Fedot

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.4/

Features & Enhancements:
* Boosting method implementation (XGBoost) by RomanKharkovskoy in https://github.com/aimclub/FEDOT/pull/1209
* Boosting method implementation (LightGBM) by RomanKharkovskoy in https://github.com/aimclub/FEDOT/pull/1264

Bugfixes:
* Sktime fix by nicl-nno in https://github.com/aimclub/FEDOT/pull/1323

What's Changed
* fix: NaNs containing forecast on some ets configs by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1292
* fix: NaNs containing data on fit stage for fast_ica by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1295
* fix: go for forecast method instead of predict in ets by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1299
* fix: indexing and seasonal_periods check that endog is at least twice the size by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1305
* Fix multitarget stratification by DRMPN in https://github.com/aimclub/FEDOT/pull/1308
* Update unit-build.yml by technocreep in https://github.com/aimclub/FEDOT/pull/1294
* Hotfix of fit time estimation for initial assumption by nicl-nno in https://github.com/aimclub/FEDOT/pull/1309
* Fix for 1296 by aPovidlo in https://github.com/aimclub/FEDOT/pull/1312
* hotfix: go for FeaturesType instead of InputData in a pipeline tuning by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1311
* Boosting method implementation (XGBoost) by RomanKharkovskoy in https://github.com/aimclub/FEDOT/pull/1209
* fix: bring back cgru integration tests by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1313
* Boosting method implementation (LightGBM) by RomanKharkovskoy in https://github.com/aimclub/FEDOT/pull/1264
* hotfix: Bug with target encoding by aPovidlo in https://github.com/aimclub/FEDOT/pull/1314
* hotfix: boostings related failure of test_*_models_check_fit_predict_correct by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1316
* hotfix: add punkt_tab to nltk LookupError check by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1319
* hotfix: pin nltk version to avoid downloading punkt_tab by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1321
* Sktime fix by nicl-nno in https://github.com/aimclub/FEDOT/pull/1323
* m4 timeseries benchmark by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1277
* fix: do not launch composition for atomized model by DRMPN in https://github.com/aimclub/FEDOT/pull/1324

**Full Changelog**: https://github.com/aimclub/FEDOT/compare/v0.7.3.2...v0.7.4

Minor fixes and improvements.

What's Changed
* Update main_concepts.rst by glad4enkonm in https://github.com/aimclub/FEDOT/pull/1275
* Fix pca by valer1435 in https://github.com/aimclub/FEDOT/pull/1267
* Fix sporadic forecasting test failure by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1280
* Update data.py by v1docq in https://github.com/aimclub/FEDOT/pull/1283
* Add validation on _window_size param in topo extractor by Lopa10ko in https://github.com/aimclub/FEDOT/pull/1284
* Update version of codecov action by DRMPN in https://github.com/aimclub/FEDOT/pull/1289
* Fix gitlab mirroring by Nunkyl in https://github.com/aimclub/FEDOT/pull/1288
* Fix gensim for PyPi release by DRMPN in https://github.com/aimclub/FEDOT/pull/1287


**Full Changelog**: https://github.com/aimclub/FEDOT/compare/v0.7.3.1...v0.7.3.2

Minor fixes and improvements, better installation

[0.7.3](https://github.com/aimclub/FEDOT/releases/tag/v0.7.3) [Latest](https://github.com/aimclub/FEDOT/releases/latest)

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.3/

Features & Enhancements:

- Update of [GOLEM](https://github.com/aimclub/GOLEM) (core framework) dependency to 0.4.0 version, that has some important features itself
- Improvement of data preprocessing
- Improvements for time series forecasting
- Better API

Full list of changes:

* Fix params setting by YamLyubov in https://github.com/aimclub/FEDOT/pull/1121
* Clean up SA module & add example by maypink in https://github.com/aimclub/FEDOT/pull/1114
* Debug mode and modified fitness evaluation for for time series by nicl-nno in https://github.com/aimclub/FEDOT/pull/1116
* Lagged implementation speedup by kasyanovse in https://github.com/aimclub/FEDOT/pull/1139
* Support of multimodal data in DataSourceSplitter + bugfix 1119 by kasyanovse in https://github.com/aimclub/FEDOT/pull/1138
* Fix low speed of lagged implementation by kasyanovse in https://github.com/aimclub/FEDOT/pull/1144
* 628 bugfix by kasyanovse in https://github.com/aimclub/FEDOT/pull/1145
* Fix random FEDOT results with set seed by kasyanovse in https://github.com/aimclub/FEDOT/pull/1143
* Replace exogenous example by YamLyubov in https://github.com/aimclub/FEDOT/pull/1108
* Fix multiobj tuning by YamLyubov in https://github.com/aimclub/FEDOT/pull/1156
* Decimal places control implemented by nicl-nno in https://github.com/aimclub/FEDOT/pull/1161
* Title for Pipeline plots by maypink in https://github.com/aimclub/FEDOT/pull/1162
* Prediction Intervals by GrigoriJasnovidov in https://github.com/aimclub/FEDOT/pull/1149
* Add MASE metric by valer1435 in https://github.com/aimclub/FEDOT/pull/1166
* Remove redundant metrics mapping by YamLyubov in https://github.com/aimclub/FEDOT/pull/1167
* Refactor convert to op by valer1435 in https://github.com/aimclub/FEDOT/pull/1176
* Fix default tuning params by valer1435 in https://github.com/aimclub/FEDOT/pull/1175
* Add desc to not implemented errors by valer1435 in https://github.com/aimclub/FEDOT/pull/1177
* `from fedot import Fedot` by MorrisNein in https://github.com/aimclub/FEDOT/pull/1179
* Boosting method implementation (Catboost) by aPovidlo in https://github.com/aimclub/FEDOT/pull/1155
* Add documented API params setters by MorrisNein in https://github.com/aimclub/FEDOT/pull/1077
* add InputData from pd and numpy by ChrisLisbon in https://github.com/aimclub/FEDOT/pull/1184
* Support for changes in GOLEM by kasyanovse in https://github.com/aimclub/FEDOT/pull/1189
* Docs fix for atomized model by nicl-nno in https://github.com/aimclub/FEDOT/pull/1197
* get predict_data.num_classes if trained_model havent num_classes by ElishaFlacon in https://github.com/aimclub/FEDOT/pull/1199
* Metrics fixes, tests, refactoring, typing improvement by MorrisNein in https://github.com/aimclub/FEDOT/pull/1214
* 922 preprocessor acceleration by IIaKyJIuH in https://github.com/aimclub/FEDOT/pull/1004
* Add direct examples by valer1435 in https://github.com/aimclub/FEDOT/pull/1192
* Improve the rule for time series pipeline verification by kasyanovse in https://github.com/aimclub/FEDOT/pull/1210
* Fixes for unit and intergration tests by nicl-nno in https://github.com/aimclub/FEDOT/pull/1238
* add window size selector by kasyanovse in https://github.com/aimclub/FEDOT/pull/1237
* Add topofeatures by valer1435 in https://github.com/aimclub/FEDOT/pull/1241

**Full Changelog**: https://github.com/aimclub/FEDOT/compare/v0.7.2...v0.7.3

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.2/

Features & Enhancements:

- Update of [GOLEM](https://github.com/aimclub/GOLEM) (core framework) dependency to 0.3.2 version, that has some important features itself

Bugfixes:

- Minor improvements and fixes

Hello, AutoML folk! We’re releasing a minor version of FEDOT that includes the following.

PyPi release: https://pypi.org/project/fedot/0.7.1/

Features & Enhancements:
- Update of [GOLEM](https://github.com/aimclub/GOLEM) (core framework) dependency to 0.3.1 version, that has some important features itself
- Added *Meta Rules* that automatically select best parameters based on dataset: with rules for early_stopping_generations, preset and cross-validation folds were added (1057)
- Improved API parameters and made documentation more clear and structured (1067, 1041)
- Improve test suite and its performance (1098)
- Improved DataMerger for textual data – now multi-column text table can be used with FEDOT (1052)

Bugfixes:
- Bug with wrong combinations of operations in pipelines for time series forecasting was fixed.
- Multiple initial assumptions support was fixed (1070)
- Various minor fixes