def test_crit_functions_dict(self):
res = make_criterion_functions({
"MASELoss": {
"baseline_method": "mean"
},
"MSE": {}
})
self.assertIsInstance(res, list)
def __init__(
self,
model_base: str,
training_data: str,
validation_data: str,
test_data: str,
params: Dict):
self.params = params
if "weight_path" in params:
params["weight_path"] = get_data(params["weight_path"])
self.model = self.load_model(model_base, params["model_params"], params["weight_path"])
else:
self.model = self.load_model(model_base, params["model_params"])
# params["dataset_params"]["forecast_test_len"] = params["inference_params"]["hours_to_forecast"]
self.training = self.make_data_load(training_data, params["dataset_params"], "train")
self.validation = self.make_data_load(validation_data, params["dataset_params"], "valid")
self.test_data = self.make_data_load(test_data, params["dataset_params"], "test")
if "GCS" in self.params and self.params["GCS"]:
self.gcs_client = get_storage_client()
else:
self.gcs_client = None
self.wandb = self.wandb_init()
self.crit = make_criterion_functions(params["metrics"])
def test_crit_functions_list(self):
res = make_criterion_functions(["MSE", "RMSE", "MAPE"])
self.assertIsInstance(res, list)