1

I have two different CSVs with time series data. One Table is continuous, starting in 01.01.2017 at 00:00. From there each row represents one hour (1. Table). The data looks kind of like this:

  1. Table aka df1:
Date,                   Volume
2017-02-03 12-PM,       9787.51
2017-02-03 01-PM,       9792.01
2017-02-03 02-PM,       9803.94
2017-02-03 03-PM,       9573.99

The other table contains events that happened and are serialized by UNIX datetime in seconds. I was able to convert it to datetime and group it by hour with this code:

df['datetime'] = pd.to_datetime(df['created_utc'], unit='s')
df['datetime'] = pd.to_datetime(df['datetime'], format="%Y-%m-%d %I-%p")
df['date_by_hour'] = df['datetime'].apply(lambda x: x.strftime('%Y-%m-%d %H:00'))

This resulted in this data:

  1. Table aka df2:
created_utc,    score,      compound,   datetime,               date_by_hour
1486120391,     156,        0.125,      2017-02-03 12:13:11,    2017-02-03 12:00:00
1486125540,     1863,       0.475,      2017-02-03 13:39:00,    2017-02-03 13:00:00
1486126013,     863,        0.889,      2017-02-03 13:46:53,    2017-02-03 13:00:00
1486130203,     23,         0.295,      2017-02-03 14:56:43,    2017-02-03 14:00:00

Now I need to map the events (2.table) to the Time Series of the 1. Table. If multiple events happened in one hour, i need to make an addition of the scores and calculate the mean average of the compound. In the end i want to have a dataframe like this:

  1. Final Dataframe
Date,                   Volume,         score,      compound,
2017-02-03 12-PM,       9787.51,        156,        0.125,
2017-02-03 01-PM,       9792.01,        2726,       0.682,
2017-02-03 02-PM,       9803.94,        23,         0.295,
2017-02-03 03-PM,       9573.99,        0,          0,

I know my code below does not work and is wrong, but I wanted to show what I was thinking how I could achieve this. I thought I could loop through each row of my events table df2 and compare if the datetime matches. If so, I would calculate score and compound. The issue is that I know that one should not loop through a dataframe and I don't know how to loop through another dataframe at the same time and perform the right calculations based on the previous rows...

for index, row in df2.iterrows():
    memory_score = 0
    memory_compound = 0
    if df1['Date'] == df2['date_by_hour']:
        df1['score'] = row['score'] + memory_score
        df1['compound'] = (row['compound'] + memory_compound) / 2

How can I get to my Final Dataframe? There must be some pandas magic that I could use to make this work and map the time series data to the right hours.

Thanks in advance and best regards

EDIT: THere can be an undefined number of events within one hour. I just chose 2 for this simple example but it could be 5000 or something or 0 in some cases.

LGR
  • 402
  • 3
  • 18

1 Answers1

2
# import necessary packages, set seed

import pandas as pd
import datetime
import random
random.seed(42)

Setup

Let's create an example df1, using this:

numdays=5
base = datetime.datetime(2017,2,3,12)
date_list = [base + datetime.timedelta(hours=x) for x in range(numdays)]

Then, using:

df1 = pd.DataFrame.from_dict({'Date': date_list,
                              'Volume': [random.randint(9000,11000) for _ in range(len(date_list))]})

This gives us:

+----+---------------------+----------+
|    | Date                |   Volume |
|----+---------------------+----------|
|  0 | 2017-02-03 12:00:00 |     9228 |
|  1 | 2017-02-03 13:00:00 |     9051 |
|  2 | 2017-02-03 14:00:00 |    10518 |
|  3 | 2017-02-03 15:00:00 |     9563 |
|  4 | 2017-02-03 16:00:00 |     9501 |
+----+---------------------+----------+

Let's create df2 as well:

random_date_list = [base + datetime.timedelta(hours=x*random.uniform(0,2)) for x in range(7)]

df2 = pd.DataFrame({'datetime':random_date_list,
                   'score':[random.randint(20,200) for _ in range(len(random_date_list))],
                   'compound': [random.uniform(0,1) for _ in range(len(random_date_list))]},
                   index=[x for x in range(len(random_date_list))])

which gives:

+----+----------------------------+---------+------------+
|    | datetime                   |   score |   compound |
|----+----------------------------+---------+------------|
|  0 | 2017-02-03 12:00:00        |      75 |   0.71602  |
|  1 | 2017-02-03 13:28:22.592742 |      79 |   0.701325 |
|  2 | 2017-02-03 14:42:24.472619 |     149 |   0.41952  |
|  3 | 2017-02-03 17:21:11.078662 |     174 |   0.449209 |
|  4 | 2017-02-03 12:41:43.838380 |      26 |   0.278191 |
|  5 | 2017-02-03 16:13:09.185509 |     163 |   0.8693   |
|  6 | 2017-02-03 12:21:27.239880 |      70 |   0.758807 |
+----+----------------------------+---------+------------+

The actual calculation

Let's create a column in df2 with datetime objects to hour resolution:

df2['Date'] = df2['datetime'].apply(lambda x: x.replace(minute=0, second=0, microsecond=0))

We can merge df1 & df2, replace NaNs with 0s:

merged = pd.merge(df1,df2,on='Date', how='outer')
merged.fillna(0,inplace=True)

Now calculate the required new columns:

newscoredf=merged.groupby('Date')[['score']].agg('sum')
newcompounddf=merged.groupby('Date')[['compound']].agg('mean')

Let's concat them, & add the Volume column as well which we left behind:

final = pd.concat([df1.set_index('Date')[['Volume']],newscoredf,newcompounddf],axis=1)

which gives you what you want.

final:

+---------------------+----------+---------+------------+
| Date                |   Volume |   score |   compound |
|---------------------+----------+---------+------------|
| 2017-02-03 12:00:00 |     9228 |     171 |   0.584339 |
| 2017-02-03 13:00:00 |     9051 |      79 |   0.701325 |
| 2017-02-03 14:00:00 |    10518 |     149 |   0.41952  |
| 2017-02-03 15:00:00 |     9563 |       0 | nan        |
| 2017-02-03 16:00:00 |     9501 |     163 |   0.8693   |
| 2017-02-03 17:00:00 |      nan |     174 |   0.449209 |
+---------------------+----------+---------+------------+

Checking: consider the rows which happened in an hour starting with 12. Score: Score: 75+26+70 = 171. Compound: (0.71602 + 0.278191 + 0.758807) / 3 = 0.584339. Both are in agreement with the first row of our result.

zabop
  • 6,750
  • 3
  • 39
  • 84