Elgato_dark/chronos-forecasting
1
0
Fork 0
mirror of https://github.com/amazon-science/chronos-forecasting synced 2026-05-24 01:58:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Update comments

Browse source
This commit is contained in:
Oleksandr Shchur 2026-01-19 10:20:08 +00:00
parent 1f405f5a2d
commit b285ee6148
3 changed files with 15 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
src/chronos/base.py
View file

@ -165,9 +165,11 @@ class BaseChronosPipeline(metaclass=PipelineRegistry):
quantile_levels
Quantile levels to compute
validate_inputs
[ADVANCED] When False, skips validation. You must ensure: (1) df and future_df (if provided)
are sorted by (id_column, timestamp_column); (2) future_df (if provided) contains exactly
prediction_length rows per item. Defaults to True.
[ADVANCED] When True (default), validates dataframes before prediction. Setting
to False is faster but data errors may silently lead to wrong predictions. When
False, you must ensure: (1) all dataframes are sorted by (id_column, timestamp_column);
(2) future_df (if provided) has the same item IDs as df with exactly
prediction_length rows of future timestamps per item.
freq
Frequency string for timestamp generation (e.g., "H", "D", "W"). Can only be used when
validate_inputs=False. When provided, skips frequency inference from the data.

8
src/chronos/chronos2/pipeline.py
View file

@ -865,9 +865,11 @@ class Chronos2Pipeline(BaseChronosPipeline):
For optimal results, consider using a batch size around 100 (as used in the Chronos-2 technical report).
- Cross-learning is most helpful when individual time series have limited historical context, as the model can leverage patterns from related series in the batch.
validate_inputs
[ADVANCED] When False, skips validation. You must ensure: (1) df and future_df (if provided)
are sorted by (id_column, timestamp_column); (2) future_df (if provided) contains exactly
prediction_length rows per item. Defaults to True.
[ADVANCED] When True (default), validates dataframes before prediction. Setting
to False is faster but data errors may silently lead to wrong predictions. When
False, you must ensure: (1) all dataframes are sorted by (id_column, timestamp_column);
(2) future_df (if provided) has the same item IDs as df with exactly
prediction_length rows of future timestamps per item.
freq
Frequency string for timestamp generation (e.g., "h", "D", "W"). Can only be used when
validate_inputs=False. When provided, skips frequency inference from the data.

8
src/chronos/df_utils.py
View file

@ -234,9 +234,11 @@ def convert_df_input_to_list_of_dicts_input(
Frequency string for timestamp generation (e.g., "H", "D", "W"). Can only be used
when validate_inputs=False. When provided, skips frequency inference from the data.
validate_inputs
[ADVANCED] When False, skips validation. You must ensure: (1) df and future_df (if provided)
are sorted by (id_column, timestamp_column); (2) future_df (if provided) contains exactly
prediction_length rows per item. Defaults to True.
[ADVANCED] When True (default), validates dataframes before prediction. Setting
to False is faster but data errors may silently lead to wrong predictions. When
False, you must ensure: (1) all dataframes are sorted by (id_column, timestamp_column);
(2) future_df (if provided) has the same item IDs as df with exactly
prediction_length rows of future timestamps per item.
Returns
-------