def setUp(self):
self.make_bucket()
buffer = io.BytesIO(b"mock")
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name="artifacts/mock.txt",
data=buffer,
length=buffer.getbuffer().nbytes,
)
def create_mock_figure(self):
file = BytesIO(
b'<svg viewBox=\'0 0 125 80\' xmlns=\'http://www.w3.org/2000/svg\'>\n'
b'<text y="75" font-size="100" font-family="serif"><![CDATA[10]]></text>\n'
b'</svg>\n')
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name="experiments/test/operators/test/figure-123456.svg",
data=file,
length=file.getbuffer().nbytes,
)
def create_mock_model(self):
model = {"model": MockModel()}
buffer = BytesIO()
dump(model, buffer)
buffer.seek(0, SEEK_SET)
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=
f"experiments/{EXPERIMENT_ID}/operators/{OPERATOR_ID}/model.joblib",
data=buffer,
length=buffer.getbuffer().nbytes,
)
def load_model(experiment_id: Optional[str] = None,
operator_id: Optional[str] = None) -> Dict[str, object]:
"""Retrieves a model from object storage.
Args:
experiment_id (str, optional): the experiment uuid. Defaults to None.
operator_id (str, optional): the operator uuid. Defaults to None.
Returns:
dict: A dictionary of models.
Raises:
TypeError: when experiment_id is undefined in args and env.
TypeError: when operator_id is undefined in args and env.
"""
if experiment_id is None:
experiment_id = get_experiment_id()
if operator_id is None:
operator_id = get_operator_id()
try:
object_name = f"{PREFIX_1}/{experiment_id}/{PREFIX_2}/{operator_id}/{MODEL_FILE}"
data = MINIO_CLIENT.get_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
)
except (NoSuchBucket, NoSuchKey):
return {}
buffer = BytesIO(data.read())
return load(buffer)
def download_artifact(name: str, path: str):
"""Downloads the given artifact to the path.
Args:
name (str): the dataset name.
path (str): destination path.
Raises:
FileNotFoundError
"""
try:
MINIO_CLIENT.fget_object(
bucket_name=BUCKET_NAME,
object_name=f"{PREFIX}/{name}",
file_path=path,
)
except (NoSuchBucket, NoSuchKey):
raise FileNotFoundError("The specified artifact does not exist")
def create_mock_dataset3(self):
with open("mock.jpg", 'wb') as imagef:
imagef.write(MOCK_IMAGE)
MINIO_CLIENT.fput_object(
bucket_name=BUCKET_NAME,
object_name="datasets/mock.jpg/mock.jpg",
file_path="mock.jpg",
)
metadata = {
"filename": "mock.jpg",
}
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name="datasets/mock.jpg/mock.jpg.metadata",
data=buffer,
length=buffer.getbuffer().nbytes,
)
def create_mock_dataset2(self):
with ZipFile("mock.zip", "w") as zipf:
zipf.writestr("mock.gif", MOCK_IMAGE)
MINIO_CLIENT.fput_object(
bucket_name=BUCKET_NAME,
object_name="datasets/mock.zip/mock.zip",
file_path="mock.zip",
)
metadata = {
"filename": "mock.zip",
}
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name="datasets/mock.zip/mock.zip.metadata",
data=buffer,
length=buffer.getbuffer().nbytes,
)
def update_dataset_metadata(name: str,
metadata: Dict[str, str],
run_id: Optional[str] = None,
operator_id: Optional[str] = None):
"""Update the metadata of a dataset.
Args:
name (str): the dataset name.
metadata (dict): metadata about the dataset.
run_id (str, optional): the run id of trainning pipeline. Defaults to None.
operator_id (str, optional): the operator uuid. Defaults to None.
"""
object_name = _metadata_filepath(name, run_id, operator_id)
# encodes metadata to JSON format
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
data=buffer,
length=buffer.getbuffer().nbytes,
)
def save_model(**kwargs):
"""Serializes and saves models.
