At first, we were unable to fit the ai model, despite changing the initial dataset and modifying the code. Google Collab still displayed the error message that it could not convert a NumPy to a Tensor. We realised later that we included data that were not numbers in the x_train variable (that contains information like the destination, the scheduled flight time, the flight distance, etc). However, we needed the IATA code (the letter code representing airports). In order to fix this issue, we assigned a number for each IATA code in the original dataset. Furthermore, we could not implement the Google Collab code into an HTML website.
Log in or sign up for Devpost to join the conversation.