Last answered:

15 Jun 2020

Posted on:

11 Jun 2020

0

Model Deployment

Hello. I am stuck in some point. How can I save my algorithm for further prediction without the need to retrain or refit it each time I open the jupyter notebook? Is that easy?  This is in fact a general question not related to any specific course. Thank you in advance.
1 answers ( 0 marked as helpful)
Instructor
Posted on:

15 Jun 2020

0
Hi Ahmed, Depends on the model you are trying to save. If you are trying to save a model in sklearn or statsmodels, you will need to use the pickle package. Then you pickle the object in question and you are done! If you are using TensorFlow (or some other library that has this functionality) you can easily save a model with the built-in functions. You can refer to the documentation of TensorFlow here: https://www.tensorflow.org/tutorials/keras/save_and_load Best, Iliya

Submit an answer