Args:
**kwargs: the models as keyword arguments.
Raises:
TypeError: when a figure is not a matplotlib figure.
Raises:
TypeError: when experiment_id is undefined in args and env.
TypeError: when operator_id is undefined in args and env.
"""
experiment_id = kwargs.get("experiment_id")
if experiment_id is None:
experiment_id = get_experiment_id()
operator_id = kwargs.get("operator_id")
if operator_id is None:
operator_id = get_operator_id()
object_name = f"{PREFIX_1}/{experiment_id}/{PREFIX_2}/{operator_id}/{MODEL_FILE}"
model_buffer = BytesIO()
dump(kwargs, model_buffer)
model_buffer.seek(0, SEEK_SET)
# ensures MinIO bucket exists
make_bucket(BUCKET_NAME)
# uploads file to MinIO
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
data=model_buffer,
length=model_buffer.getbuffer().nbytes,
)
def list_datasets() -> List[str]:
"""Lists dataset names from object storage.
Returns:
list: A list of all datasets names.
"""
datasets = []
# ensures MinIO bucket exists
make_bucket(BUCKET_NAME)
objects = MINIO_CLIENT.list_objects_v2(BUCKET_NAME, PREFIX + "/")
for obj in objects:
name = obj.object_name[len(PREFIX) + 1:-1]
datasets.append(name)
return datasets
def list_metrics(experiment_id: Optional[str] = None,
operator_id: Optional[str] = None,
run_id: Optional[str] = None) -> List[Dict[str, object]]:
"""Lists metrics from object storage.
Args:
experiment_id (str, optional): the experiment uuid. Defaults to None.
operator_id (str, optional): the operator uuid. Defaults to None.
run_id (str, optional): the run id. Defaults to None.
Returns:
list: A list of metrics.
Raises:
TypeError: when experiment_id is undefined in args and env.
TypeError: when operator_id is undefined in args and env.
"""
if experiment_id is None:
experiment_id = get_experiment_id()
if operator_id is None:
operator_id = get_operator_id()
# ensures MinIO bucket exists
make_bucket(BUCKET_NAME)
if run_id is None:
# gets run_id from env variable
# Attention: returns None if env is unset
run_id = get_run_id()
elif run_id == "latest":
try:
metadata = stat_metadata(experiment_id, operator_id)
run_id = metadata.get("run_id")
except FileNotFoundError:
return []
try:
object_name = operator_filepath(METRICS_FILE, experiment_id, operator_id, run_id)
data = MINIO_CLIENT.get_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
)
except (NoSuchBucket, NoSuchKey):
raise FileNotFoundError(f"No such file or directory: '{experiment_id}'")
return load(data)
def create_mock_dataset1(self, size=1e2):
header = ",".join(self.mock_columns()) + "\n"
rows = "\n".join([
",".join([str(v) for v in self.mock_values()])
for x in range(int(size))
])
buffer = BytesIO((header + rows).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name="datasets/mock.csv/mock.csv",
data=buffer,
length=buffer.getbuffer().nbytes,
)
metadata = {
"columns": self.mock_columns(),
"featuretypes": self.mock_featuretypes(),
"filename": "mock.csv",
"run_id": RUN_ID,
}
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name="datasets/mock.csv/mock.csv.metadata",
data=buffer,
length=buffer.getbuffer().nbytes,
)
MINIO_CLIENT.copy_object(
bucket_name=BUCKET_NAME,
object_name=
f"datasets/mock.csv/runs/{RUN_ID}/operators/{OPERATOR_ID}/mock.csv/mock.csv",
object_source=f"/{BUCKET_NAME}/datasets/mock.csv/mock.csv",
)
MINIO_CLIENT.copy_object(
bucket_name=BUCKET_NAME,
object_name=
f"datasets/mock.csv/runs/{RUN_ID}/operators/{OPERATOR_ID}/mock.csv/mock.csv.metadata",
object_source=f"/{BUCKET_NAME}/datasets/mock.csv/mock.csv.metadata",
)
def make_bucket(self):
try:
MINIO_CLIENT.make_bucket(BUCKET_NAME)
except BucketAlreadyOwnedByYou:
pass
def empty_bucket(self):
for obj in MINIO_CLIENT.list_objects(BUCKET_NAME,
prefix="",
recursive=True):
MINIO_CLIENT.remove_object(BUCKET_NAME, obj.object_name)
def stat_dataset(name: str,
run_id: Optional[str] = None,
operator_id: Optional[str] = None) -> Dict[str, str]:
"""Retrieves the metadata of a dataset.
