I am trying to generate a grid of subplots based off of a Pandas groupby object. I would like each plot to be based off of two columns of data for one group of the groupby object. Fake data set:
C1,C2,C3,C4
1,12,125,25
2,13,25,25
3,15,98,25
4,12,77,25
5,15,889,25
6,13,56,25
7,12,256,25
8,12,158,25
9,13,158,25
10,15,1366,25
I have tried the following code:
import pandas as pd
import csv
import matplotlib as mpl
import matplotlib.pyplot as plt
import math
#Path to CSV File
path = "..\\fake_data.csv"
#Read CSV into pandas DataFrame
df = pd.read_csv(path)
#GroupBy C2
grouped = df.groupby('C2')
#Figure out number of rows needed for 2 column grid plot
#Also accounts for odd number of plots
nrows = int(math.ceil(len(grouped)/2.))
#Setup Subplots
fig, axs = plt.subplots(nrows,2)
for ax in axs.flatten():
for i,j in grouped:
j.plot(x='C1',y='C3', ax=ax)
plt.savefig("plot.png")
But it generates 4 identical subplots with all of the data plotted on each (see example output below):
I would like to do something like the following to fix this:
for i,j in grouped:
j.plot(x='C1',y='C3',ax=axs)
next(axs)
but I get this error
AttributeError: 'numpy.ndarray' object has no attribute 'get_figure'
I will have a dynamic number of groups in the groupby object I want to plot, and many more elements than the fake data I have provided. This is why I need an elegant, dynamic solution and each group data set plotted on a separate subplot.
Sounds like you want to iterate over the groups and the axes in parallel, so rather than having nested for loops (which iterates over all groups for each axis), you want something like this:
for (name, df), ax in zip(grouped, axs.flat):
df.plot(x='C1',y='C3', ax=ax)
You have the right idea in your second code snippet, but you're getting an error because axs is an array of axes, but plot expects just a single axis. So it should also work to replace next(axs) in your example with ax = axs.next() and change the argument of plot to ax=ax.
Related
I have two data frames (df1 and df2). Each have the same 10 variables with different values.
I created box plots of the variables in the data frames like so:
df1.boxplot()
df2.boxplot()
I get two graphs of 10 box plots next to each other for each variable. The actual output is the second graph, however, as obviously Python just runs the code in order.
Instead, I would either like these box plots to appear side by side OR ideally, I would like 10 graphs (one for each variable) comparing each variable by data frame (e.g. one graph for the first variable with two box plots in it, one for each data frame). Is that possible just using python library or do I have to use Matplotlib?
Thanks!
To get graphs, standard Python isn't enough. You'd need a graphical library such as matplotlib. Seaborn extends matplotlib to ease the creation of complex statistical plots. To work with Seaborn, the dataframes should be converted to long form (e.g. via pandas' melt) and then combined into one large dataframe.
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
import numpy as np
# suppose df1 and df2 are dataframes, each with the same 10 columns
df1 = pd.DataFrame({i: np.random.randn(100).cumsum() for i in 'abcdefghij'})
df2 = pd.DataFrame({i: np.random.randn(150).cumsum() for i in 'abcdefghij'})
# pd.melt converts the dataframe to long form, pd.concat combines them
df = pd.concat({'df1': df1.melt(), 'df2': df2.melt()}, names=['source', 'old_index'])
# convert the source index to a column, and reset the old index
df = df.reset_index(level=0).reset_index(drop=True)
sns.boxplot(data=df, x='variable', y='value', hue='source', palette='turbo')
This creates boxes for each of the original columns, comparing the two dataframes:
Optionally, you could create multiple subplots with that same information:
sns.catplot(data=df, kind='box', col='variable', y='value', x='source',
palette='turbo', height=3, aspect=0.5, col_wrap=5)
By default, the y-axes are shared. You can disable the sharing via sharey=False. Here is an example, which also removes the repeated x axes and creates a common legend:
g = sns.catplot(data=df, kind='box', col='variable', y='value', x='source', hue='source', dodge=False,
palette='Reds', height=3, aspect=0.5, col_wrap=5, sharey=False)
g.set(xlabel='', xticks=[]) # remove x labels and ticks
g.add_legend()
PS: If you simply want to put two pandas boxplots next to each other, you can create a figure with two subplots, and pass the axes to pandas. (Note that pandas plotting is just an interface towards matplotlib.)
fig, (ax1, ax2) = plt.subplots(ncols=2, figsize=(15, 5))
df1.boxplot(ax=ax1)
ax1.set_title('df1')
df2.boxplot(ax=ax2)
ax2.set_title('df2')
plt.tight_layout()
plt.show()
I have a dictionary of dataframes where the key is the name of each dataframe and the value is the dataframe itself.
I am looking to iterate through the dictionary and quickly plot the top 10 rows in each dataframe. Each dataframe would have its own plot. I've attempted this with the following:
for df in dfs:
data = dfs[df].head(n=10)
sns.barplot(data=data, x='x_col', y='y_col', color='indigo').set_title(df)
This works, but only returns a plot for the last dataframe in the iteration. Is there a way I can modify this so that I am also able to return the subsequent plots?
