module plotting.plotting_validate_graph#

Short summary#

module mlprodict.plotting.plotting_validate_graph

Functions to help visualizing performances.

source on GitHub

Functions#

function

truncated documentation

_model_name

Extracts the main component of a model, removes suffixes such Classifier, Regressor, CV.

plot_validate_benchmark

Plots a graph which summarizes the performances of a benchmark validating a runtime for ONNX.

Documentation#

Functions to help visualizing performances.

source on GitHub

mlprodict.plotting.plotting_validate_graph._model_name(name)#

Extracts the main component of a model, removes suffixes such Classifier, Regressor, CV.

Parameters:

name – string

Returns:

shorter string

source on GitHub

mlprodict.plotting.plotting_validate_graph.plot_validate_benchmark(df)#

Plots a graph which summarizes the performances of a benchmark validating a runtime for ONNX.

Parameters:

df – output of function summary_report

Returns:

fig, ax

from logging import getLogger
from pandas import DataFrame
import matplotlib.pyplot as plt
from mlprodict.onnxrt.validate import enumerate_validated_operator_opsets, summary_report
from mlprodict.tools.plotting import plot_validate_benchmark

rows = list(enumerate_validated_operator_opsets(
    verbose=0, models={"LinearRegression"}, opset_min=11,
    runtime=['python', 'onnxruntime1'], debug=False,
    benchmark=True, n_features=[None, 10]))

df = DataFrame(rows)
piv = summary_report(df)
fig, ax = plot_validate_benchmark(piv)
plt.show()

source on GitHub