Introducing model deployment
When you start a DL project, your primary focus should be on data collection, data processing, and model development. The process of model development from data collection to model training always happens in an offline setting. However, wouldn’t it be nice if the awesome model that you worked so hard on turned into something that other people can interact with? Even great models won’t have much impact if they remain in the notebooks. Also, you wouldn’t run your notebook every time new data comes in, right? So, how do you do it? The simple answer is model deployment, which is the process of integrating the model into an existing production environment to make appropriate business decisions based on data. It is the second-to-last stage of the DL life cycle before model monitoring and is the most important step for solving business challenges. A model that’s deployed in production has several advantages:
- It will serve the...
