Skip to content

Commit a680f96

Browse files
solegallisolegalli
authored
fixes wording in datetime user guide (#356)
1 parent e6a13f8 commit a680f96

File tree

1 file changed

+60
-60
lines changed

1 file changed

+60
-60
lines changed

‎docs/user_guide/datetime/DatetimeFeatures.rst‎

Lines changed: 60 additions & 60 deletions
Original file line numberDiff line numberDiff line change
@@ -57,7 +57,7 @@ second datetime variable in our dataset.
5757
5858
dtfs = DatetimeFeatures(
5959
variables="var_date2",
60-
features_to_extract=["month", "month_end", "day_of_the_year"]
60+
features_to_extract=["month", "month_end", "day_of_year"]
6161
)
6262
6363
df_transf = dtfs.fit_transform(toy_df)
@@ -66,11 +66,11 @@ second datetime variable in our dataset.
6666
6767
.. code:: python
6868
69-
var_date1 var_date2_month var_date2_month_end var_date2_doty
70-
0 May-1989 6 0 173
71-
1 Dec-2020 2 0 41
72-
2 Jan-1999 8 0 215
73-
3 Feb-2002 10 1 305
69+
var_date1 var_date2_month var_date2_month_end var_date2_day_of_year
70+
0 May-1989 6 0 173
71+
1 Dec-2020 2 0 41
72+
2 Jan-1999 8 0 215
73+
3 Feb-2002 10 1 305
7474
7575
7676
With `transform()`, the features extracted from the datetime variable are added to the
@@ -214,25 +214,25 @@ And now we mistakenly extract only date features.
214214
.. code:: python
215215
216216
dfts = DatetimeFeatures(
217-
features_to_extract=["year", "month", "day_of_the_week"],
217+
features_to_extract=["year", "month", "day_of_week"],
218218
)
219219
df_transf = dfts.fit_transform(toy_df)
220220
221221
print(df_transf)
222222
223223
.. code:: python
224224
225-
not_a_dt var_time1_year var_time1_month var_time1_dotw var_time2_year \
226-
0 not 2021 12 2 2021
227-
1 a 2021 12 2 2021
228-
2 date 2021 12 2 2021
229-
3 time 2021 12 2 2021
225+
not_a_dt var_time1_year var_time1_month var_time1_day_of_week var_time2_year \
226+
0 not 2021 12 2 2021
227+
1 a 2021 12 2 2021
228+
2 date 2021 12 2 2021
229+
3 time 2021 12 2 2021
230230
231-
var_time2_month var_time2_dotw
232-
0 12 2
233-
1 12 2
234-
2 12 2
235-
3 12 2
231+
var_time2_month var_time2_day_of_week
232+
0 12 2
233+
1 12 2
234+
2 12 2
235+
3 12 2
236236
237237
The transformer will still create features derived from today's date (the date of
238238
creating the docs).
@@ -313,15 +313,15 @@ To do this, we leave the parameter `features_to_extract` to `None`.
313313
1 2018-01-01 01:00:00 12/01/90 23:01:02 02/28/97 10:10:55 1
314314
2 2018-01-01 02:00:00 04/25/01 11:59:21 11/11/03 17:30:00 1
315315
316-
var_dt1_year var_dt1_dotw var_dt1_dotm var_dt1_hour var_dt1_minute \
317-
0 2018 0 1 0 0
318-
1 2018 0 1 1 0
319-
2 2018 0 1 2 0
316+
var_dt1_year var_dt1_day_of_week var_dt1_day_of_month var_dt1_hour \
317+
0 2018 0 1 0
318+
1 2018 0 1
319+
2 2018 0 1 2
320320
321-
var_dt1_second
322-
0 0
323-
1 0
324-
2 0
321+
var_dt1_minute var_dt1_second
322+
0 0 0
323+
1 0 0
324+
2 0 0
325325
326326
Our new dataset contains the original features plus the new variables extracted
327327
from them.
@@ -336,8 +336,8 @@ We can find the group of features extracted by the transformer in its attribute.
336336
337337
['month',
338338
'year',
339-
'day_of_the_week',
340-
'day_of_the_month',
339+
'day_of_week',
340+
'day_of_month',
341341
'hour',
342342
'minute',
343343
'second']
@@ -368,10 +368,10 @@ We can also extract all supported features automatically.
368368
1 1 1 2018
369369
2 1 1 2018
370370
371-
var_dt1_woty var_dt1_dotw ... var_dt1_month_end var_dt1_quarter_start \
372-
0 1 0 ... 0 1
373-
1 1 0 ... 0 1
374-
2 1 0 ... 0 1
371+
var_dt1_week var_dt1_day_of_week ... var_dt1_month_end var_dt1_quarter_start \
372+
0 1 0 ... 0 1
373+
1 1 0 ... 0 1
374+
2 1 0 ... 0 1
375375
376376
var_dt1_quarter_end var_dt1_year_start var_dt1_year_end \
377377
0 0 1 0
@@ -400,10 +400,10 @@ We can find the group of features extracted by the transformer in its attribute.
400400
'quarter',
401401
'semester',
402402
'year',
403-
'week_of_the_year',
404-
'day_of_the_week',
405-
'day_of_the_month',
406-
'day_of_the_year',
403+
'week',
404+
'day_of_week',
405+
'day_of_month',
406+
'day_of_year',
407407
'weekend',
408408
'month_start',
409409
'month_end',
@@ -467,29 +467,29 @@ from the dataset.
467467
468468
.. code:: python
469469
470-
var_date_month var_date_year var_date_dotw var_date_dotm \
471-
0 6 2012 3 21
472-
1 2 1998 1 10
473-
2 8 2010 1 3
474-
3 10 2020 5 31
470+
var_date_month var_date_year var_date_day_of_week var_date_day_of_month \
471+
0 6 2012 3 21
472+
1 2 1998 1 10
473+
2 8 2010 1 3
474+
3 10 2020 5 31
475475
476476
var_time1_hour var_time1_minute var_time1_second var_dt_month \
477477
0 12 34 45 8
478478
1 23 1 2 12
479479
2 11 59 21 4
480480
3 8 44 23 4
481481
482-
var_dt_year var_dt_dotw var_dt_dotm var_dt_hour var_dt_minute \
483-
0 2000 3 31 12 34
484-
1 1990 5 1 23 1
485-
2 2001 2 25 11 59
486-
3 2001 2 25 11 59
482+
var_dt_year var_dt_day_of_week var_dt_day_of_month var_dt_hour \
483+
0 2000 3 31 12
484+
1 1990 5 1 23
485+
2 2001 2 25 11
486+
3 2001 2 25 11
487487
488-
var_dt_second
489-
0 45
490-
1 2
491-
2 21
492-
3 21
488+
var_dt_minute var_dt_second
489+
0 34 45
490+
1 1 2
491+
2 59 21
492+
3 59 21
493493
494494
As you can see, we do not have the constant features in the transformed dataset.
495495

