What is ChatGPT Code Interpreter? Let me Explain:
ChatGPT is no longer just a clever language model. OpenAI is testing the ability to run code and use third-party plugins within ChatGPT, and the results are nothing short of magical. The Code Interpreter (CI) plugin can generate code, run code, upload and download files ranging from CSV data to images and evaluate output all within the ChatGPT interface.
A Plugin that Replaces Python
At present, the CI runs Python with a small selection of libraries, but even with basic libraries it can do some amazing things. For instance, it can analyze output and input it into another function, allowing you to string together different sections of code, taking the output of one and feeding it to another. The Pac-Man gif test showed how it uses an algorithm to generate a maze, converts it into blocks, finds the exit, makes it look like Pac-Man and generates a gif.
Previously, using ChatGPT to create code meant taking the output and testing it in another environment. With the CI plugin, a lot of development can be done within ChatGPT without leaving the UI. In addition, it can analyze data such as spreadsheets and perform functions like graphing, making it a very powerful tool for both coding and research.
Examples of ChatGPT Code Interpreter
Some interesting experiments have been run with the ChatGPT Code Interpreter plugin. For example, tests have been done to show ChatGPT’s sound capabilities by generating a Shepard tone, a sound illusion that seems like it’s constantly rising in pitch. ChatGPT was also asked to generate a session of Conway’s Game of Life and convert it to music. For an entire album of musical math, ChatGPT generated an album cover for the Game of Life song. While further tests generated a functioning QR code image, a game of life simulation that resulted in a QR code and created an ASCII image from an uploaded image.
It was also possible to generate a bouncing block, a snowstorm, a session of Game of Life, a rotating cube and even a simulation of the orbits of the inner planets. The Matplotlib library helped ChatGPT Code Interpreter simulate a game of checkers as well as the start of a chess game.
What Else?
With OCR as one of the libraries in CI, tests were done to see if OCR was capable of recognizing text in images including screen grabs. While it wasn’t possible to load any machine learning libraries into CI, basic statistical text generation was possible using n-grams. By giving Bot Andrew’s book, it could create a prediction algorithm and then predict the next word in a text sequence.
As more people get access to ChatGPT Code Interpreter, we’re looking forward to seeing mind-blowing applications that haven’t been previously conceived. For now, the ChatGPT Code Interpreter plugin provides a multitude of possibilities.
