In the timeline plot I’m making, I want date tickers to show only specified dates. (In my example I show tickers for events ‘A’, but it can be any list on tickers). I found how to do it when x-axis data is numeric (upper subplot in my example), but this won’t work with timestamp date type (bottom plot).
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import matplotlib.ticker as ticker
myData = pd.DataFrame({'date':['2019-01-15','2019-02-10','2019-03-20','2019-04-17','2019-05-23','2019-06-11'],'cnt':range(6),'event':['a','b','a','b','a','b']})
myData['date'] = [pd.Timestamp(j) for j in myData['date']]
start = pd.Timestamp('2019-01-01')
stop = pd.Timestamp('2019-07-01')
inxa = myData.loc[myData['event'] == 'a'].index
inxb = myData.loc[myData['event'] == 'b'].index
# create two plots, one with 'cnt' as x-axis, the other 'dates' on x-axis.
fig, ax = plt.subplots(2,1,figsize=(16,9))
ax[0].plot((0,6),(0,0), 'k')
ax[1].plot((start, stop),(0,0))
for g in inxa:
ax[0].plot((myData.loc[g,'cnt'],myData.loc[g,'cnt']),(0,1),c='r')
ax[1].plot((myData.loc[g,'date'],myData.loc[g,'date']),(0,1),c='r')
for g in inxb:
ax[0].plot((myData.loc[g,'cnt'],myData.loc[g,'cnt']),(0,2),c='b')
ax[1].plot((myData.loc[g,'date'],myData.loc[g,'date']),(0,2),c='b')
xlist0 = myData.loc[myData['event']=='a','cnt']
xlist1 = myData.loc[myData['event']=='a','date']
ax[0].xaxis.set_major_locator(ticker.FixedLocator(xlist0))
# ax[1].xaxis.set_major_locator(**???**)
Couldn't find a sufficient duplicate, maybe I didn't look hard enough. There are a number of ways to do this:
Converting to numbers first or using the underlying values of a Pandas DateTime Series
xticks = [mdates.date2num(z) for z in xlist1]
# or
xticks = xlist1.values
and at least a couple ways to use it/them
ax[1].xaxis.set_major_locator(ticker.FixedLocator(xticks))
ax[1].xaxis.set_ticks(xticks)
Date tick labels
How to set the xticklabels for date in matplotlib
how to get ticks every hour?
...
Related
I have two lists representing dates and values respectively:
dates = ['10/6/2020',
'10/7/2020',
'10/8/2020',
'10/9/2020',
'10/12/2020',
'10/13/2020',
'10/14/2020',
'10/15/2020',
'10/16/2020',
'10/19/2020']
and
values = ['40.660',
'39.650',
'41.010',
'41.380',
'39.950',
'40.790',
'41.050',
'40.370',
'40.880',
'40.860']
I want to use seaborn/matplotlib to plot them without using pandas. Is that possible? I've made a few attempts but it doesn't seem to be going too well.
Here's what I've got so far:
def plots(values=values,dates=dates):
sns.lineplot(x=dates,y=sorted(values)[::-1])
sns.scatterplot(x=dates,y=sorted(values)[::-1])
plt.show()
return
crude_data = plots()
But it gives me this:
This is obviously wrong but I don't know how to fix it. The x-axis is also messy, and I'd like to fix that as well and make it more legible without expanding the width of the graph if possible. If that's impossible, then a way to do so while expanding the graph would be happily accepted as well.
Cheers!
Just parse dates from string to datetime.
Otherwise they are treated as strings and lose all their properties besides sortability (which can be wrong, depending on date format).
Also add xticks rotation for better x axis labels.
edit:
Also as others noticed, your numeric data is in string type aswell.
import seaborn as sns
import matplotlib.pyplot as plt
import datetime
dates = ['10/6/2020',
'10/7/2020',
'10/8/2020',
'10/9/2020',
'10/12/2020',
'10/13/2020',
'10/14/2020',
'10/15/2020',
'10/16/2020',
'10/19/2020']
values = ['40.660',
'39.650',
'41.010',
'41.380',
'39.950',
'40.790',
'41.050',
'40.370',
'40.880',
'40.860']
dates = [datetime.datetime.strptime(date, '%m/%d/%Y') for date in dates]
values = [float(val) for val in values]
def plots(values=values,dates=dates):
sns.lineplot(x=dates,y=sorted(values)[::-1])
sns.scatterplot(x=dates,y=sorted(values)[::-1])
plt.xticks(rotation=45)
plt.show()
return crude_data = plots()
You can use datetime library.