Args:
name (str): the dataset name.
run_id (str, optional): the run id of trainning pipeline. Defaults to None.
operator_id (str, optional): the operator uuid. Defaults to None.
Returns:
dict: The metadata.
Raises:
FileNotFoundError: If dataset does not exist in the object storage.
"""
metadata = {}
# remove /tmp/data/ from dataset name
# because in jupyter we use dataset with full path
name = name.replace("/tmp/data/", "")
if run_id == "latest":
metadata = stat_dataset(name)
run_id = metadata.get("run_id")
if run_id is None:
# gets run_id from env variables
# Attention: returns None if env is unset
run_id = get_run_id()
if run_id and operator_id:
# get metadata for a specific operator of a run, if exists
object_name = _metadata_filepath(name, run_id, operator_id)
elif run_id:
# if no metadata was generated by the operator,
# get the last one generated by the pipeline flow
object_name = _metadata_filepath(name, run_id)
if not metadata_exists(name, run_id):
# if it is at the beginning of a run,
# there will be no metadata generated by run_id
object_name = _metadata_filepath(name)
else:
# unable to get run_id automatically,
# this function is probably being called out of a run
object_name = _metadata_filepath(name)
else:
# get path according to received parameters
run_id = None if run_id == "root" else run_id
object_name = _metadata_filepath(name, run_id, operator_id)
try:
# reads the .metadata file
data = MINIO_CLIENT.get_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
)
# decodes the metadata (which is in JSON format)
metadata = loads(data.read())
except (NoSuchBucket, NoSuchKey):
raise FileNotFoundError("The specified dataset does not exist")
return metadata
def tearDown(self):
MINIO_CLIENT.remove_object(
bucket_name=BUCKET_NAME,
object_name="artifacts/mock.txt",
)
def save_metrics(experiment_id: Optional[str] = None,
operator_id: Optional[str] = None,
run_id: Optional[str] = None,
**kwargs):
"""Saves metrics of an experiment to the object storage.
Args:
experiment_id (str, optional): the experiment uuid. Defaults to None
operator_id (str, optional): the operator uuid. Defaults to None
run_id (str, optional): the run id. Defaults to None.
**kwargs: the metrics dict.
Raises:
TypeError: when experiment_id is undefined in args and env.
TypeError: when operator_id is undefined in args and env.
"""
if experiment_id is None:
experiment_id = get_experiment_id()
if operator_id is None:
operator_id = get_operator_id()
if run_id is None:
# gets run_id from env variables
# Attention: returns None if env is unset
run_id = get_run_id()
# ensures MinIO bucket exists
make_bucket(BUCKET_NAME)
if run_id:
metadata = {}
try:
metadata = stat_metadata(experiment_id, operator_id)
if run_id == "latest":
run_id = metadata.get("run_id")
except FileNotFoundError:
pass
metadata["run_id"] = run_id
# encodes metadata to JSON format and uploads to MinIO
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=f'experiments/{experiment_id}/operators/{operator_id}/.metadata',
data=buffer,
length=buffer.getbuffer().nbytes,
)
object_name = operator_filepath(METRICS_FILE, experiment_id, operator_id, run_id)
encoded_metrics = []
# retrieves the metrics saved previosuly
try:
data = MINIO_CLIENT.get_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
)
encoded_metrics = loads(data.read())
except NoSuchKey:
pass
# appends new metrics
encoded_metrics.extend(_encode_metrics(kwargs))
# puts metrics into buffer
buffer = BytesIO(dumps(encoded_metrics).encode())
# uploads metrics to MinIO
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
data=buffer,
length=buffer.getbuffer().nbytes,
)
def save_dataset(name: str,
data: Union[pd.DataFrame, BinaryIO] = None,
df: pd.DataFrame = None,
metadata: Optional[Dict[str, str]] = None,
run_id: Optional[str] = None,
operator_id: Optional[str] = None):
"""Saves a dataset and its metadata to the object storage.
