def test_raise_exception(sequential_model):
with TempDir(chdr=True, remove_on_exit=True) as tmp:
path = tmp.path("model")
with pytest.raises(IOError, match="No such file or directory"):
mlflow.pytorch.load_model(path)
with pytest.raises(
TypeError,
match="Argument 'pytorch_model' should be a torch.nn.Module"):
mlflow.pytorch.save_model([1, 2, 3], path)
mlflow.pytorch.save_model(sequential_model, path)
with pytest.raises(
RuntimeError,
match=f"Path '{os.path.abspath(path)}' already exists"):
mlflow.pytorch.save_model(sequential_model, path)
from mlflow import sklearn
import sklearn.neighbors as knn
path = tmp.path("knn.pkl")
knn = knn.KNeighborsClassifier()
with open(path, "wb") as f:
pickle.dump(knn, f)
path = tmp.path("knn")
sklearn.save_model(knn, path=path)
with pytest.raises(MlflowException,
match='Model does not have the "pytorch" flavor'):
mlflow.pytorch.load_model(path)
def sv_model(params, model):
mlflow.log_params(params)
mlflow.set_tag("running_from_jupyter", "True")
#mlflow.log_artifact("./models")
#mlflow.sklearn.log_model(model, "model")
path = f"models/{params['model']+params['label']}_{datetime.now().strftime('%Y-%m-%d_%H%M%S')}"
save_model(sk_model=model, path=path)
def _save_model(self):
"""Save model in `./models` as `<model_type>_<label>_<time>`
"""
if self.model:
time_now = datetime.now().strftime('%Y-%m-%d_%H%M%S')
model_type = self.params["model"]
label = self.params["label"]
path = self.model_path / f"{model_type}_{label}_{time_now}"
save_model(sk_model=self.model, path=path)
logger.info(f"Saved model to {path}.")
def test_model_save_load(self):
with TempDir(chdr=True, remove_on_exit=True) as tmp:
model_path = tmp.path("knn.pkl")
with open(model_path, "wb") as f:
pickle.dump(self._knn, f)
path = tmp.path("knn")
sklearn.save_model(self._knn, path=path)
x = sklearn.load_model(path)
xpred = x.predict(self._X)
np.testing.assert_array_equal(self._knn_predict, xpred)
# sklearn should also be stored as a valid pyfunc model
# test pyfunc compatibility
y = pyfunc.load_pyfunc(path)
ypred = y.predict(self._X)
np.testing.assert_array_equal(self._knn_predict, ypred)
def test_raise_exception(sequential_model):
with TempDir(chdr=True, remove_on_exit=True) as tmp:
path = tmp.path("model")
with pytest.raises(IOError):
mlflow.pytorch.load_model(path)
with pytest.raises(TypeError):
mlflow.pytorch.save_model([1, 2, 3], path)
mlflow.pytorch.save_model(sequential_model, path)
with pytest.raises(RuntimeError):
mlflow.pytorch.save_model(sequential_model, path)
from mlflow import sklearn
import sklearn.neighbors as knn
path = tmp.path("knn.pkl")
knn = knn.KNeighborsClassifier()
with open(path, "wb") as f:
pickle.dump(knn, f)
path = tmp.path("knn")
sklearn.save_model(knn, path=path)
with pytest.raises(MlflowException):
mlflow.pytorch.load_model(path)
'classification',
'classification',
'classification',
'classification',
]
})
helper_features = ['ew_'+str(x) for x in range(2,14)] + \
['lw_distinct_series', 'lw_distinct_episodes'] + \
['genre_share_drama', 'genre_share_comedy', 'genre_share_sport', 'genre_share_music']
# Create the missing value imputer to learn sensible substitute values
mvi = missing_values.missing_value_imputer(impute_strategies=impute_strategies,
helper_features=helper_features)
sklearn.save_model(mvi, f"missing_value_imputer-{date.today()}")
sklearn.log_model(mvi, f"missing_value_imputer-{date.today()}")
# Train the imputer on df (the training set)
df_impute = mvi.train(df)
# Replacing missing value columns in fresh data with imputed columns
non_imputed = [
c for c in df.columns if c not in mvi.impute_strategies.colname.values
]
df = pd.concat([df[non_imputed], df_impute], axis=1)
# Pickle the imputer for use with model scoring
utils.pickler(mvi, pickle_dir + '/prep/missing_value_imputer')
print("Missing value imputation completed\n")
