Hi, I am currently working on HSASynthesizer trying to fit my data. I modified the one of the sdtypes of my column to ‘phone_number’ and when I fit the data it gives me this message error :
ValueError: Phone number (638)617-7185 is represented in national format. Please provide ``default_country`` for nationally represented numbers when creating the transformer instance.
The ‘phone_numbers’ were created from DayZSynthesizer. When I create data using only French phones numbers it gives me the same thing. I feel like that the synthesizer doesn’t recognize the national format so I may have to fiddle around with python to change the number to “+…” since there is no help on this topic in the documentation.
SDV Synthesizers (like HSASynthesizer) currently expect you to pass in internationally formatted phone number values with the international code pre-pended. Some examples:
For an American number: +1 972 369 4678
For a French number: +33 5 20 21 29 13
It looks like DayZSynthesizer isn’t generating consistently formatted French phone numbers right now. We use the Faker library under the hood and I replicated the issue down to the Faker level. This is a limitation we have right now that we can hopefully overcome in the future.
I can create a workaround script if you’d like to generate French phone numbers directly and then use with HSASynthesizer. If you have a lot of different countries’ phone numbers to generate, then of course that will be less useful!
So let me know and I can try my best to unblock you.
Hi Srini!
Thanks for all the explanations. It’s fine I will make a little script to work around it
I understood that you guys used the Faker librairy and don’t worry about that. As I said I trying to push the limit of the tools
Have a great day!
