* introducing BatchContainer
* BatchContainer basic functionality done
* pass test_input to _convert
* introduce convert_batch API
* use convert_batch in the benchmark
* store _batch_size attribute
* test working
* run black, add concat output option, fix benchmark
* fix getattr
* fix operator benchmark
* support transform and decision function
* make sure input is tuple not list
* fix torch backend prediction
* begin fixing tests
* squeeze and ravel on onnx regression output
* all tests in test_extra_conf.py working
* restore BATCH_SIZE and k neighbor test
* fix onnxml test
* run black on test_extra_conf.py
* fix test_sklearn_normalizer_converter.py
* fix test_lightgbm_converter.py
* fixing more onnxml tests
* fixed remaining onnxml tests
* use format, fix pylint
* fix typo
* add document
* add missing doc
* fix typo
* doc update, remove unused stuff
* trying to get pipeline working
* updating string
* rtol atol
* pinning to macos version, reverint to 1 cache. removing rtol for now..
* macos 10.14
* rtol back
* Fix problem with inputs of shape == 0
Use proper data types for scalers
Remove ONNX_INPUTS since is not used
* Add missing test
fix problem with N_features when no test input is passed
* add explisti case for concat
* trying to make pytest run on all except ubu w/python 3.7
* fix problem with concat types in pt < 1.6
* add coverage for 3.5
* remove unecessary type conversions
Co-authored-by: Matteo Interlandi <mainterl@microsoft.com>
* add containers for onnx models
* add tvm_installed, initial work on topology
* add containers
add tvm backend to supported
add few tests
* fix type error in TVM
tree_trav and perf_tree_trav now work
* Add TVM_MAX_FUSE_DEPTH option
Add BATCH_SIZE option
Tree trav generate indexes based on batch size (if available)
TVM takes the max fuse detph configuration if set
* initial attempt at sphinx
* moving from shell script to just single command in yml
* adding make command, deleting sh
* ignoreing website/sphinx/_build/
* fixed missing hb prefix. testing out pipeline that will fail so hard
* fixing readme
* excluding web files from coverage. fixingdeps
* inner makefile
* adding secret for gh-pages for deploy
* using makefile
* changing subfolder to push
* excluding web files from coverage. fixingdeps
* fixing broken link, fixing doc tree depth
* purging pdoc
* add benchmark code for trees
* device can be added directly to convert
* add code for tvm
* refactoring of the tree benchmark files
* add operators scripts
few fixes in the tree bench
* no need for adding input schemas to onnx anymore
* remove ONNX_INITIAL_TYPES
* add support for onnx models with multiple inputs
* add test case when onnx input data type is not supported
* add missing test
fix a bug in cast for onnx
* add capability of setting number of threads to the container
* remove tuples from setup
* add psutil
* fix import problem
* fix xgb import
* add missing skipIf for onnx tests
* add back spark deps
* addressing Karla's comments
* wip
* fixing tests
* fixing formatting issues
* adding support to pass pyspark dataframe as test data
* adding support for pipelinemodel
* named input selection
* adding support for named inputs in spark-ml
* fixing issue
* fixing name error
* end-to-end spark-ml pipeline working
* more tests
* addressing review comments
* adding pyspark to ci/cd
* fixing bug
* sparkml requires torch >= 1.6.0 which has automatic type promotion
* input selection happening inside pytorch container
* fixing pytorch version
* reusing Concat from pipeline_implementations.py
* fixing Concat issue
* add support for pandas inputs
fix a couple of bugs
add support for double and long onnx input types
* makes OHE working over multiple inputs
add more tests
Karla Saur
Matteo Interlandi
masahi
Supun Nakandala
Karla Saur