Eng 467 ai xplain sdk update finetune functional tests to cover all new finetunable models

[xainaz]

Added createdAt attribute to model class, and implemented necessary changes in Model Factory.
As for the finetune functional test, I replaced the changes with the original code that ran the test on the 3 models in data, since there were no new models in the past 3 weeks.
2 models are passing ( llama2 7b, vectara) and 1 is failing : air

[thiago-aixplain]

Where are the functional tests?

[thiago-aixplain]

Please check whether this to_dict method works with a datetime attribute.

[xainaz]

It seems to work, this is what it returns:
{'id': '64e615671567f848804985e1', 'name': 'GPT2', 'supplier': <Supplier.OPENAI: {'id': 1777, 'name': 'OpenAI', 'code': 'openai'}>, 'additional_info': {'parameters': {'language': ['en']}}, 'createdAt': datetime.datetime(2023, 8, 30, 10, 53, 10, 827000, tzinfo=datetime.timezone.utc)}
I added createdAt to to_dict

[thiago-aixplain]

You should not test all the models within the same method. You should call test_end2end for each model in finetunable_llms_from_last_4_weeks. Check how to use a fixture together with the python command yield. There is an example here: https://github.com/aixplain/aiXplain/blob/main/tests/functional/pipelines/designer_test.py#L17