This question already has answers here:
How to plot multiple Seaborn Jointplot in Subplot
(4 answers)
Closed 4 years ago.
Using seaborn, I created two Facetgrid (seaborn.axisgrid.FacetGrid) plots as follows:
#load dataset
import pandas as pd
iris=pd.read_csv('https://raw.githubusercontent.com/mwaskom/seaborn-data/master/iris.csv')
#plotting
import matplotlib.pyplot as plt
import seaborn as sns
g1 = sns.FacetGrid(pd.melt(iris.loc[:,['sepal_length','sepal_width','species']], id_vars='species'), col='variable')
g1.map(sns.boxplot, 'species','value',order=['setosa','versicolor','virginia'])
g2 = sns.FacetGrid(iris, col="species")
g2.map(plt.scatter, "sepal_length", "sepal_width")
Two separate plots were generated:
and
I would like to have them in a single figure (and save as a single pdf file). However, I could not find any way to concatenate the plots (through python). I can always use Illustrator/Inkscape or bash commands to concatenate separate plots. However I want to avoid that and do all the steps in python alone.
I tried fig.axes.append (fig.axes.append(g1.axes.flatten()[0])), but the subplots ended up overlapping each other.
Question: Is it possible to concatenate two seaborn-Facegrid plots?
Personally when I need to have multiple figures displayed together I will have an html output with embedded references to each of the plots I want to display. That way you can review the analysis all in one place on a browser.
Related
I've used pandas.read_csv to generate a 1000-row dataframe with 32 columns. I'm looking to plot a histogram or bar chart (depending on data type) of each column. For columns of type 'int64', I've tried doing matplotlib.pyplot.hist(df['column']) and df.hist(column='column'), as well as calling matplotlib.pyplot.hist on df['column'].values and df['column'].to_numpy(). Weirdly, nthey all take areally long time (>30s) and when I've allowed them to complet, I get unit-height bars in multiple colors, as if there's some sort of implicit grouping and they're all being separated into different groups. Any ideas about what I can do to get a normal histogram? Unfortunately I closed the charts so I can't show you an example right now.
Edit - this seems to be a much bigger problem with Int columns, and casting them to float fixes the problem.
Follow these two steps:
import the Histogram class from the Matplotlib library
use the "plot" method, which will accept a dataframe as argument
import matplotlib.pyplot as plt
plt.hist(df['column'], color='blue', edgecolor='black', bins=int(45/1))
Here's the source.
This question already has answers here:
How to change the number or rows and columns in my catplot
(2 answers)
Seaborn multiple barplots
(2 answers)
subplotting with catplot
(1 answer)
Closed 4 months ago.
I have this dataset table
And i want to plot profit made by different sub_category in different region.
now i am using this code to make a plot using seaborn
sns.barplot(data=sub_category_profit,x="sub_category",y="profit",hue="region")
I am getting a extreamly huge plot like this output
is there is any way i can get sub-plots of this like a facet-gird. Like subplots of different sub_category. I have used the facet grid function but it is the also not working properly.
g=sns.FacetGrid(data=sub_category_profit,col="sub_category")
g.map(sns.barplot(data=sub_category_profit,x="region",y="profit"))
I am getting the following output
As you can see in the facet grid output the plots are very small and the bar graph is just present on one grid.
See docs on seaborn.FacetGrid, particularly the posted example, where you should not pass the data again in the map call but simply the plot function and x and y variables to draw plots to corresponding facets.
Also, consider the col_wrap argument since you do not specify row to avoid the very wide plot output.
g=sns.FacetGrid(data=sub_category_profit, col="sub_category", col_wrap=4)
g.map_dataframe(sns.barplot, x="region", y="profit")
I intended to plot a trend of daily BTC prices, which is recorded as a csv file, consisting of a sequence of 3892 numbers.
Being a csv file, I could use microsoft excel to plot the trend.
It looks like this;
Meanwhile, I attempted to do the same thing using pandas and matplotlib in a jupyter notebook, which doesn't look good;
What's wrong with my implementation?
One more thing is the implementing time ; it takes a quite a while, like one or two minute.
This is not the case when I put a code and implement like this ;
plt.plot(list(range(1000)),list(range(1000)))
Here is the full code for my attempt.
import pandas as pd
from matplotlib import pyplot as plt
df=pd.read_csv('BTC_inv_daily.csv')
Close=df['Close'].tolist()
N=len(Close)
plt.plot(list(range(N)),Close)
plt.show()
I have the following Dataset and I wanna create a plot, which to columns compares with each other.
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
ds=pd.read_csv('h-t-t-p-:bit.ly/uforeports') #My DataSet
ds.head(5) # Only the fist 5 rows to show
ds1= ds.head(4).drop(['Colors Reported','State'],axis=1) # Droping of unnecesssary rows
print(ds1)
Now I wanna compare "City" and "Shape Reported" with help of plotting. I found something with Pandas but this is not so elegant!
x=ds.loc[0:100,['State']]
y=ds.loc[0:100,['Shape Reported']]
x.apply(pd.value_counts).plot(kind='bar', subplots=True)
y.apply(pd.value_counts).plot(kind='bar', subplots=True)
Do you know a better solution with Matplotlib to this problem?
This is what I want
It's not exactly clear how you want to compare them.
The simplest way of drawing a bar chart is:
df['State'].value_counts().plot.bar()
df['Shape Reported'].value_counts().plot.bar()
If you just want to do it for the first 100 rows as in your example, just add head(100):
df['State'].head(100).value_counts().plot.bar()
df['Shape Reported'].head(100).value_counts().plot.bar()
EDIT:
To compare the two values you can plot a bivariate distribution plot. This is easily done with seaborn:
import seaborn
sns.displot(df,x='State', y='Shape Reported', height=6, aspect=1.33)
Result:
I'm using the "PyPlot" package in Julia, and I want to add gridlines at specified locations. I'm not familiar enough with Python/Matlab to use their documentation pages to help - the commands differ in Julia. I want a basic plot, with gridlines on both axes at intervals of 1:
using PyPlot
fig=figure("Name")
grid("on")
scatter([1,2,3,4],[4,5,6,7])
Help appreciated...
PyPlot is just an interface to Matplotlib, so the commands
to customize the grid are Matplotlib's commands.
One way to configure the gridlines on both axes at intervals of 1 (for the given data) is:
using PyPlot
fig=figure(figsize=[6,3])
ax1=subplot(1,1,1) # creates a subplot with just one graphic
ax1[:xaxis][:set_ticks](collect(1:4)) # configure x ticks from 1 to 4
ax1[:yaxis][:set_ticks](collect(4:7)) # configure y ticks from 4 to 7
grid("on")
scatter([1,2,3,4],[4,5,6,7])
This code was tested inside an IJulia's notebook, and produces the following output:
Take a look at Various Julia plotting examples using PyPlot.
tested with Julia Version 0.4.3
The values where grid lines are drawn can be controlled by passing an array to the xticks() and yticks() functions.
A simple example:
using PyPlot
fig=figure("Name")
grid("on")
xticks(0:5)
yticks(3:8)
scatter([1,2,3,4],[4,5,6,7])
If you want it to be more flexible you can figure out the limits based on your data and set the tick interval to something else.
One little more dynamic way to configure the x-axis of the grid could be:
x_data = [1,2,3,4]
x_tick_interval = 2;
x_tick_start = minimum(xdata)
x_tick_end = maximum(xdata)
xticks(x_tick_start:x_tick_interval:x_tick_end)