Examples Gallery#
Benchmark inference for scikit-learn models
Train and deploy a scikit-learn pipeline
Convert a pipeline with a LightGBM classifier
Store arrays in one onnx graph
Intermediate results and investigation
Black list operators when converting
Forward backward on a neural network on GPU
Train a scikit-learn neural network with onnxruntime-training on GPU
Train a linear regression with forward backward
Train a linear regression with onnxruntime-training
SerializeToString and ParseFromString
Forward backward on a neural network on GPU (Nesterov) and penalty
Implement a new converter using other converters
Batch predictions vs one-off predictions
Change the number of outputs by adding a parser
One model, many possible conversions with options
Two ways to implement a converter
Choose appropriate output of a classifier
Convert a pipeline with a XGBoost model
Profiling of ONNX graph with onnxruntime
Batch predictions without parallelization
Convert a pipeline with a LightGBM regressor
Issues when switching to float
Benchmark onnxruntime optimization
Benchmark, comparison scikit-learn - forward-backward
Benchmark, comparison scikit-learn - onnxruntime-training
Dealing with discrepancies (tf-idf)
Converter for WOEEncoder from categorical_encoder
sphx_glr_gyexamples_plot_quantization.py
Benchmark onnxruntime API: run or run_with_ort_values
Benchmark, comparison torch - forward-backward
Benchmark, comparison sklearn - forward-backward - classification
Benchmark inference for a linear regression
Fast design with a python runtime
Compares numpy to onnxruntime on simple functions
Benchmark and profile of operator Slice
Add a parser to handle dataframes
Multithreading with onnxruntime
Train a linear regression with onnxruntime-training on GPU in details
Benchmark onnxruntime API: eager mode
Train a linear regression with onnxruntime-training in details
Multithreading with onnxruntime and big models