I have been working with scikit-learn for some time now and I was wondering if there is a way after training a model to push the pickle-d or joblib output to a database (preferably joblib for reasons mentioned here: What are the different use cases of joblib versus pickle?), or the only way to do so is using a object storage service such as Amazon S3? I am pretty sure about the second but I would like to know if there is another tool that allows saving a model to database.
Even if it is possible to save the result to a database (not sure how that could be done as a joblib result of a saved model consists of five files) would that be efficient to begin with? Why joblib creates more than one file as a result an those contents be concatenated into one file
joblib.dump(svm, '/...temp/svc.pkl')
Sample output after the execution of the dump above (content of one of the files created):
934e 554d 5059 0100 4600 7b27 6465 7363
7227 3a20 273c 6638 272c 2027 666f 7274
7261 6e5f 6f72 6465 7227 3a20 4661 6c73
652c 2027 7368 6170 6527 3a20 2832 2c29
2c20 7d20 2020 2020 2020 2020 2020 200a
0000 0000 0000 f03f 0000 0000 0000 f03f
Thanks!
