A consilience of interdisciplinary scripts in finance, data science, and mathematics that attempt to harness the power of computational neural networks and other machine learning tools.
This repository aims to serve as a centralized space for the following 3 main concepts subject to semantic analysis implemented by a series of tools. These tools may vary from simple statistics based methods to more nuanced physics models that aim to incorporate into modern artificial intelligence algorithms. Examples of these are evolutionary computation, adversarial neural networks and finally, using ChatGPT to help out in making sense of all of this.
The main directory structure is as follows:
To relieve websocket backpressure in this project Javascript is recommended to regulate API byte flow, Newtonsoft.Json is a great NuGet package for this. Anaconda DataSpell is the recommended program to run the files in this repository.
Financial:
The above folders are initial starting points, other subdirectories that will most likely emerge will be input/output paths.
The scripts in the notebooks were written mostly by ChatGPT through asking it a series of prompts in formats such as "Write me a python script that...[functions]". Beyond this dozens of debugging runs also improved the code by asking ChatGPT "The following Exception was returned: ['Enter exception', traceback line:'Enter traceback line' ]. Here is the code: 'feed code as prompt again'". Essentially this is a form of regressive analysis where the script is being called on itself and solved for the variating evolving bugs. A lot of these scripts need more debugging and will be updated as time permits.