onnxmltools: Convert your model into ONNX

ONNXMLTools enables you to convert models from different machine learning toolkits into ONNX. Currently the following toolkits are supported:

• Apple Core ML, (onnx-coreml does the reverse conversion from onnx to Apple Core ML) (up to version 3.1)

• catboost

• h2o (a subset only)

• Keras

• LightGBM

• libsvm

• scikit-learn (subset of models convertible to ONNX)

• SparkML

• XGBoost

onnxmltools leverages existing converting library, sklearn-onnx, tensorflow-onnx and implements converters for the other libraries.

• Tutorial
  • Step 1: Train a model using your favorite framework
  • Step 2: Convert or export the model into ONNX format
  • Step 3: Load and run the model using ONNX Runtime
  • Target opset
• API Summary
  • Converters
  • Utils
• Gallery of examples

onnxmltools converts models in ONNX format which can be then used to compute predictions with the backend of your choice. Every converter is tested with: onnxruntime (does also exist with GPU: onnxruntime-gpu). Here is a typical example which trains a model, converts into ONNX and finally uses onnxruntime to predict.

# Train a model.
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
iris = load_iris()
X, y = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y)
clr = RandomForestClassifier()
clr.fit(X_train, y_train)

# Convert into ONNX format with onnxmltools
from onnxmltools import convert_sklearn
from onnxmltools.utils import save_model
from onnxconverter_common.data_types import FloatTensorType
initial_type = [('float_input', FloatTensorType([1, 4]))]
onx = convert_sklearn(clr, initial_types=initial_type)
save_model(onx, "rf_iris.onnx")

# Compute the prediction with ONNX Runtime
import onnxruntime as rt
import numpy
sess = rt.InferenceSession("rf_iris.onnx")
input_name = sess.get_inputs()[0].name
label_name = sess.get_outputs()[0].name
pred_onx = sess.run([label_name], {input_name: X_test.astype(numpy.float32)})[0]