Skip to content

🐛 FillGapsAR fix interpolation and boundaries effect when resampling … #8

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

Merged
BaptisteDE merged 1 commit into from
Jan 9, 2025
Merged

🐛 FillGapsAR fix interpolation and boundaries effect when resampling … #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 2 additions & 0 deletions tests/test_processing.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -656,6 +656,8 @@ def test_pd_fill_gap(self):
toy_df_15min_hole = toy_df_15min.copy()
toy_df_15min_hole.loc[hole_backast, "Temp_1"] = np.nan
toy_df_15min_hole.loc[hole_forecast, "Temp_1"] = np.nan
toy_df_15min_hole.iloc[:12, 0] = np.nan
toy_df_15min_hole.iloc[-12:, 0] = np.nan

filler = FillGapsAR(resample_at_td="1h")
res = filler.fit_transform(toy_df_15min_hole.copy())
Expand Down
27 changes: 21 additions & 6 deletions tide/processing.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1155,11 +1155,18 @@ def _fit_and_fill_x(self, X, biggest_group, col, idx, backcast):
if self.resample_at_td is not None:
x_fit = X.loc[biggest_group, col].resample(self.resample_at_td).mean()
idx_origin = idx
idx = pd.date_range(idx[0], idx[-1], freq=self.resample_at_td)
if not backcast and x_fit.index[-1] == idx[0]:
x_fit = x_fit[:-1]
elif x_fit.index[0] == idx[-1]:
x_fit = x_fit[1:]
if backcast:
idx = pd.date_range(
idx[0],
x_fit.index[0] - pd.Timedelta(self.resample_at_td),
freq=self.resample_at_td,
)
else:
idx = pd.date_range(
x_fit.index[-1] + pd.Timedelta(self.resample_at_td),
idx[-1],
freq=self.resample_at_td,
)
else:
x_fit = X.loc[biggest_group, col]
idx_origin = None
Expand All @@ -1169,7 +1176,15 @@ def _fit_and_fill_x(self, X, biggest_group, col, idx, backcast):
to_predict.name = col
X.loc[idx, col] = bc_model.predict(to_predict).to_numpy().flatten()
if self.resample_at_td is not None:
X.loc[idx_origin, col] = X.loc[idx_origin, col].interpolate()
beg = idx_origin[0] - idx_origin.freq
end = idx_origin[-1] + idx_origin.freq
# Interpolate linearly between inferred values and using neighbor data
X.loc[idx_origin, col] = X.loc[beg:end, col].interpolate()
# If gap is at boundaries
if beg < X.index[0]:
X.loc[idx_origin, col] = X.loc[idx_origin, col].bfill()
if end > X.index[-1]:
X.loc[idx_origin, col] = X.loc[idx_origin, col].ffill()

def _fit_implementation(self, X: pd.Series | pd.DataFrame, y=None):
self.model_ = MODEL_MAP[self.model_name]
Expand Down
Loading