import matplotlib.pyplot as plt
from datetime import datetime
import seaborn as sns
dates = ['10/6/2020',
'10/7/2020',
'10/8/2020',
'10/9/2020',
'10/12/2020',
'10/13/2020',
'10/14/2020',
'10/15/2020',
'10/16/2020',
'10/19/2020']
x = [datetime.strptime(i, '%m/%d/%Y') for i in dates]
values = ['40.660',
'39.650',
'41.010',
'41.380',
'39.950',
'40.790',
'41.050',
'40.370',
'40.880',
'40.860']
nvalues = [float(i) for i in values]
def plots(values=nvalues,dates=x):
sns.lineplot(x=dates,y=sorted(values)[::-1])
sns.scatterplot(x=dates,y=sorted(values)[::-1])
plt.xticks(rotation=45, ha='right')
plt.show()
return
import matplotlib.pyplot as plt
from datetime import datetime
crude_data = plots()
Output:
I made 3 changes to your code:
I set the y axis data to be float instead of string by removing the '',
I removed sorted from the two plot lines, and
I added sort=False to the lineplot call
import matplotlib.pyplot as plt
import seaborn as sns
dates = ['10/6/2020',
'10/7/2020',
'10/8/2020',
'10/9/2020',
'10/12/2020',
'10/13/2020',
'10/14/2020',
'10/15/2020',
'10/16/2020',
'10/19/2020']
values = [40.660,
39.650,
41.010,
41.380,
39.950,
40.790,
41.050,
40.370,
40.880,
40.860]
def plots(values=values,dates=dates):
sns.lineplot(x=dates,y=values[::-1], sort=False)
s = sns.scatterplot(x=dates,y=values[::-1])
plt.xticks(rotation=45, ha='right')
plt.show()
return s
crude_data = plots()
This gives:
I am trying to create a heat map from pandas dataframe using seaborn library. Here, is the code:
test_df = pd.DataFrame(np.random.randn(367, 5),
index = pd.DatetimeIndex(start='01-01-2000', end='01-01-2001', freq='1D'))
ax = sns.heatmap(test_df.T)
ax.xaxis.set_major_locator(mdates.MonthLocator())
ax.xaxis.set_minor_locator(mdates.DayLocator())
ax.xaxis.set_major_formatter(mdates.DateFormatter('%b'))
ax.xaxis.set_minor_formatter(mdates.DateFormatter('%d'))
However, I am getting a figure with nothing printed on the x-axis.
Seaborn heatmap is a categorical plot. It scales from 0 to number of columns - 1, in this case from 0 to 366. The datetime locators and formatters expect values as dates (or more precisely, numbers that correspond to dates). For the year in question that would be numbers between 730120 (= 01-01-2000) and 730486 (= 01-01-2001).
So in order to be able to use matplotlib.dates formatters and locators, you would need to convert your dataframe index to datetime objects first. You can then not use a heatmap, but a plot that allows for numerical axes, e.g. an imshow plot. You may then set the extent of that imshow plot to correspond to the date range you want to show.
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
df = pd.DataFrame(np.random.randn(367, 5),
index = pd.DatetimeIndex(start='01-01-2000', end='01-01-2001', freq='1D'))
dates = df.index.to_pydatetime()
dnum = mdates.date2num(dates)
start = dnum[0] - (dnum[1]-dnum[0])/2.
stop = dnum[-1] + (dnum[1]-dnum[0])/2.
extent = [start, stop, -0.5, len(df.columns)-0.5]
fig, ax = plt.subplots()
im = ax.imshow(df.T.values, extent=extent, aspect="auto")
ax.xaxis.set_major_locator(mdates.MonthLocator())
ax.xaxis.set_minor_locator(mdates.DayLocator())
ax.xaxis.set_major_formatter(mdates.DateFormatter('%b'))
fig.colorbar(im)
plt.show()
I found this question when trying to do a similar thing and you can hack together a solution but it's not very pretty.
For example I get the current labels, loop over them to find the ones for January and set those to just the year, setting the rest to be blank.
This gives me year labels in the correct position.
xticklabels = ax.get_xticklabels()
for label in xticklabels:
text = label.get_text()
if text[5:7] == '01':
label.set_text(text[0:4])
else:
label.set_text('')
ax.set_xticklabels(xticklabels)
Hopefully from that you can figure out what you want to do.
