UI + AI = documentation, or the story of an experiment
-
40 min
The report is dedicated to the development of a neural network model for generating technical instructions based on design layouts. The research examines three models — ChatGPT, YandexGPT, and Llama — which were fine-tuned using specially prepared datasets and prompts.
Furthermore, the report showcases the results of the fine-tuning process and conducts a detailed analysis of the quality of the generated technical information. This analysis focuses on assessing the accuracy and correctness of the created documents, which is crucial for the practical application of the developed approach.
In
conclusion, the report provides insights into the quality and effectiveness of
using the neural network model for automating the creation of technical
documentation for UI layouts. It also explores potential prospects for its
further development and application in real-world scenarios.