Andrey Getmanov

MultiModalData class improvement

3

Now csv files with text and table columns can be read and separated to various data sources just in one motion - MultiModalData.from_csv method added - text fields are defined...

Incorrect Binary Categorical Preprocessor work

2

1. Incorrectly defines categorical features from test_pipeline_for_side_task_predict (categorical_ids = [1, 2, 3, 4, 5, 7, 8], true cat_ids = [1, 2, 3, 4, 5, 6, 8] 2. column_uniques contains 1.0...

Multimodal data processing requires automatic definition of text fileds when they are in the same table with usual numerical data. It can be problematic because sometimes categorical features can be...

Current API does not support multitasking on multimodal data. However, the probable user may want to solve two tasks (regression and classification) simultaneously. So this opportunity should be added like...

Search space of hyperparameters is not defined at current moment. To effectively integrate CNN to FEDOT pipelines, its hyperparameters should be found as other parameters of pipeline.

Improvement of automatic text detection

1

Automatic text detection now is more effective, accurate and robust. 1. Every column which possibly contains text now is checked on the tf-idf vocabulary size. If this size is more...

Add unit tests for ALL InputData and Data classes methods #1200

5

## Summary Solution of #1200 issue. Add unit tests for: - [x] 1. Classification tasks: - [x] 1.1 Binary classification - [x] 1.2 Multitarget classification - [x] 2. Regression task...

- deprecated create_correct_path function is deleted in all of its variations - paths in tests and examples of pipeline loading are simplified

Pipeline checker doesn't check for existence of data source nodes in multimodal pipelines, but without them pipeline will crash

Hello and thank your for your project! I work as a developer on the AutoML open source framework [FEDOT](https://github.com/aimclub/FEDOT). We wanted to create a map of our stargazers, so I...