def test_decode(content_type):
decoder = Mock()
with patch.dict(encoder._dmatrix_decoders_map, {content_type: decoder},
clear=True):
encoder.decode(42, content_type)
decoder.assert_called_once_with(42)
def input_fn(input_data, content_type):
if content_type == content_types.JSON:
obj = json.loads(input_data)
features = obj['instances'][0]['features']
array = np.array(features).reshape((1, -1))
return xgb.DMatrix(array)
else:
return xgb_encoders.decode(input_data, content_type)
def input_fn(input_data, content_type):
if content_type == content_types.JSON:
print("Recieved content type is json")
print("input_data is", input_data)
obj = json.loads(input_data)
print("obj", obj)
array = np.array(obj)
return xgb.DMatrix(array)
else:
print("content type is not json")
return xgb_encoders.decode(input_data, content_type)
def default_input_fn(input_data, content_type):
"""Take request data and de-serializes the data into an object for prediction.
When an InvokeEndpoint operation is made against an Endpoint running SageMaker model server,
the model server receives two pieces of information:
- The request Content-Type, for example "application/json"
- The request data, which is at most 5 MB (5 * 1024 * 1024 bytes) in size.
The input_fn is responsible to take the request data and pre-process it before prediction.
Args:
input_data (obj): the request data.
content_type (str): the request Content-Type.
Returns:
(obj): data ready for prediction. For XGBoost, this defaults to DMatrix.
"""
return xgb_encoders.decode(input_data, content_type)
def test_decode_error():
with pytest.raises(_errors.UnsupportedFormatError):
encoder.decode(42, _content_types.OCTET_STREAM)
def test_decode_with_complex_csv_content_type(content_type):
dmatrix_result = encoder.decode("42.0,6.0,9.0\n42.0,6.0,9.0", content_type)
assert type(dmatrix_result) is xgb.DMatrix
