1
$\begingroup$

I have trained an ML model with a good accuracy but what next?

  1. I am facing difficulty in answering this question, how will you present your model.
  2. Which framework do you use
  3. How do you make sure the model is learning continuously.
  4. Where can I find definite guide for this approach?

I am good at building and predicting ML, DL & NLP. But stuck to deploy further.

$\endgroup$
0
$\begingroup$

If one understands the business problem you are trying to solve, then this shouldn't be that hard.

  1. You can present the results of your model. Understand your data first, is it skewed, balanced? Understand your model, check the feature importance. Do they make sense with the domain knowledge you have? For example, if your data is highly imbalanced and you have accuracy of 0.99, then it's highly likely that your model is classifying almost everything as one class. Try and make sense of the model, may be try running some shallow models to establish a baseline.
  2. There's no free lunch. You can't have frameworks for interpreting results, you can have a good set of metrics that you can use but metrics change based on how your data is.
  3. Model is learning continuously?? You mean re-training the model. You need to establish a pipeline for it. You can have a python project for this or use DAG mechanisms like AirFlow.
  4. There's no definite guide, a few text books might help but this comes with experience and understanding the concepts of machine learning thouroughly.
| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.