Args:
name (str): the dataset name.
data (pandas.DataFrame, BinaryIO, optional): the dataset contents as a
pandas.DataFrame or an `BinaryIO` buffer. Defaults to None.
df (pandas.DataFrame, optional): the dataset contents as an `pandas.DataFrame`.
df exists only for compatibility with existing components.
Use "data" for all types of datasets. Defaults to None.
metadata (dict, optional): metadata about the dataset. Defaults to None..
run_id (str, optional): the run id. Defaults to None.
operator_id (str, optional): the operator uuid. Defaults to None.
Raises:
PermissionError: If dataset was read only.
"""
# ensures MinIO bucket exists
make_bucket(BUCKET_NAME)
if run_id is None:
# gets run_id from env variables
# Attention: returns None if env is unset
run_id = get_run_id()
if operator_id is None:
# gets operator_id from env variables
# Attention: returns None if env is unset
operator_id = get_operator_id(raise_for_none=False)
# df exists only for compatibility with existing components
# from now on one must use "data" for all types of datasets
if df is not None:
data = df
try:
# gets metadata (if dataset exists)
stored_metadata = stat_dataset(name, run_id)
metadata_should_be_updated = False
# update stored metadata values
if metadata:
stored_metadata.update(metadata)
elif isinstance(data, pd.DataFrame):
metadata_should_be_updated = True
metadata = stored_metadata
except FileNotFoundError:
metadata_should_be_updated = False
# builds metadata dict:
# sets filename and run_id
if metadata is None:
metadata = {}
metadata["filename"] = name
if isinstance(data, pd.DataFrame):
# sets metadata specific for pandas.DataFrame:
# columns, featuretypes
metadata["columns"] = data.columns.tolist()
metadata["total"] = len(data.index)
if "featuretypes" not in metadata:
metadata["featuretypes"] = infer_featuretypes(data)
# if the metadata was given (set manually), ignore updates, otherwise
# search for changes and then update current featuretypes to be even with columns
if metadata_should_be_updated:
previous_metadata = stat_dataset(name, run_id)
previous_columns = previous_metadata["columns"]
previous_featuretypes = previous_metadata["featuretypes"]
column_to_type = dict(zip(previous_columns, previous_featuretypes))
new_featuretypes = []
for new_column in metadata["columns"]:
if new_column in column_to_type:
new_featuretypes.append(column_to_type[new_column])
else:
new_featuretypes.append(
infer_featuretypes(pd.DataFrame(data[new_column]))[0])
metadata["featuretypes"] = new_featuretypes
if run_id:
metadata["run_id"] = run_id
# When saving a dataset of a run, also
# set the run_id in datasets/<name>.metadata
# This enables load_dataset by run="latest"
try:
root_metadata = stat_dataset(name, "root")
except FileNotFoundError:
root_metadata = {}
root_metadata["run_id"] = run_id
object_name = _metadata_filepath(name)
# encodes metadata to JSON format
buffer = BytesIO(dumps(root_metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
data=buffer,
length=buffer.getbuffer().nbytes,
)
# create a run metadata to save the last operator id
# to dataset get loaded on next step of the pipeline flow
metadata["operator_id"] = operator_id
object_name = _metadata_filepath(name, run_id=run_id)
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
data=buffer,