I would like to consolidate tick data stored in a pandas dataframe to the open high low close format but not time related, but aggregated for every 100 ticks. After that I would like to display them in a candlestick chart using matlibplot.
I solved this already for a time related aggregation using a pandas dataset with two values: TIMESTAMP and PRICE. The TIMESTAMP already has the pandas date format so I work with that:
df["TIMESTAMP"]= pd.to_datetime(df["TIMESTAMP"])
df = df.set_index(['TIMESTAMP'])
data_ohlc = df['PRICE'].resample('15Min').ohlc()
Is there any function, that resamples datasets in the ohlc format not using a time frame, but a count of ticks?
After that it comes to visualization, so for plotting I have to change date format to mdates. The candlestick_ohlc function requires a mdate format:
data_ohlc["TIMESTAMP"] = data_ohlc["TIMESTAMP"].apply(mdates.date2num)
from mpl_finance import candlestick_ohlc
candlestick_ohlc(ax1,data_ohlc.values,width=0.005, colorup='g', colordown='r',alpha=0.75)
So is there any function to display a candle stick chart without mdates because by aggregating tick data there would be no time relation?
As there seems to be no build in function for this problem I wrote one myself. The given dataframe needs to have the actual values in the column "PRICE":
def get_pd_ohlc(mydf, interval):
## use a copy, so that the new column doesn't effect the original dataset
mydf = mydf.copy()
## Add a new column to name tick interval
interval = [(1+int(x/interval)) for x in range(mydf["PRICE"].count())]
mydf["interval"] = interval
##Step 1: Group
grouped = mydf.groupby('interval')
##Step 2: Calculate different aggregations
myopen = grouped['PRICE'].first()
myhigh = grouped['PRICE'].max()
mylow = grouped['PRICE'].min()
myclose = grouped['PRICE'].last()
##Step 3: Generate Dataframe:
pd_ohlc = pd.DataFrame({'OPEN':myopen,'HIGH':myhigh,'LOW':mylow,'CLOSE':myclose})
return(pd_ohlc)
pd_100 = get_pd_ohlc(df,100)
print (pd_100.head())
I also found a solution to display ist. Module mpl_finance has a function candlestick2_ohlc, that does not need any datetime information. Here is the code:
#Making plot
import matplotlib.pyplot as plt
from mpl_finance import candlestick2_ohlc
fig = plt.figure()
plt.rcParams['figure.figsize'] = (16,8)
ax1 = plt.subplot2grid((6,1), (0,0), rowspan=12, colspan=1)
#Making candlestick plot
candlestick2_ohlc(ax1, pd_ohlc['OPEN'], pd_ohlc['HIGH'],
pd_ohlc['LOW'], pd_ohlc['CLOSE'], width=0.5,
colorup='#008000', colordown='#FF0000', alpha=1)
I have data that shows some values collected on three different dates: 2015-01-08, 2015-01-09 and 2015-01-12. For each date there are several data points that have timestamps.
Date/times are in a list and it looks as follows:
['2015-01-08-09:00:00', '2015-01-08-10:00:00', '2015-01-08-11:00:00', '2015-01-08-12:00:00', '2015-01-08-13:00:00', '2015-01-09-14:00:00', '2015-01-09-15:00:00', '2015-01-09-16:00:00', '2015-01-12-09:00:00', '2015-01-12-10:00:00', '2015-01-12-11:00:00']
On the other hand I have corresponding values (floats) in another list:
[12210.0, 12210.0, 12180.0, 12240.0, 12250.0, 12420.0, 12390.0, 12400.0, 12380.0, 12450.0, 12460.0]
To put all this together and plot a graph I use following code:
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import matplotlib.dates as md
import dateutil
from matplotlib.font_manager import FontProperties
timestamps = ['2015-01-08-09:00:00', '2015-01-08-10:00:00', '2015-01-08-11:00:00', '2015-01-08-12:00:00', '2015-01-08-13:00:00', '2015-01-09-14:00:00', '2015-01-09-15:00:00', '2015-01-09-16:00:00', '2015-01-12-09:00:00', '2015-01-12-10:00:00', '2015-01-12-11:00:00']
ticks = [12210.0, 12210.0, 12180.0, 12240.0, 12250.0, 12420.0, 12390.0, 12400.0, 12380.0, 12450.0, 12460.0]
plt.subplots_adjust(bottom=0.2)
plt.xticks( rotation=90 )
dates = [dateutil.parser.parse(s) for s in timestamps]
ax=plt.gca()
ax.set_xticks(dates)
ax.tick_params(axis='x', labelsize=8)
xfmt = md.DateFormatter('%H:%M:%S')
ax.xaxis.set_major_formatter(xfmt)
plt.plot(dates, ticks, label="Price")
plt.xlabel("Date and time", fontsize=12)
plt.ylabel("Price", fontsize=12)
plt.suptitle("Price during last three days", fontsize=12)
plt.legend(loc=0,prop={'size':8})
plt.savefig("figure.pdf")
When I try to plot these datetimes and values I get a messy graph with the line going back and forth.
