[FSTORE-1439] Merge hopsworks-api, feature-store-api and machine-learning-api #230
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…atures (#1011) * accept datetime object in filter * removed unused varible * support for unbinded features * Update python/tests/core/test_arrow_flight_client.py Co-authored-by: Ralf <[email protected]> * Update python/tests/core/test_arrow_flight_client.py Co-authored-by: Ralf <[email protected]> * simplified diff * Update python/hsfs/core/arrow_flight_client.py Co-authored-by: Ralf <[email protected]> --------- Co-authored-by: Ralf <[email protected]>
…ing (#1008) * cleaned up read printout * udpated message * filter only userwarnings
Co-authored-by: Fabio Buso <[email protected]>
…997) Co-authored-by: Fabio Buso <[email protected]>
…on and path fix (#992) Co-authored-by: Ralf <[email protected]>
…asets (#1021) * add spine group support
* change wait default for fg.insert/save * fixed test * added get_state to job class * added final state and updated docs * updated docs * reverted import change * moved job config below job * added wait parameter * fixed tests * do not override write_options with wait= parameter * one more place --------- Co-authored-by: Moritz Meister <[email protected]>
…spark engine (#1031) Co-authored-by: Fabio Buso <[email protected]>
…ure views (#1029)
* working state * refactored query parameters * removed numeric parameter * fix bug * look in all features in all featuregroups * support excluded features * added features property to featureview * include transformation functions in repr * fix column check on join * updated warning messages * more refactoring, fixed tests * fixed tests, added filter check * fixed test * fixed ambiguity issue with filter params * fixed tests and cleanup * check full filter * fix append feature and check filter there as well * don't check merged filter on .filter * working state * removed checks * some renaming * made lookup function private/protected * added docs * fixed cached check * updated error messages, reverted transformation function test changes * added tests * added add_features_to_collection method * collect features on request * fixed comment
* removed pandas version pinning * patched pandas, patches jdbc tests, added unittests * removed pandas dep from dev * removed pandas1sqlachemy2, pandas2sqlalchemy2 * rename action * added Dev-pandas1 for ci testing on pandas 1.5.3 * added fsspec dependency, required by pandas.read * fixed statistics serialization
…040) Co-authored-by: Fabio Buso <[email protected]>
* Decamelize dict to pass correct args * Fix id in from_ge_type * Added ids to dict for conversions * allow extra kwargs
… path (#1358) * Append MATERIAL_DIRECTORY env var value to certificates path * Also include MATERIAL_DIR in token.jwt
* Fix test_base_client.py It was impure, changing environ variable and thereby messing other tests. This commit wraps these tests in a decorator saving and restoring os.environ. * Move changes_environ to tests/utils.py
* Refactor mysql dependencies out of util * Fix tests for mysql * Start cleaning some dependencies under type_checking
* Start removing hive related dependencies * More hive cleanups * Update ruff pre-commit hook * Start fixing pytest * REmove hive related tests * Review
…imary_key" (#1357) * primary_keys -> primary_key * backwards compatibility * backwards compatibility * Review and fix missed foc change --------- Co-authored-by: davitbzh <[email protected]> Co-authored-by: Victor Jouffrey <[email protected]> Co-authored-by: Victor Jouffrey <[email protected]>
* feature logging * clean up * default write option * doc * fix * fix style * revert pd.StringDtype() * fix doc * fix doc * doc and copy df * get untransformed features * rename transformed_features * get untransformed features * return batch df * address comments * rename internal methods * fix style * add warning * fix circular import * set logging_enabled to True * fix style
* Start Kafka refactor * Continue refactor of kafka/avro out of python engine * Minor tidy up of args * Tidy up python engine * Minor refactoring of multi_part_insert and kafka headers * Options to get_kafka_config with spark options * Fix mocking * Move and fix kafka tests * More fixing of mocking * More work on unit test fixing * Fix with import reload * Fix mocking in test_feature_group_writer * Fix mocking in engine/test_python --------- Co-authored-by: Aleksey Veresov <[email protected]>
…ion in spark (#1368)
* left join * snowflake * test_join --------- Co-authored-by: davitbzh <[email protected]> Co-authored-by: Dhananjay Mukhedkar <[email protected]>