@@ -562,7 +562,7 @@ timezone.
562562
toy_df = pd.DataFrame({"var_tz": var_tz})
563563
564564
dfts = DatetimeFeatures(
565-
features_to_extract=["day_of_the_month", "hour"],
565+
features_to_extract=["day_of_month", "hour"],
566566
drop_original=False,
567567
utc=True,
568568
)
@@ -573,10 +573,10 @@ timezone.
573573
574574
.. code:: python
575575
576-
var_tz var_tz_dotm var_tz_hour
577-
0 2000-08-31 12:34:45-04:00 31 16
578-
1 1990-12-01 23:01:02-05:00 2 4
579-
2 2001-04-25 11:59:21-04:00 25 15
576+
var_tz var_tz_day_of_month var_tz_hour
577+
0 2000-08-31 12:34:45-04:00 31 16
578+
1 1990-12-01 23:01:02-05:00 2 4
579+
2 2001-04-25 11:59:21-04:00 25 15
580580
581581
582582
**Case 3**: given a variable like *var_tz* in the example above, we now want
@@ -587,7 +587,7 @@ is the default option.
587587
.. code:: python
588588
589589
dfts = DatetimeFeatures(
590-
features_to_extract=["day_of_the_month", "hour"],
590+
features_to_extract=["day_of_month", "hour"],
591591
drop_original=False,
592592
utc=None,
593593
)
@@ -598,10 +598,10 @@ is the default option.
598598
599599
.. code:: python
600600
601-
var_tz var_tz_dotm var_tz_hour
602-
0 2000-08-31 12:34:45-04:00 31 12
603-
1 1990-12-01 23:01:02-05:00 1 23
604-
2 2001-04-25 11:59:21-04:00 25 11
601+
var_tz var_tz_day_of_month var_tz_hour
602+
0 2000-08-31 12:34:45-04:00 31 12
603+
1 1990-12-01 23:01:02-05:00 1 23
604+
2 2001-04-25 11:59:21-04:00 25 11
605605
606606
Note that the hour extracted from the variable differ in this dataframe respect to the
607607
one obtained in **Case 2**.

0 commit comments

Comments
 (0)