compression, these models can mimic human language in a compelling manner. I also investigate possible strategies to optimize this amazing technology - including zero-shot learning, one-shot learning, few-shot learning, and fine-tuning. Entering the era of prompt engineering and larger models, we're...
This post provides a comprehensive guide to building a do-it-yourself (DIY) Alexa using an ESP32 and Wit.ai. It illustrates how to create a wake word detection system, use Python for machine learning and employ TensorFlow for the 'wake' word identification. It also covers the usage of Wit.ai for intent recognition and managing commands. The post is fully backed with code snippets, examples and video tutorials to deliver an interactive learning experience to readers.
In this blog, I venture beyond binary classification and delve into categorical classification using TensorFlow. Specifically, I show how to generate and classify images into four categories: blank, square, circle, and triangle. Crucially, I highlight the use of one-hot encoding for labeling and the softmax activation function in our model. I provide examples of code, results of my model's predictions, and link to the full code in my GitHub repository. Feel free to try it out and explore this exciting domain of machine learning further.
In this playful tutorial for binary classification aided by visuals, we create a Python generator that generates alternating images of squares and circles, which we then classify using TensorFlow. We take special care to point out, too, that to get binary classification up and running, we must use a particular activation function and loss function. After training our model, illustrations show that it's excellent at identifying images as either squares or circles. So, if you're into Machine Learning and has a sense of humor, don’t forget to check out the 'hot dog or not hot dog' bit and visit the...
This blog showcases the progress of my idea to create an augmented reality Sudoku solver using technology that enables us to solve puzzles in our browser rather than with dedicated apps. I have developed an AR Sudoku solver with a simple image processing pipeline. It identifies and extracts Sudoku puzzles from pictures, recognizes each cell's numbers, solves the puzzle, and renders the solution on top of the original image. This process is accomplished by converting the image to greyscale, conducting thresholding, OCR-processing, and puzzle-solving. I've also done parallel image location and extraction. It's a technical journey that's achieved some high-quality,...
In this technical tutorial, we walk through the process of wiring up the iPhone's camera to CoreML using Vision Kit, allowing us to run machine learning models against the camera input. We outline the necessary steps in creating a new Xcode project; capturing video frames; using AVFoundation, Vision, and CoreML; and dissecting the video frames using Vision magic. We also illustrate the process of running our Vision requests and displaying the expected outputs.