length=buffer.getbuffer().nbytes,
)
path = _data_filepath(name, run_id, operator_id)
if isinstance(data, pd.DataFrame):
# uploads dataframe to MinIO as a .csv file
temp_file = tempfile.NamedTemporaryFile(dir='.', delete=False)
data.to_csv(temp_file.name, header=True, index=False)
MINIO_CLIENT.fput_object(bucket_name=BUCKET_NAME,
object_name=path.lstrip(f"{BUCKET_NAME}/"),
file_path=temp_file.name)
temp_file.close()
os.remove(temp_file.name)
else:
# uploads raw data to MinIO
buffer = BytesIO(data.read())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=path.lstrip(f"{BUCKET_NAME}/"),
data=buffer,
length=buffer.getbuffer().nbytes,
)
object_name = _metadata_filepath(name, run_id, operator_id)
# encodes metadata to JSON format
buffer = BytesIO(dumps(metadata).encode())
MINIO_CLIENT.put_object(
bucket_name=BUCKET_NAME,
object_name=object_name,
data=buffer,
length=buffer.getbuffer().nbytes,
)
def load_dataset(
name: str,
run_id: Optional[str] = None,
operator_id: Optional[str] = None,
page: Optional[int] = None,
page_size: Optional[int] = None) -> Union[pd.DataFrame, BinaryIO]:
"""Retrieves the contents of a dataset.
If run_id exists, then loads the dataset from the specified run.
If the dataset does not exist for given run_id/operator_id return the
'original' dataset
Args:
name (str): the dataset name.
run_id (str, optional): the run id of training pipeline. Defaults to None.
operator_id (str, optional): the operator uuid. Defaults to None.
Returns:
The contents of a dataset. Either a `pandas.DataFrame` or an `BinaryIO` buffer.
Raises:
FileNotFoundError: If dataset does not exist in the object storage.
"""
if run_id is None:
# gets run_id from env variable
# Attention: returns None if env is unset
run_id = get_run_id()
elif run_id == "latest":
metadata = stat_dataset(name)
run_id = metadata.get("run_id")
# when the dataset does not exist for given run_id/operator_id
# must return the 'original' dataset
# unset run_id so data_filepath points to the 'original' dataset
if run_id and operator_id:
try:
metadata = stat_dataset(name, run_id, operator_id)
except FileNotFoundError:
run_id = None
elif run_id:
try:
run_metadata = stat_dataset(name, run_id)
operator_id = run_metadata.get("operator_id")
except FileNotFoundError:
run_id = None
# builds the path to the dataset file
path = _data_filepath(name, run_id, operator_id)
if page_size and page_size > 0:
nrows = page_size
else:
nrows = None
if page and page > 0:
skiprows = (page - 1) * page_size
else:
skiprows = None
try:
metadata = stat_dataset(name, run_id, operator_id)
dataset = pd.read_csv(S3FS.open(path),
header=0,
index_col=False,
nrows=nrows,
skiprows=skiprows)
dtypes = dict((column, "object") for column, ftype in zip(
metadata["columns"], metadata["featuretypes"])
if ftype in [CATEGORICAL, DATETIME])
dataset = dataset.astype(dtypes)
except (UnicodeDecodeError, pd.errors.EmptyDataError,
pd.errors.ParserError):
# reads the raw file
data = MINIO_CLIENT.get_object(
bucket_name=BUCKET_NAME,
object_name=path.lstrip(f"{BUCKET_NAME}/"),
)
return BytesIO(data.read())
except KeyError:
# metadata file does not contains "columns" or "featuretypes"
# ignore this error and return dataset without cast its type
pass
except FileNotFoundError:
raise FileNotFoundError("The specified dataset does not exist")
return dataset
