class _DataInterface(BaseDataInterface):
input_name = os.getenv(
'INPUT_NAME',
_ModelConfigurations().options['input_name'])
output_name = os.getenv(
'OUTPUT_NAME',
_ModelConfigurations().options['output_name'])
def __init__(self, model_runners):
self.model_runners = model_runners
self.classifiers = OrderedDict()
self.input_name = None
self.channel = None
self.stub = None
self.model_spec_name = os.getenv("MODEL_SPEC_NAME", _ModelConfigurations().options["model_spec_name"])
self.model_spec_signature_name = os.getenv("MODEL_SPEC_SIGNATURE_NAME", _ModelConfigurations().options["model_spec_signature_name"])
self.load_model()
def test_ModelConfigurations():
prediction_types = [e.value for e in PREDICTION_TYPE]
assert os.path.exists(_ModelConfigurations().interface_filepath)
assert isinstance(_ModelConfigurations().interface_dict, Dict)
assert isinstance(_ModelConfigurations().model_name, str)
assert isinstance(_ModelConfigurations().io, Dict)
assert isinstance(_ModelConfigurations().model_runners, List)
assert isinstance(_ModelConfigurations().runner, str)
assert _ModelConfigurations().prediction_type in prediction_types
def predict(self, input_data: Image) -> np.ndarray:
logger.info(f'run predict proba in {self.__class__.__name__}')
_prediction = input_data
for k, v in self.classifiers.items():
if v['runner'] == MODEL_RUNTIME.SKLEARN.value:
_prediction = np.array(v['predictor'].transform(_prediction))
else:
request = predict_pb2.PredictRequest()
request.model_spec.name = self.model_spec_name
request.model_spec.signature_name = self.model_spec_signature_name
request.inputs[_DataInterface().input_name].CopyFrom(tf.make_tensor_proto(
_prediction, shape=_ModelConfigurations().io['input_shape']))
result = self.stub.Predict(request, TIMEOUT_SECOND)
_prediction = np.array(
result.outputs[_DataInterface().output_name].float_val)
output = _prediction
return output
from typing import List, Sequence
import numpy as np
import onnxruntime as rt
from src.app.configurations import _ModelConfigurations
from src.app.constants import MODEL_RUNTIME
from src.app.ml.base_predictor import BaseData, BaseDataInterface, BaseDataConverter, BasePredictor
from src.app.ml.save_helper import load_labels
import logging
logger = logging.getLogger(__name__)
LABELS = load_labels(_ModelConfigurations().options["label_filepath"])
class _Data(BaseData):
test_data: List[List[int]] = [[5.1, 3.5, 1.4, 0.2]]
labels: List[str] = LABELS
class _DataInterface(BaseDataInterface):
pass
class _DataConverter(BaseDataConverter):
pass
class _Classifier(BasePredictor):
def __init__(self, model_runners):
self.model_runners = model_runners
class _DataInterface(BaseDataInterface):
input_name = os.getenv("INPUT_NAME",
_ModelConfigurations().options["input_name"])
output_name = os.getenv("OUTPUT_NAME",
_ModelConfigurations().options["output_name"])
