GitLab

* [dask] run one training task on each worker

* add comment on pure

* missing ticks

* empty commit

* include test for prediction with raw_score

* close client

* initial comments

* update data creation and include ranking task

* linting

* update _create_data

* compare unique raw_predictions with values in leaves_df

* Some mypy fixes

* address James' comments

* Re-introduce pass in empty classes

* Update compat.py

Remove extra lines

* [python-package] add type hints on Booster.set_network()

* change behavior

* Update Parallel-Learning-Guide.rst

* Update test.sh

* fix path

* address review comments

* [dask] raise more informative error for duplicates in 'machines'

* uncomment

* avoid test failure

* Revert "avoid test failure"

This reverts commit 9442bdf00f193a19a923dc0deb46b7822cb6f601.

* include support for init_score

* use dataframe from init_score and test difference with and without init_score in local model

* revert refactoring

* initial docs. test between distributed models with and without init_score

* remove ranker from tests

* test value for root node and change docs

* comma

* re-include parametrize

* fix incorrect merge

* use single init_score and the booster_ attribute

* use np.float64 instead of float

* include support for column array as label

* remove nested ifs

* fix linting errors

* include tests for sklearn regressors

* include docstring for numpy_1d_array_to_dtype

* include . at end of docstring

* remove pandas import and test for regression, classification and ranking

* check predictions of sklearn models as well

* test training only in dask. drop pandas series tests

* use PANDAS_INSTALLED and pd_Series

* inline imports

* use col array in fit for test_dask

* include review comments

* use socket.bind with port 0 and client.run to find random open ports

* include test for found ports

* find random open ports as default

* parametrize local_listen_port. type hint to _find_random_open_port. fid open ports only on workers with data.

* make indentation consistent and pass list of workers to client.run

* remove socket import

* change random port implementation

* fix test

* [dask] allow tight control over ports

* getting there, getting there

* fix params maybe

* fixing params

* remove unnecessary stuff

* fix tests

* fixes

* some minor changes

* fix flaky test

* linting

* more linting

* clarify parameter description

* add warning

* revert docs change

* Update python-package/lightgbm/dask.py

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* trying to fix stuff

* this is working

* update tests

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* indent
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [docs] Change some 'parallel learning' references to 'distributed learning'

* found a few more

* one more reference

Approximately %80 of runtime when loading "low column count, high row
count" DataFrames into Datasets is consumed in `np.fromiter`, called
as part of the `Dataset.get_field` method.

This is particularly pernicious hotspot, as unlike other ctypes-based
methods this is a hot loop over a python iterator loop and causes
significant GIL-contention in multi-threaded applications.

Replace `np.fromiter` with a direct call to `np.ctypeslib.as_array`,
which allows a single-shot `copy` of the underlying array.

This reduces the load time of a ~35 million row categorical dataframe
with 1 column from ~5 seconds to ~1 second, and allows multi-threaded
execution.

* run isort in CI linting job

* workaround conda compatibility issues

* minor mypy type errors fixed

* fix some warnings from mypy

* fix 3 mypy warnings

* selectively ignored some mypy errors

* minor mypy type errors fixed

* minor mypy type errors fixed

* minor mypy type errors fixed

* added import

* Update python-package/lightgbm/callback.py

* Apply suggestions from code review

* Apply suggestions from code review
Co-authored-by: James Lamb <jaylamb20@gmail.com>
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update dask.py

* Update sklearn.py

* got fit() working

* add predict()

* predict_proba()

* remove custom objective docs

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* fix capitalization

* Update tests/python_package_test/test_dask.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* fix tests

* fix tests

* fix test comments

* simplify tests

* Apply suggestions from code review

* add support for pandas categorical columns

* remove commented code

* quotes

* syntax error

* fix shape for ranker test

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update tests/python_package_test/test_dask.py

* trying

* fix tests

* remove unnecessary debugging stuff

* skip accuracy checks on categorical

* use category columns as categorical features
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update test_dask.py

* Update dask.py

* Update .vsts-ci.yml

* Revert "Update .vsts-ci.yml"

This reverts commit 98422be5b5095f0585de333b5b5545356776ef88.

* Update dask.py

* Update dask.py

* starting on Dask client

* more docs stuff

* fix pickling

* just copy docstrings

* fit docs

* switch test order

* linting

* use client kwarg

* remove inner set_params()

* add type hints

* fix type hints

* remove commented code

* reorder

* fix tests, add client_ property

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* fix tests

* linting

* simplify
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* start supporting Python 3.9

* Update .vsts-ci.yml

* Update .vsts-ci.yml

* Update .appveyor.yml

* Update .vsts-ci.yml

* Update .appveyor.yml

* fixed conflicts

* continue