Open arya-hemanshu opened 6 years ago
This needs more discussion as our current architecture states that data needs to go to database and from there backend will pick up the data and then serve it to user.
However when data is small enough that training and predicting wouldn't require much time, the response can be send directly to backend which can relay it to frontend, and simultaneously can be saved in database, which will increase response time
i) Prediction array ii) Timestamp for the predictions iii) Request id iv) Timestamp of the response