matplotlib mySQL空白图

Question

I made this Python script. The idea is using this Python script so that matplotlib can generate a vertical bar chart for me using data from a MySQL database. Another thing I want to do is saving the png file instead of displaying it (I am trying to run this code via putty), which I think I did it correctly. When I run the code, I got some error which I have no idea what it means. Hopefully someone can point out where I did wrong.

#!/usr/bin/python
# matplotlib pyplot module
import MySQLdb
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# connect to MySQL database
conn = MySQLdb.connect(host="localhost", user="xx", passwd="xxxx", db="r1array")

# prepare a cursor
cur = conn.cursor()

# this is the query we will be making
query = """
SELECT TimeStamp,Pac 
FROM SolarData 
WHERE TimeStamp >= "2014-01-03" 
  AND TimeStamp < "2014-01-04";
"""

# execute the query
cur.execute(query)

# retrieve the whole result set
data = cur.fetchall()

# close cursor and connection
cur.close()
conn.close()

# unpack data in TimeStamp (x axis) and Pac (y axis)
TimeStamp, Pac = zip(*data)

# graph code
plt.bar(TimeStamp, Pac, align='center')

# set title, X/Y labels
plt.title("PVIC R1 panel")
plt.xlabel("Time of Day")
plt.ylabel("Pac")
fig = plt.gcf()

# plt.xticks(TimeStamp, (hour))
fig.set_size_inches(20.5,10.5)
plt.grid(True)
plt.draw()
fig.savefig('test2.png', dpi=100)

The error message I got is

Traceback (most recent call last):
      File "python_sql_image.py", line 41, in <module>
        plt.savefig('test.png')
      File "/usr/lib/python2.6/site-packages/matplotlib/pyplot.py", line 356, in savefig
        return fig.savefig(*args, **kwargs)
      File "/usr/lib/python2.6/site-packages/matplotlib/figure.py", line 1032, in savefig
        self.canvas.print_figure(*args, **kwargs)
      File "/usr/lib/python2.6/site-packages/matplotlib/backend_bases.py", line 1476, in print_figure
        **kwargs)
      File "/usr/lib/python2.6/site-packages/matplotlib/backends/backend_agg.py", line 358, in print_png
        FigureCanvasAgg.draw(self)
      File "/usr/lib/python2.6/site-packages/matplotlib/backends/backend_agg.py", line 314, in draw
        self.figure.draw(self.renderer)
      File "/usr/lib/python2.6/site-packages/matplotlib/artist.py", line 46, in draw_wrapper
        draw(artist, renderer, *kl)
      File "/usr/lib/python2.6/site-packages/matplotlib/figure.py", line 773, in draw
        for a in self.axes: a.draw(renderer)
      File "/usr/lib/python2.6/site-packages/matplotlib/artist.py", line 46, in draw_wrapper
        draw(artist, renderer, *kl)
      File "/usr/lib/python2.6/site-packages/matplotlib/axes.py", line 1735, in draw
        a.draw(renderer)
      File "/usr/lib/python2.6/site-packages/matplotlib/artist.py", line 46, in draw_wrapper
        draw(artist, renderer, *kl)
      File "/usr/lib/python2.6/site-packages/matplotlib/axis.py", line 736, in draw
        for tick, loc, label in self.iter_ticks():
      File "/usr/lib/python2.6/site-packages/matplotlib/axis.py", line 680, in iter_ticks
        majorLabels = [self.major.formatter(val, i) for i, val in enumerate(majorLocs)]
      File "/usr/lib/python2.6/site-packages/matplotlib/dates.py", line 417, in __call__
        return self._formatter(x, pos)
      File "/usr/lib/python2.6/site-packages/matplotlib/dates.py", line 293, in __call__
        dt = num2date(x, self.tz)
      File "/usr/lib/python2.6/site-packages/matplotlib/dates.py", line 249, in num2date
        if not cbook.iterable(x): return _from_ordinalf(x, tz)
      File "/usr/lib/python2.6/site-packages/matplotlib/dates.py", line 170, in _from_ordinalf
        dt = datetime.datetime.fromordinal(ix)
    ValueError: ordinal must be >= 1

especially the last line, I don't get this.

ValueError: ordinal must be >= 1

This is the data I called from MySQL database.

