-
Several PRs on github are failing now because of this error.
Seems related to these problems in pytest
- https://github.com/pytest-dev/pytest/issues/5502#issuecomment-526348052
- https://github.c…
-
1. Draw schemes that support the available use cases that Fedot' chain maintains (or planning to maintain in future).
2. Add the corresponding new page in documentation.
-
In this test the 'auto' preset in FEDOT API is choosing between 'best quality' and 'fast train' presets depending on the fit time. Timeout is set to 0.01 min to avoid the choice of 'best quality' pres…
-
Some content in https://fedot.readthedocs.io/ is obsoled and should be updated:
- Installation guide
- Quickstart
- New cases: texts + images
- Components of FEDOT
-
There is a need to assemble the example cases/multi_modal_genre_prediction.py via Fedot API.
-
https://github.com/DataCanvasIO/HyperTS
-
In the current version save_mode is True by default.
However, it can cause the unexpected quality decrease for user.
May be it's better to use 'false' as default case.
-
There are several flaws exists in validation of pipeline for time-series.
The possible corrections are:
1) Refactor has_no_data_flow_conflicts_in_ts_pipeline (https://github.com/nccr-itmo/FEDOT/…
-
A lot of notebooks starts from advanced, overcomplicated examples. In can cause the negative feedback from users un-famliar with FEDOT.
-
Currently link to https://fedot.onti.actcognitive.org/ gives "404 Not Found" error.