Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Refactor/hist fc general #1801

Merged
merged 10 commits into from
May 28, 2023
Merged

Refactor/hist fc general #1801

Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
a268974
simple time reduction from removing redundant slicing
dennisbader Apr 19, 2023
3a229be
move the min_start_time truncation out of the loop
dennisbader Apr 19, 2023
78cbb32
replace self by model to allow for moving historical forecasts out of…
dennisbader Apr 19, 2023
9984680
replace series.drop_*() with series.slice() for performance boost
dennisbader Apr 19, 2023
b632542
Merge branch 'master' into refactor/historical_forecast
dennisbader May 27, 2023
7d20683
make same state as test
dennisbader May 27, 2023
2ad3e3a
optimize some time series methods
dennisbader May 27, 2023
4cc7d88
used improved splitting in historical forecasts
dennisbader May 27, 2023
d38b8f9
Merge branch 'master' into refactor/hist_fc_general
dennisbader May 27, 2023
2373f29
Update CHANGELOG.md
dennisbader May 28, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
replace series.drop_*() with series.slice() for performance boost
  • Loading branch information
@dennisbader
dennisbader committed Apr 19, 2023
commit 998468068dec1f63419e7c2921c185c57b5dd718
4 changes: 2 additions & 2 deletions darts/models/forecasting/forecasting_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1006,7 +1006,7 @@ def retrain_func(

# adjust the start of the series depending on whether we train (at some point), or predict only
if min_timestamp_series > series_.time_index[0]:
series_ = series_.drop_before(min_timestamp_series - 1 * series_.freq)
series_ = series_.slice(min_timestamp_series, series_.end_time())

if len(series) == 1:
# Only use tqdm if there's no outer loop
Expand All @@ -1026,7 +1026,7 @@ def retrain_func(
# iterate and forecast
for _counter, pred_time in enumerate(iterator):
# drop everything after `pred_time` to train on / predict with shifting input
train_series = series_.drop_after(pred_time)
train_series = series_.slice(min_timestamp_series, pred_time - series_.freq)

# optionally, apply moving window (instead of expanding window)
if train_length_ and len(train_series) > train_length_:
Expand Down