Skip to content

Commit

Permalink
address deprecation warnings (#69)
Browse files Browse the repository at this point in the history
  • Loading branch information
@jmoralez
jmoralez authored Mar 6, 2024
1 parent 081d5a4 commit a66dcea
Show file tree
Hide file tree
Showing 2 changed files with 33 additions and 22 deletions.
48 changes: 28 additions & 20 deletions nbs/preprocessing.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,8 @@
"import pandas as pd\n",
"\n",
"from utilsforecast.compat import DataFrame, pl, pl_DataFrame, pl_Series\n",
"from utilsforecast.processing import group_by, repeat"
"from utilsforecast.processing import group_by, repeat\n",
"from utilsforecast.validation import _is_int_dtype"
]
},
{
Expand Down Expand Up @@ -156,7 +157,7 @@
" starts = _determine_bound_pl(start, times_by_id, 'min')\n",
" ends = _determine_bound_pl(end, times_by_id, 'max')\n",
" grid = pl_DataFrame({id_col: times_by_id[id_col]})\n",
" if starts.is_integer():\n",
" if _is_int_dtype(starts):\n",
" grid = grid.with_columns(\n",
" pl.int_ranges(starts, ends + freq, step=freq, eager=True).alias(time_col)\n",
" )\n",
Expand Down Expand Up @@ -186,8 +187,10 @@
" raise NotImplementedError('Multiple of a business day')\n",
" freq = 'D'\n",
" if offset.n > 1:\n",
" freq = freq.replace(str(offset.n), '') \n",
" if not hasattr(offset, 'delta'):\n",
" freq = freq.replace(str(offset.n), '')\n",
" try:\n",
" pd.Timedelta(offset)\n",
" except ValueError:\n",
" # irregular freq, try using first letter of abbreviation\n",
" # such as MS = 'Month Start' -> 'M', YS = 'Year Start' -> 'Y'\n",
" freq = freq[0]\n",
Expand Down Expand Up @@ -241,7 +244,7 @@
"text/markdown": [
"---\n",
"\n",
"[source](https://github.com/Nixtla/utilsforecast/blob/main/utilsforecast/preprocessing.py#L55){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"[source](https://github.com/Nixtla/utilsforecast/blob/main/utilsforecast/preprocessing.py#L56){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### fill_gaps\n",
"\n",
Expand All @@ -267,7 +270,7 @@
"text/plain": [
"---\n",
"\n",
"[source](https://github.com/Nixtla/utilsforecast/blob/main/utilsforecast/preprocessing.py#L55){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"[source](https://github.com/Nixtla/utilsforecast/blob/main/utilsforecast/preprocessing.py#L56){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### fill_gaps\n",
"\n",
Expand Down Expand Up @@ -1238,8 +1241,8 @@
"data": {
"text/html": [
"<div><style>\n",
".dataframe > thead > tr > th,\n",
".dataframe > tbody > tr > td {\n",
".dataframe > thead > tr,\n",
".dataframe > tbody > tr {\n",
" text-align: right;\n",
" white-space: pre-wrap;\n",
"}\n",
Expand Down Expand Up @@ -1290,8 +1293,8 @@
"data": {
"text/html": [
"<div><style>\n",
".dataframe > thead > tr > th,\n",
".dataframe > tbody > tr > td {\n",
".dataframe > thead > tr,\n",
".dataframe > tbody > tr {\n",
" text-align: right;\n",
" white-space: pre-wrap;\n",
"}\n",
Expand Down Expand Up @@ -1342,8 +1345,8 @@
"data": {
"text/html": [
"<div><style>\n",
".dataframe > thead > tr > th,\n",
".dataframe > tbody > tr > td {\n",
".dataframe > thead > tr,\n",
".dataframe > tbody > tr {\n",
" text-align: right;\n",
" white-space: pre-wrap;\n",
"}\n",
Expand Down Expand Up @@ -1394,8 +1397,8 @@
"data": {
"text/html": [
"<div><style>\n",
".dataframe > thead > tr > th,\n",
".dataframe > tbody > tr > td {\n",
".dataframe > thead > tr,\n",
".dataframe > tbody > tr {\n",
" text-align: right;\n",
" white-space: pre-wrap;\n",
"}\n",
Expand Down Expand Up @@ -1446,8 +1449,8 @@
"data": {
"text/html": [
"<div><style>\n",
".dataframe > thead > tr > th,\n",
".dataframe > tbody > tr > td {\n",
".dataframe > thead > tr,\n",
".dataframe > tbody > tr {\n",
" text-align: right;\n",
" white-space: pre-wrap;\n",
"}\n",
Expand Down Expand Up @@ -1496,8 +1499,8 @@
"data": {
"text/html": [
"<div><style>\n",
".dataframe > thead > tr > th,\n",
".dataframe > tbody > tr > td {\n",
".dataframe > thead > tr,\n",
".dataframe > tbody > tr {\n",
" text-align: right;\n",
" white-space: pre-wrap;\n",
"}\n",
Expand Down Expand Up @@ -1617,8 +1620,13 @@
" assert max_dates[0] == expected_end\n",
"\n",
"n_periods = 100\n",
"for freq in ['Y', 'YS', 'M', 'MS', 'W', 'W-TUE', 'D', 's', 'ms', 1, 2, '20D', '30s', '2Y', '3YS', '30min', 'B']:\n",
" if isinstance(freq, str): \n",
"freqs = ['YE', 'YS', 'ME', 'MS', 'W', 'W-TUE', 'D', 's', 'ms', 1, 2, '20D', '30s', '2YE', '3YS', '30min', 'B']\n",
"try:\n",
" pd.tseries.frequencies.to_offset('YE')\n",
"except ValueError:\n",
" freqs = [f.replace('YE', 'Y').replace('ME', 'M') for f in freqs if isinstance(f, str)]\n",
"for freq in freqs:\n",
" if isinstance(freq, (pd.offsets.BaseOffset, str)): \n",
" dates = pd.date_range('1900-01-01', periods=n_periods, freq=freq)\n",
" offset = pd.tseries.frequencies.to_offset(freq)\n",
" else:\n",
Expand Down
7 changes: 5 additions & 2 deletions utilsforecast/preprocessing.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@

from .compat import DataFrame, pl, pl_DataFrame, pl_Series
from .processing import group_by, repeat
from .validation import _is_int_dtype

# %% ../nbs/preprocessing.ipynb 4
def _determine_bound(bound, freq, times_by_id, agg) -> np.ndarray:
Expand Down Expand Up @@ -100,7 +101,7 @@ def fill_gaps(
starts = _determine_bound_pl(start, times_by_id, "min")
ends = _determine_bound_pl(end, times_by_id, "max")
grid = pl_DataFrame({id_col: times_by_id[id_col]})
if starts.is_integer():
if _is_int_dtype(starts):
grid = grid.with_columns(
pl.int_ranges(starts, ends + freq, step=freq, eager=True).alias(
time_col
Expand Down Expand Up @@ -133,7 +134,9 @@ def fill_gaps(
freq = "D"
if offset.n > 1:
freq = freq.replace(str(offset.n), "")
if not hasattr(offset, "delta"):
try:
pd.Timedelta(offset)
except ValueError:
# irregular freq, try using first letter of abbreviation
# such as MS = 'Month Start' -> 'M', YS = 'Year Start' -> 'Y'
freq = freq[0]
Expand Down

0 comments on commit a66dcea

Please sign in to comment.