Note
Click here to download the full example code
Train, convert and predict with ONNX Runtime¶
This example demonstrates an end to end scenario starting with the training of a machine learned model to its use in its converted from.
Train a logistic regression¶
The first step consists in retrieving the iris datset.
from sklearn.datasets import load_iris
iris = load_iris()
X, y = iris.data, iris.target
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y)
Then we fit a model.
from sklearn.linear_model import LogisticRegression
clr = LogisticRegression()
clr.fit(X_train, y_train)
Out:
LogisticRegression()
We compute the prediction on the test set and we show the confusion matrix.
from sklearn.metrics import confusion_matrix
pred = clr.predict(X_test)
print(confusion_matrix(y_test, pred))
Out:
[[ 8 0 0]
[ 0 9 1]
[ 0 2 18]]
Conversion to ONNX format¶
We use module sklearn-onnx to convert the model into ONNX format.
from skl2onnx import convert_sklearn
from skl2onnx.common.data_types import FloatTensorType
initial_type = [('float_input', FloatTensorType([None, 4]))]
onx = convert_sklearn(clr, initial_types=initial_type)
with open("logreg_iris.onnx", "wb") as f:
f.write(onx.SerializeToString())
We load the model with ONNX Runtime and look at its input and output.
import onnxruntime as rt
sess = rt.InferenceSession("logreg_iris.onnx", providers=rt.get_available_providers())
print("input name='{}' and shape={}".format(
sess.get_inputs()[0].name, sess.get_inputs()[0].shape))
print("output name='{}' and shape={}".format(
sess.get_outputs()[0].name, sess.get_outputs()[0].shape))
Out:
input name='float_input' and shape=[None, 4]
output name='output_label' and shape=[None]
We compute the predictions.
input_name = sess.get_inputs()[0].name
label_name = sess.get_outputs()[0].name
import numpy
pred_onx = sess.run([label_name], {input_name: X_test.astype(numpy.float32)})[0]
print(confusion_matrix(pred, pred_onx))
Out:
[[ 8 0 0]
[ 0 11 0]
[ 0 0 19]]
The prediction are perfectly identical.
Probabilities¶
Probabilities are needed to compute other relevant metrics such as the ROC Curve. Let’s see how to get them first with scikit-learn.
prob_sklearn = clr.predict_proba(X_test)
print(prob_sklearn[:3])
Out:
[[2.50532892e-03 8.48834843e-01 1.48659828e-01]
[1.80424972e-04 2.23589233e-01 7.76230342e-01]
[9.47235933e-01 5.27628799e-02 1.18664935e-06]]
And then with ONNX Runtime. The probabilies appear to be
prob_name = sess.get_outputs()[1].name
prob_rt = sess.run([prob_name], {input_name: X_test.astype(numpy.float32)})[0]
import pprint
pprint.pprint(prob_rt[0:3])
Out:
[{0: 0.00250532990321517, 1: 0.8488348722457886, 2: 0.1486598700284958},
{0: 0.00018042475858237594, 1: 0.22358904778957367, 2: 0.77623051404953},
{0: 0.9472358822822571, 1: 0.052762895822525024, 2: 1.1866491149703506e-06}]
Let’s benchmark.
from timeit import Timer
def speed(inst, number=10, repeat=20):
timer = Timer(inst, globals=globals())
raw = numpy.array(timer.repeat(repeat, number=number))
ave = raw.sum() / len(raw) / number
mi, ma = raw.min() / number, raw.max() / number
print("Average %1.3g min=%1.3g max=%1.3g" % (ave, mi, ma))
return ave
print("Execution time for clr.predict")
speed("clr.predict(X_test)")
print("Execution time for ONNX Runtime")
speed("sess.run([label_name], {input_name: X_test.astype(numpy.float32)})[0]")
Out:
Execution time for clr.predict
Average 5.1e-05 min=4.58e-05 max=6.63e-05
Execution time for ONNX Runtime
Average 2.42e-05 min=2.33e-05 max=3.53e-05
2.4194435000879365e-05
Let’s benchmark a scenario similar to what a webservice experiences: the model has to do one prediction at a time as opposed to a batch of prediction.
def loop(X_test, fct, n=None):
nrow = X_test.shape[0]
if n is None:
n = nrow
for i in range(0, n):
im = i % nrow
fct(X_test[im: im+1])
print("Execution time for clr.predict")
speed("loop(X_test, clr.predict, 100)")
def sess_predict(x):
return sess.run([label_name], {input_name: x.astype(numpy.float32)})[0]
print("Execution time for sess_predict")
speed("loop(X_test, sess_predict, 100)")
Out:
Execution time for clr.predict
Average 0.00499 min=0.00471 max=0.0057
Execution time for sess_predict
Average 0.00116 min=0.00105 max=0.00138
0.001157615160000205
Let’s do the same for the probabilities.