It looks like the dates are being ignored and only timestamps are taken in account which is the reason for the messy chart. I tried to edit the datetimes to have the same date and consecutive timestamps and it fixed the chart. However, I must have dates as well..
What am I doing wrong?
When I try to plot these datetimes and values I get a messy graph with the line going back and forth.
Your plots are going all over the place because plt.plot connects the dots in the order you give it. If this order is not monotonically increasing in x, then it looks "messy". You can sort the points by x first to fix this. Here is a minimal example:
import numpy as np
import pylab as plt
X = np.random.random(20)
Y = 2*X+np.random.random(20)
idx = np.argsort(X)
X2 = X[idx]
Y2 = Y[idx]
fig,ax = plt.subplots(2,1)
ax[0].plot(X,Y)
ax[1].plot(X2,Y2)
plt.show()
I have the following code to plot a chart with matplotlib
#!/usr/bin/env python
import matplotlib.pyplot as plt
import urllib2
import json
req = urllib2.urlopen("http://localhost:17668/retrieval/data/getData.json? pv=LNLS:ANEL:corrente&donotchunk")
data = json.load(req)
secs = [x['secs'] for x in data[0]['data']]
vals = [x['val'] for x in data[0]['data']]
plt.plot(secs, vals)
plt.show()
The secs arrays is epoch time.
What I want is to plot the data in the x axis (secs) as a date (DD-MM-YYYY HH:MM:SS).
How can I do that?
To plot date-based data in matplotlib you must convert the data to the correct format.
One way is to first convert your data to datetime objects, for an epoch timestamp you should use datetime.datetime.fromtimestamp().
You must then convert the datetime objects to the right format for matplotlib, this can be handled using matplotlib.date.date2num.
Alternatively you can use matplotlib.dates.epoch2num and skip converting your date to datetime objects in the first place (while this will suit your use-case better initially, I would recommend trying to keep date based date in datetime objects as much as you can when working, it will save you a headache in the long run).
Once you have your data in the correct format you can plot it using plot_date.
Finally to format your x-axis as you wish you can use a matplotlib.dates.DateFormatter object to choose how your ticks will look.
import matplotlib.pyplot as plt
import matplotlib.dates as mdate
import numpy as np
# Generate some random data.
N = 40
now = 1398432160
raw = np.array([now + i*1000 for i in range(N)])
vals = np.sin(np.linspace(0,10,N))
# Convert to the correct format for matplotlib.
# mdate.epoch2num converts epoch timestamps to the right format for matplotlib
secs = mdate.epoch2num(raw)
fig, ax = plt.subplots()
# Plot the date using plot_date rather than plot
ax.plot_date(secs, vals)
# Choose your xtick format string
date_fmt = '%d-%m-%y %H:%M:%S'
# Use a DateFormatter to set the data to the correct format.
date_formatter = mdate.DateFormatter(date_fmt)
ax.xaxis.set_major_formatter(date_formatter)
# Sets the tick labels diagonal so they fit easier.
fig.autofmt_xdate()
plt.show()
You can change the ticks locations and formats on your plot:
import matplotlib.pyplot as plt
import matplotlib.ticker as mtick
import time
secs = [10928389,102928123,383827312,1238248395]
vals = [12,8,4,12]
plt.plot(secs,vals)
plt.gcf().autofmt_xdate()
plt.gca().xaxis.set_major_locator(mtick.FixedLocator(secs))
plt.gca().xaxis.set_major_formatter(
mtick.FuncFormatter(lambda pos,_: time.strftime("%d-%m-%Y %H:%M:%S",time.localtime(pos)))
)
plt.tight_layout()
plt.show()