-
Fedot crushing after fitting it with ts full of zero values.
Code to reproduce:
```
horizon = 10
task_params = TsForecastingParams(forecast_length=horizon)
task = Task(TaskTypesEnum.ts_foreca…
-
Pipeline with stl_arima can not be serialized correctly with pipeline.save()
![image](https://user-images.githubusercontent.com/5711814/149188284-e7352f35-8a9d-4bd8-8543-9d558d68296b.png)
-
When we forecast futures values, there is always uncertainty around the values that needed to be quantified. This is called risk. It would be nice if we can provide a probabilistic forecast around the…
-
The example for FEDOT-GOLEM-iOPt should be prepared.
The efficiency should be compared with existing HyperOpt tuner.
-
`tsfresh` gives access to a whole range of features engineered specifically for time-series problems, and also has sensible feature selection algorithms built in. Is there a way to use tsfresh as a no…
-
Для задачи прогнозирования временных рядов генерируется только первое и последнее поколение. Пример скрипта
import numpy as np
import pandas as pd
from fedot.api.main import Fedot
from fedot.c…
-
In [this ](fedot/core/pipelines/verification.py) PR `initial_assumption` with smoothing+autoregression was removed due to `Unable to allocate 44.9 GiB for an array with shape (2010082226, 3) and data …
-
I open the documentation using the link from the ReadMe file, I get a 404 error.
-
Some bugs arised during time series benchmark experiments (the post will be updated periodically):
```
OptimisationTimer - Composition time: 1.8 min
Exception "no default __reduce__ due to non-tr…
-
The utilities module is duplicated now:
![image](https://user-images.githubusercontent.com/5711814/188181347-36ed4817-cff6-434d-ac4d-1ed66f5ff7cb.png)