* hopsworks_udf first version * working code for running hopsworks udf without saving in backend using python client * removing debugging logs * statistics working with python client * basic functionality working with backend * code with statistics working and saved to backend * working code for feature vector * reformatted and documented Hopswork UDF class * unit tests for transformation functions * clearning transformations engine and adding unit tests * feature view api formated * reformatting and fixing feature_view_engine * reformatted and added unit tests for feature view * updating documentation for feature store * updating documentation for feature store * fixed tests for training datatset features * reformatted and added unit tests for python engine * most unit tests fixed * all unit tests working * removed print * adding test for hopsworks_udf * correcting merge for vector server * reformatting with ruff * fixing vector server * fixing docs * fixing vector server * fixing building in transformations * correcting get feature vector * adding missed changes for build in transformations * shallow copying scope dictonary to not overwrite statistics variable for different udf's having same statistics parameter name * adding deep copy to create multiple transfromation functions with different features * sorting transformation function to maintain consistent order * sorting transformation functions in transformation function engine to mainatin same order * using feature view transformation functions * addressing review comments * using PYARROW_EXTENSION_ENABLE during import rather than as a function * skiping transformation function test in windows spark udf failing due to dependencies with greater expectation * changing transformed_feature_vector_col_name to transformed_features to obtain feature names after transfromations * adding property transformed_features in feature view to obtain feature names after transfromations * updating doc string and adding property decorator missed during rebase * refactoring transformation functions to update parsing of statistics parameters and also renaming decorator name * refactoring transformation functions to update parsing of statistics parameters and also renaming decorator name * reformating with ruff * adding statistics to udf only if required * convrting extended statistics to dictonary * sorting built in label encoder to maintain consistency * adding type hints for class TransformationStatistics * adapating to backend update of reaturning output_types, transformation_features and statistics_argument_names as Lists * fixing unit tests * removign space in doc string * replace - from output column names with _ * revreting unwanted spark test _ replace changes * adding missed import * correcting to_dict feature view * reverting python.py unintentional changes during rebase * rebase and adding back missed import
…ning-api * Remove redundant java-related files * Move java code of hsfs * Move locust_benchmark of hsfs * Move python code of hsfs * Move python code of hsml * Move hsml tests * Remove redundant __init__.py in hsml/tests * Move hsfs tests * Merge hsfs tests * Fix the problem with impure tests changing backend_fixtures One of the tests is impure, changing backend_fixtures, which breaks test_model. This commit makes backend_fixtures to return a deep copy of backend_fixtures_json isntead of returning them directly. * Fix ruff check of test_base_client * Move hsfs utils * Merge hsfs docs * Fix pyproject dependencies for hsfs docs * Merge the rest of hsfs docs * Merge hsfs mkdocs * Merge hsml docs * Add files generated by auto_doc to gitignore * Merge hsfs pyproject.toml * Merge hsml pyproject.toml * Remove hsfs and hsml pyproject.toml * Remove hive from pyproject dependencies * Fix mistypes in pyproject * Fix documentation generation * Fix circular dependencies * Fix circular dependency in hsml (CONNECTION_SAAS_HOSTNAME) * Fix circular dependency in hsml (model/util) * Ruff fix hsml/util * Fix docgen There is a very strange bug: hsml.python for some reason is available only as python.hsml.python. * Fix hsml.python import errors * Fix import problem in docgen * Ruff fix auto_doc *Revert the move of requirements-docs to an extra See PR #209 * Move workflows * Merge workflows * Remove redundant github files * Skip test_login until Robin is back * Fix test_connection * Fix test_hopsworks_udf * Fix import problem in docgen * Fix pyproject optional docgen dependency * Revert the move of requirements-docs to an extra See PR #209 * Rename unit_tests_pandas to unit_tests_pandas1 for clarity * Rename unit_tests_typechecked * Merge workflows * Fix import problem in docgen * Merge and remove the rest of python files * Merge gitignore * Move docker and jenkins files * Merge CONTRIBUTING * Merge README and remove hsfs and hsml subdirectories * Revert the move of requirements-docs to an extra See PR #209 * Add hopsworks_common
kennethmhc
approved these changes
Jul 17, 2024
aversey
force-pushed
the
dev-merge
branch
3 times, most recently
from
474ebff to
67484a9
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.