Pandas Dataframe not populating in loop

Question

I'm creating the dataframe df_stats and I want to fill it with variables from each t for t in t_list. When I run this df_stats doesn't populate with values but if I run the line df_stats.append({... independantly it populates one row of data with the values if the current t. What am I missing to populate df_statswith a row of data from each t1 int_list`?

import pandas as pd
import matplotlib.pyplot as plt
from datetime import datetime, timedelta
import glob

#add all data files into large df so all dates are acessible
path = 'C:\Users\data' # use your path
all_files = glob.glob(path + "/*.csv")
li = []
for filename in all_files:
    df_data = pd.read_csv(filename, index_col=None, header=0)
    li.append(df_data)

df_data = pd.concat(li, axis=0, ignore_index=True)
df_data['datetime'] = pd.to_datetime(df_data['TimeStamp'] )
df = df_data[(df_data['datetime']>= datetime(2017, 11,9, 00,00, 00)) &
         (df_data['datetime']< datetime(2017, 11, 9, 23,50, 00))]

##want a time array for all of the datetimes in the df
t_list = df.groupby("datetime").all().index

df_stats = pd.DataFrame(columns = ['t', 'min_ws', 'max_ws', 'mean_ws','stdev_ws',
 'TI_var_ws', 'min_power', 'max_power', 'mean_power', 'stdev_pwr', 'TI_var_pwr'])

for t in t_list:
    df_t = df[(df['datetime']>=t) & (df['datetime']<t_end)]

    #calc min/max for setting scale on images
    t = t
    min_ws = df['wtc_AcWindSp_mean'].min()
    max_ws = df['wtc_AcWindSp_mean'].max()
    mean_ws = df['wtc_AcWindSp_mean'].mean()
    stdev_ws = df['wtc_AcWindSp_mean'].std()
    TI_var_ws = stdev_ws/mean_ws

    min_power = df['wtc_ActPower_mean'].min()
    max_power = df['wtc_ActPower_mean'].max()
    mean_power = df_t['wtc_ActPower_mean'].mean()
    stdev_pwr = df_t['wtc_ActPower_mean'].std()
    TI_var_pwr = stdev_pwr/mean_power

    df_stats.append({'t':t, 'min_ws':min_ws, 'max_ws':max_ws, 'mean_ws':mean_ws,'stdev_ws':stdev_ws,
    'TI_var_ws':TI_var_ws, 'min_power':min_power,...'max_power':max_power, 'mean_power': mean_power,
    'stdev_pwr':stdev_pwr, 'TI_var_pwr':TI_var_pwr}, ignore_index=True)

[Never call `DataFrame.append` or `pd.concat` inside a for-loop. It leads to quadratic copying.](https://stackoverflow.com/a/36489724/1422451) — Parfait, Dec 20 '19 at 20:58
What is *t_end* and is it a typo that use `df` and not `df_t` in all of the `for` loop aggregates. — Parfait, Dec 20 '19 at 21:03

fsl · Accepted Answer · 2019-12-20T20:55:15.747

You need to reassign the DataFrame as append always returns a new object:

https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.append.html

df_stats = df_stats.append({'t':t, 'min_ws':min_ws, 'max_ws':max_ws, 'mean_ws':mean_ws,'stdev_ws':stdev_ws,
    'TI_var_ws':TI_var_ws, 'min_power':min_power,...'max_power':max_power, 'mean_power': mean_power,
    'stdev_pwr':stdev_pwr, 'TI_var_pwr':TI_var_pwr}, ignore_index=True)

That said, you are likely better off creating the index from scratch, something like:

# Pass an index argument
df_stats = pd.DataFrame(index=range(len(t_list)), columns = ['t', 'min_ws', 'max_ws', 'mean_ws','stdev_ws',
 'TI_var_ws', 'min_power', 'max_power', 'mean_power', 'stdev_pwr', 'TI_var_pwr'])

# ...

for i, t in enumerate(t_list):

    # ...

    df.iloc[i] = {'t':t, 'min_ws':min_ws, 'max_ws':max_ws, 'mean_ws':mean_ws,'stdev_ws':stdev_ws,
    'TI_var_ws':TI_var_ws, 'min_power':min_power,...'max_power':max_power, 'mean_power': mean_power,
    'stdev_pwr':stdev_pwr, 'TI_var_pwr':TI_var_pwr}

Thank you for your helpful answer, I chose the `iloc` method and forgot to accept the answer, apologies! — squar_o, May 20 '20 at 11:51

Pandas Dataframe not populating in loop

1 Answers1