 TimeStamp           | Pac  |
+---------------------+------+
| 2014-01-03 07:55:02 |    0 |
| 2014-01-03 08:00:02 |    0 |
| 2014-01-03 08:05:02 |    2 |
| 2014-01-03 08:10:02 |   11 |
| 2014-01-03 08:15:03 |   72 |
| 2014-01-03 08:20:04 |  162 |
| 2014-01-03 08:25:04 |  307 |
| 2014-01-03 08:30:04 |  464 |
| 2014-01-03 08:40:01 |  665 |
| 2014-01-03 08:45:01 |  764 |
| 2014-01-03 08:50:02 |  844 |
| 2014-01-03 08:55:01 |  931 |
| 2014-01-03 09:00:02 | 1017 |
| 2014-01-03 09:05:02 | 1098 |
| 2014-01-03 09:10:01 | 1176 |
| 2014-01-03 09:15:02 | 1272 |
| 2014-01-03 09:20:01 | 1351 |
| 2014-01-03 09:25:02 | 1432 |
| 2014-01-03 09:30:01 | 1528 |
| 2014-01-03 09:35:02 | 1653 |
| 2014-01-03 09:40:02 | 1738 |
| 2014-01-03 09:45:01 | 1821 |
| 2014-01-03 09:50:02 | 1895 |
| 2014-01-03 09:55:01 | 2015 |
| 2014-01-03 10:00:02 | 2153 |
| 2014-01-03 10:05:01 | 2300 |
| 2014-01-03 10:10:02 | 2482 |
| 2014-01-03 10:15:02 | 2635 |
| 2014-01-03 10:20:01 | 2750 |
| 2014-01-03 10:25:02 | 2923 |
| 2014-01-03 10:30:01 | 3059 |
| 2014-01-03 10:35:02 | 3201 |
| 2014-01-03 10:40:01 | 3275 |
| 2014-01-03 10:45:02 | 3360 |

What I need is a bar chart. Right now I am able to get an image but it's just a wrong image.

Answer 1

If you're on a recent version of pandas, you should use the read_sql function...

import matplotlib.pyplot as plt
import pandas
import MySQLdb

# connect to MySQL database 
conn = MySQLdb.connect(host="localhost", user="xx", passwd="xxxx", db="r1array")


# this is the query we will be making 
query = """ 
SELECT TimeStamp, Pac  
FROM SolarData  
WHERE TimeStamp >= "2014-01-03"
  AND TimeStamp < "2014-01-04"; 
"""

df = pandas.read_sql(query, conn, index_col=['TimeStamp'])
fig, ax = plt.subplots()
df.plot(ax=ax)
conn.close()

Setting your TimeStamp column as the index of the dataframe will let pandas know that it goes on the x-axis.

I converted a small subset of data you posted to a CSV, and all other things equal, I get this:

response to comments:

Since you really want a bar plot, here's a solution to the problem you're having. The default width for bar plots in matplotlib is 0.8. You can find this by looking at the docstring for pyplot.bar . Point is, 0.8 of what? Well when you have a time axis, that's 0.8 days, or a little over 19 hours. So just set that width kwarg to something more sensible.

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.dates as mdates


Pac = [
    0,0,2,11,72,162,307,464,665,764,844,931,1017,1098,1176,1272,1351,1432,
    1528,1653,1738,1821,1895,2015,2153,2300,2482,2635,2750,2923,3059,3201,
    3275,3360
]
TimeStamp = mdates.num2date(mdates.datestr2num([
    '2014-01-03 07:55:02', '2014-01-03 08:00:02', '2014-01-03 08:05:02', '2014-01-03 08:10:02', 
    '2014-01-03 08:15:03', '2014-01-03 08:20:04', '2014-01-03 08:25:04', '2014-01-03 08:30:04', 
    '2014-01-03 08:40:01', '2014-01-03 08:45:01', '2014-01-03 08:50:02', '2014-01-03 08:55:01', 
    '2014-01-03 09:00:02', '2014-01-03 09:05:02', '2014-01-03 09:10:01', '2014-01-03 09:15:02', 
    '2014-01-03 09:20:01', '2014-01-03 09:25:02', '2014-01-03 09:30:01', '2014-01-03 09:35:02', 
    '2014-01-03 09:40:02', '2014-01-03 09:45:01', '2014-01-03 09:50:02', '2014-01-03 09:55:01', 
    '2014-01-03 10:00:02', '2014-01-03 10:05:01', '2014-01-03 10:10:02', '2014-01-03 10:15:02', 
    '2014-01-03 10:20:01', '2014-01-03 10:25:02', '2014-01-03 10:30:01', '2014-01-03 10:35:02', 
    '2014-01-03 10:40:01', '2014-01-03 10:45:02'
]))

fig = plt.figure()
ax = fig.add_subplot()
ax.bar(TimeStamp, Pac, align='center', width=0.002)

And that gives me: