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

[FEAT] NeuralForecast Compatibility and Example Notebook #188

Merged
merged 8 commits into from
Apr 24, 2023
Merged

[FEAT] NeuralForecast Compatibility and Example Notebook #188

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
2bd67df
added median checking in reconcile
dluuo Apr 24, 2023
04a51bc
updated drop_cols logic
dluuo Apr 24, 2023
1e7c300
Merge branch 'Nixtla:main' into main
dluuo Apr 24, 2023
28cd35c
added MLFrameworksExample notebook
dluuo Apr 24, 2023
026b639
Merge branch 'main' of https://github.com/dluuo/hierarchicalforecast
dluuo Apr 24, 2023
a049808
updated MLFrameworksExamble nb documentation
dluuo Apr 24, 2023
fe355e0
removed extra -median check
dluuo Apr 24, 2023
e2438a7
updated sidebar.yml with new example nb
dluuo Apr 24, 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
11 changes: 8 additions & 3 deletions hierarchicalforecast/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,9 +45,14 @@ def _reverse_engineer_sigmah(Y_hat_df, y_hat, model_name):
In the future, we might deprecate this function in favor of a
direct usage of an estimated $\hat{sigma}_{h}$
"""
drop_cols = ['ds', 'y'] if 'y' in Y_hat_df.columns else ['ds']

drop_cols = ['ds']
if 'y' in Y_hat_df.columns:
drop_cols.append('y')
if model_name+'-median' in Y_hat_df.columns:
drop_cols.append(model_name+'-median')
model_names = Y_hat_df.drop(columns=drop_cols, axis=1).columns.to_list()
pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name)]
pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name or '-median' in name)]
dluuo marked this conversation as resolved.
Show resolved Hide resolved
pi_model_name = [pi_name for pi_name in pi_model_names if model_name in pi_name]
pi = len(pi_model_name) > 0

Expand Down Expand Up @@ -143,7 +148,7 @@ def _prepare_fit(self,
if Y_hat_df[model_names].isnull().values.any():
raise Exception('`Y_hat_df` contains null values')

pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name)]
pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name or '-median' in name)]
model_names = [name for name in model_names if name not in pi_model_names]

# TODO: Complete y_hat_insample protection
Expand Down
11 changes: 8 additions & 3 deletions nbs/core.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,9 +138,14 @@
" In the future, we might deprecate this function in favor of a \n",
" direct usage of an estimated $\\hat{sigma}_{h}$\n",
" \"\"\"\n",
" drop_cols = ['ds', 'y'] if 'y' in Y_hat_df.columns else ['ds']\n",
"\n",
" drop_cols = ['ds']\n",
" if 'y' in Y_hat_df.columns:\n",
" drop_cols.append('y')\n",
" if model_name+'-median' in Y_hat_df.columns:\n",
" drop_cols.append(model_name+'-median')\n",
" model_names = Y_hat_df.drop(columns=drop_cols, axis=1).columns.to_list()\n",
" pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name)]\n",
" pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name or '-median' in name)]\n",
" pi_model_name = [pi_name for pi_name in pi_model_names if model_name in pi_name]\n",
" pi = len(pi_model_name) > 0\n",
"\n",
Expand Down Expand Up @@ -243,7 +248,7 @@
" if Y_hat_df[model_names].isnull().values.any():\n",
" raise Exception('`Y_hat_df` contains null values')\n",
" \n",
" pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name)]\n",
" pi_model_names = [name for name in model_names if ('-lo' in name or '-hi' in name or '-median' in name)]\n",
" model_names = [name for name in model_names if name not in pi_model_names]\n",
" \n",
" # TODO: Complete y_hat_insample protection\n",
Expand Down
Loading