By default, seaborn.barplot() plots data on the current Axes. If you didn't specify the Axes to plot on, the latter will override the previous one. To overcome this, you can either create a new figure in each loop or plot on a different axis by specifying the ax argument.
import matplotlib.pyplot as plt
for df in dfs:
data = dfs[df].head(n=10)
plt.figure() # Create a new figure, current axes also changes.
sns.barplot(data=data, x='x_col', y='y_col', color='indigo').set_title(df)
I have an excel file with the following data:
My code so far:
import pandas as pd
import matplotlib as plt
df=pd.read_excel('file.xlsx', header=[0,1], index_col=[0])
Firstly, am I reading in my file correctly to have a multi index using Main (A,B,C) as the first level and Value (X,Y) as the second level.
Using Pandas and Matplotlib - how do I plot individual scatter plot for Main (A,B,C) with each x,y as the scatter values (imaged below) . I can do it messily calling each column in an individual plot function.
Is there a nicer way to do it with multi-indexing or group by?
This should help:
df = df.set_index(['Main1', 'Main2']).value
df.unstack.plot(kind='line', stacked=True)
I have a dataframe that consists of a bunch of x,y data that I'd like to see in scatter form along with a line. The dataframe consists of data with its form repeated over multiple categories. The end result I'd like to see is some kind of grid of the plots, but I'm not totally sure how matplotlib handles multiple subplots of overplotted data.
Here's an example of the kind of data I'm working with:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
category = np.arange(1,10)
total_data = pd.DataFrame()
for i in category:
x = np.arange(0,100)
y = 2*x + 10
data = np.random.normal(0,1,100) * y
dataframe = pd.DataFrame({'x':x, 'y':y, 'data':data, 'category':i})
total_data = total_data.append(dataframe)
We have x data, we have y data which is a linear model of some kind of generated dataset (the data variable).
I had been able to generate individual plots based on subsetting the master dataset, but I'd like to see them all side-by-side in a 3x3 grid in this case. However, calling the plots within the loop just overplots them all onto one single image.
Is there a good way to take the following code block and make a grid out of the category subsets? Am I overcomplicating it by doing the subset within the plot call?
plt.scatter(total_data['x'][total_data['category']==1], total_data['data'][total_data['category']==1])
plt.plot(total_data['x'][total_data['category']==1], total_data['y'][total_data['category']==1], linewidth=4, color='black')
If there's a simpler way to generate the by-category scatter plus line, I'm all for it. I don't know if seaborn has a similar or more intuitive method to use than pyplot.
You can use either sns.FacetGrid or manual plt.plot. For example:
g = sns.FacetGrid(data=total_data, col='category', col_wrap=3)
g = g.map(plt.scatter, 'x','data')
g = g.map(plt.plot,'x','y', color='k');
Gives:
Or manual plt with groupby:
fig, axes = plt.subplots(3,3)
for (cat, data), ax in zip(total_data.groupby('category'), axes.ravel()):
ax.scatter(data['x'], data['data'])
ax.plot(data['x'], data['y'], color='k')
gives:
When exploring a I often use Pandas' DataFrame.hist() method to quickly display a grid of histograms for every numeric column in the dataframe, for example:
import matplotlib.pyplot as plt
import pandas as pd
from sklearn import datasets
data = datasets.load_iris()
df = pd.DataFrame(data.data, columns=data.feature_names)
df.hist(bins=50, figsize=(10,7))
plt.show()
Which produces a figure with separate plots for each column:
I've tried the following:
import pandas as pd
import seaborn as sns
from sklearn import datasets
data = datasets.load_iris()
df = pd.DataFrame(data.data, columns=data.feature_names)
for col_id in df.columns:
sns.distplot(df[col_id])
But this produces a figure with a single plot and all columns overlayed:
Is there a way to produce a grid of histograms showing the data from a DataFrame's columns with Seaborn?
You can take advantage of seaborn's FacetGrid if you reorganize your dataframe using melt. Seaborn typically expects data organized this way (long format).
g = sns.FacetGrid(df.melt(), col='variable', col_wrap=2)
g.map(plt.hist, 'value')
There is no equivalent as seaborn displot itself will only pick 1-D array, or list, maybe you can try generating the subplots.
fig, ax = plt.subplots(2, 2, figsize=(10, 10))
for i in range(ax.shape[0]):
for j in range(ax.shape[1]):
sns.distplot(df[df.columns[i*2+j]], ax=ax[i][j])
https://seaborn.pydata.org/examples/distplot_options.html
Here is an example how you can show 4 graphs using subplot, with seaborn.
Anothert useful SEABORN method to quickly display a grid of histograms for every numeric column in the dataframe for you could be the quick,clean and handy sns.pairplot()
try:
sns.pairplot(df)
this has a lot of cool parameters you can explor like Hue etc
pairplot example for iris dataset
if you DON'T want the scatters you can actually create a customised grid really really quickly using sns.PairGrid(df)
this creates an empty grid with all the spaces and you can map whatever you want on them :g = sns.pairgrid(df)
`g.map(sns.distplot)` or `g.map_diag(plt.scatter)`
etc
I ended up adapting jcaliz's to make it work more generally, i.e. not just when the DataFrame has four columns, I also added code to remove any unused axes and ensure axes appear in alphabetical order (as with df.hist()).
size = int(math.ceil(len(df.columns)**0.5))
fig, ax = plt.subplots(size, size, figsize=(10, 10))
for i in range(ax.shape[0]):
for j in range(ax.shape[1]):
data_index = i*ax.shape[1]+j
if data_index < len(df.columns):
sns.distplot(df[df.columns.sort_values()[data_index]], ax=ax[i][j])
for i in range(len(df.columns), size ** 2):
fig.delaxes(ax[i // size][i % size])