Plotting multiple columns with Python from CSV file

Question

I have a csv file with 25 columns and 90 rows. The first column is in date format such that it appears as 01-10-2014 for example. The rest of the columns contain numbers that correspond to those dates. The first row is the name for each column. My plan was to write a code that took in all the information and gave me the flexibility to plot any two columns against each other as a scatter xy plot. I have copied my code so far below.

The code works fine if I remove 'weekly' and 'f8' from the dtype. It plots the date in the X-axis and the first column 'daily' in the Y-axis. I assumed I could then just add additional columns like I did with the 'weekly' as shown below. However, I get the error:

Traceback (most recent call last):
  File "plot_dataset_second.py", line 10, in <module>
    Date = [DT.datetime.strptime(key,"%d-%m-%Y") for (key, value) in data]
ValueError: too many values to unpack

Sample data:

Date    Daily Installs  Weekly Installs Mean Install Duration
01-10-14    153 153 47.71
02-10-14    630 783 51.9
03-10-14    50  833 49.94
04-10-14    973 1805    51.43

import numpy as np
import matplotlib.pyplot as plt
import datetime as DT

data= np.genfromtxt('dataset1_changed.csv', delimiter=',',
    dtype={'names': ('Date', 'daily', 'weekly'),'formats': ('S10', 'f8', 'f8')} )



Date = [DT.datetime.strptime(key,"%d-%m-%Y") for (key, value) in data]
daily = [value for (key, value) in data]
weekly = [value for (key, value) in data]
#y = [value for (key, value) in data]

fig = plt.figure()
ax = fig.add_subplot(111)
ax.grid()

fig.autofmt_xdate()

plt.plot(Date,daily,'b--o--')
plt.xlabel('Date')
plt.ylabel('Daily Count')
plt.title('Daily Count since February')
plt.show()

Have you considered using `pandas`? It would be *ideal* for this kind of thing. — Ffisegydd, Dec 09 '14 at 14:17
Please post a few rows and columns of the data including the header row. — wwii, Dec 09 '14 at 14:31
@wwii I have added a small sample of the table, hope it comes out ok. — Marmo, Dec 09 '14 at 14:45
@Ffisegydd I haven't used pandas before but if you can show me an example, I am happy to try it. — Marmo, Dec 09 '14 at 14:46

score 0 · Answer 1 · answered Dec 09 '14 at 23:49

I was able to find a work around that works for my own problem above. It is not elegant but does the job. Thank you to everyone for the comments and help. Here is the code for anyone interested:

import csv
import datetime as dt
import matplotlib.pyplot as plt


Date,Daily_Installs,Weekly_Installs = [],[],[]
csv_reader = csv.reader(open('dataset1_changed.csv', 'rU'))
for line in csv_reader:
    Date.append(dt.datetime.strptime(line[0],'%d-%m-%Y'))
    Daily_Installs.append(int(line[1]))
    Weekly_Installs.append(int(line[2]))

fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(Date,Daily_Installs,'o-')
fig.autofmt_xdate()

plt.show()

Plotting multiple columns with Python from CSV file

1 Answers1