print("Execution time for predict_proba")
speed("loop(X_test, clr.predict_proba, 100)")
def sess_predict_proba(x):
return sess.run([prob_name], {input_name: x.astype(numpy.float32)})[0]
print("Execution time for sess_predict_proba")
speed("loop(X_test, sess_predict_proba, 100)")
Out:
Execution time for predict_proba
Average 0.00729 min=0.00697 max=0.00805
Execution time for sess_predict_proba
Average 0.00109 min=0.00106 max=0.00114
0.0010851338200001236
This second comparison is better as ONNX Runtime, in this experience, computes the label and the probabilities in every case.
Benchmark with RandomForest¶
We first train and save a model in ONNX format.
from sklearn.ensemble import RandomForestClassifier
rf = RandomForestClassifier()
rf.fit(X_train, y_train)
initial_type = [('float_input', FloatTensorType([1, 4]))]
onx = convert_sklearn(rf, initial_types=initial_type)
with open("rf_iris.onnx", "wb") as f:
f.write(onx.SerializeToString())
We compare.
sess = rt.InferenceSession("rf_iris.onnx", providers=rt.get_available_providers())
def sess_predict_proba_rf(x):
return sess.run([prob_name], {input_name: x.astype(numpy.float32)})[0]
print("Execution time for predict_proba")
speed("loop(X_test, rf.predict_proba, 100)")
print("Execution time for sess_predict_proba")
speed("loop(X_test, sess_predict_proba_rf, 100)")
Out:
Execution time for predict_proba
Average 0.845 min=0.839 max=0.853
Execution time for sess_predict_proba
Average 0.00131 min=0.00129 max=0.00133
0.0013066128300005175
Let’s see with different number of trees.
measures = []
for n_trees in range(5, 51, 5):
print(n_trees)
rf = RandomForestClassifier(n_estimators=n_trees)
rf.fit(X_train, y_train)
initial_type = [('float_input', FloatTensorType([1, 4]))]
onx = convert_sklearn(rf, initial_types=initial_type)
with open("rf_iris_%d.onnx" % n_trees, "wb") as f:
f.write(onx.SerializeToString())
sess = rt.InferenceSession("rf_iris_%d.onnx" % n_trees, providers=rt.get_available_providers())
def sess_predict_proba_loop(x):
return sess.run([prob_name], {input_name: x.astype(numpy.float32)})[0]
tsk = speed("loop(X_test, rf.predict_proba, 100)", number=5, repeat=5)
trt = speed("loop(X_test, sess_predict_proba_loop, 100)", number=5, repeat=5)
measures.append({'n_trees': n_trees, 'sklearn': tsk, 'rt': trt})
from pandas import DataFrame
df = DataFrame(measures)
ax = df.plot(x="n_trees", y="sklearn", label="scikit-learn", c="blue", logy=True)
df.plot(x="n_trees", y="rt", label="onnxruntime",
ax=ax, c="green", logy=True)
ax.set_xlabel("Number of trees")
ax.set_ylabel("Prediction time (s)")
ax.set_title("Speed comparison between scikit-learn and ONNX Runtime\nFor a random forest on Iris dataset")
ax.legend()
Out:
5
Average 0.0745 min=0.0741 max=0.0748
Average 0.00103 min=0.000996 max=0.00105
10
Average 0.115 min=0.114 max=0.116
Average 0.00103 min=0.001 max=0.00106
15
Average 0.156 min=0.154 max=0.157
Average 0.00109 min=0.00107 max=0.00112
20
Average 0.197 min=0.196 max=0.197
Average 0.00108 min=0.00105 max=0.00113
25
Average 0.236 min=0.235 max=0.238
Average 0.00109 min=0.00106 max=0.00113
30
Average 0.277 min=0.277 max=0.278
Average 0.00112 min=0.00107 max=0.00116
35
Average 0.319 min=0.318 max=0.321
Average 0.00114 min=0.00113 max=0.00116
40
Average 0.359 min=0.358 max=0.361
Average 0.00114 min=0.0011 max=0.00118
45
Average 0.399 min=0.397 max=0.401
Average 0.00116 min=0.00113 max=0.00118
50
Average 0.439 min=0.437 max=0.441
Average 0.00116 min=0.00115 max=0.00121
<matplotlib.legend.Legend object at 0x7fa5848e9040>
Total running time of the script: ( 3 minutes 57.930 seconds)
