Image Courtesy: Google Cloud
A new Google AI is able to identify baking recipes. This project is headed by Sara Robinson, who works on AI for Google Cloud. Sara holds a keen interest in baking various items.
In this interesting innovation, the model has been programmed in such a way that it makes additions to the recipes and creates one of its own.
How it Functions?
Sara and Dale’s joint effort to create this Google AI has been a joyride for them. Robinson compiled a list of recipes and formed a TensorFlow model which accepts the list of ingredients resulting into a flow of predictions.
For eg. “97% bread, 2%cake, 1%cookie”.
This new recipe has been named ‘Cakie’ and another hybrid recipe is now called ‘Breakie’.
The most important aspect of developing this model is that it involves no scientific process or much complication of that kind.
Dale and Sara expressed, “For this project, we decided to use a Google Cloud tool called AutoML Tables. It’s a no-code way to build machine learning models on tabular data, like the kind you’d find in a spreadsheet or database. We chose AutoML Tables because it’s both easy to use and just got an upgrade of new, built-in explainability tools like Feature Attribution (more on that in a bit).”
Around the month of June, in 2020, at Google’s AI Principles’ first anniversary, its users were made aware of the ideology behind its upcoming models i.e. to be “respectful, social and beneficial”.
However, AI powered models come with some limitations. Therefore, it is expected that the baking recipe prediction model may not give out miniscule details, leaving users to rely somewhat on their own creativity too.