Python For Loop Create DataFrame

Question

Hihi, I'm reasonably new to Python, more a R guy. But I'm required to use python for a task.

However, I encountered a situation where I need to subset data into different dataFrame by the date. "in future not 3 exactly, therefore want to create a loop to do this" I want to potentially create 3 dataframes like

df_train_30 which contains start day  0 - 30
df_train_60 which contains start day 30 - 60
df_train_90 which contains start day 60 - 90

but not sure how to achieve this... pls help.

idea in code below

today = pd.to_datetime('now')
df_train['START_DATE'] = pd.to_datetime(df_train['START_DATE'])
previous_day_del = 0

for day_del in (30,60,90):
    **'df_train_' + str(day_del)** = df_train[(df_train['START_DATE']>= today - timedelta(days=day_del)) & (df_train['START_DATE']< today - timedelta(days=previous_day_del))]
    previous_day_del = day_del

kindly post your expected output. I'm guessing it has to be as a dataframe? — sammywemmy
– sammywemmy, Commented Sep 2, 2020 at 5:53
yes, my expected output is having 3 different data frames like 1) df_train_30: contains results of START_DATE between SYSDATE - 30 to SYSDATE, 2) df_train_60: contains results of START_DATE between SYSDATE - 60 to SYSDATE -30, 3) df_train_90: contains results of START_DATE between SYSDATE - 90 to SYSDATE -60 — humphrey chan
– humphrey chan, Commented Sep 2, 2020 at 5:57

Arjun Sohanlal · Accepted Answer · 2020-09-02 06:00:20Z

1

You could probably store it into a dictionary - it's easier to manage rather than dynamically generated variables. A dictionary's more of an object with key-value pairs, and the values can be of almost any type, including even dataframes. Here's a quick guide on Python dictionaries that you could look at.

In your example, you could probably go ahead with this:

today = pd.to_datetime('now')
df_train['START_DATE'] = pd.to_datetime(df_train['START_DATE'])
previous_day_del = 0
   
# Creating an empty dictionary here called dict_train
dict_train = {}

for day_del in (30,60,90):
    dict_train[day_del] = df_train[(df_train['START_DATE']>= today -timedelta(days=day_del)) & (df_train['START_DATE']< today - timedelta(days=previous_day_del))]
    previous_day_del = day_del

Hope this helps, cheers! :)

answered Sep 2, 2020 at 6:00

Arjun Sohanlal

213 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

humphrey chan Over a year ago

thousand thanks! Yes, exactly what I needed. thanks for the dictionaries link. was wondering how to turn dictionary back to df but seems found in there.

Collectives™ on Stack Overflow

Python For Loop Create DataFrame

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related