Should we allow TSDataset to hold non-continuous series?

[iKintosh]

Checklist

• [X] I have searched the issues.
• [X] I have searched the docs.

❓ Questions and Help

Some tasks require us to delete some time periods from dataset (for example, the beginning of COVID in 2020). Should we allow our users to upload datasets with missing values as well?

[julia-shenshina]

We already do 🤔

How to check

from etna.datasets import generate_ar_df, TSDataset

df = generate_ar_df(periods=100, start_time="2020-01-01", n_segments=2)
df = df[~((df.segment == "segment_0") & (df.timestamp.between("2020-01-10", "2020-01-30")))]
df = TSDataset.to_dataset(df)
ts = TSDataset(df, "D")
ts.plot()

[alex-hse-repository]

Will work on it in the separate track