I used to come across this word called "Inference" frequently, whenever I read an article/blog about Machine learning. So I thought it could be helpful if I put it in words.

Basically, any machine learning app development involves two different phases.

1. Model Training: This part involves preparation of your ML model, annotating (labelling) your inputs, training your model and validating it to make it better.

2. Inference: Once your training model is ready, you will use it in real time. i.e. Inference Phase. So now your ML inference accepts your input and processes it with the help of your model to give predicted output.