Data Visualization with pandas

Bar plot

Bar plots are useful to visualize relationship between a categorical and numerical variables.

# Import matplotlib.pyplot with alias plt
import matplotlib.pyplot as plt

# Look at the first few rows of data
print(avocados.head())

# Get the total number of avocados sold of each size
nb_sold_by_size = avocados.groupby('size')['nb_sold'].sum()

# Create a bar plot of the number of avocados sold by size
nb_sold_by_size.plot(kind="bar")

# Show the plot
plt.show()

Line plots

Line plots are great to visualize change in numerical variable such as sales over time

# Import matplotlib.pyplot with alias plt
import matplotlib.pyplot as plt

# Get the total number of avocados sold on each date
nb_sold_by_date = avocados.groupby('date')['nb_sold'].sum()

# Create a line plot of the number of avocados sold by date
nb_sold_by_date.plot(kind="line")

# Show the plot
plt.show()

Scatter plots

Scatter plots help to visualize relationship between two numerical variables.

# Scatter plot of nb_sold vs avg_price with title
avocados.plot(kind="scatter",x="nb_sold",y="avg_price",title="Number of avocados sold vs. average price")

# Show the plot
plt.show()

Layering histograms on top of each other

# Modify bins to 20
avocados[avocados["type"] == "conventional"]["avg_price"].hist(bins=20,alpha=0.5)

# Modify bins to 20
avocados[avocados["type"] == "organic"]["avg_price"].hist(bins=20,alpha=0.5)

# Add a legend
plt.legend(["conventional", "organic"])

# Show the plot
plt.show()

Heatmap with seaborn

Let's say you have 2 dimensional numpy array V_result and you want to plot it as a heat map

import seaborn as sns
import matplotlib.pyplot as plt

states_row_length = 20
states_col_length = 20
V_result = np.zeros((states_row_length, states_col_length))
for row in range(states_row_length):
    for col in range(states_col_length):
        V_result[row, col] = state_values[State(row + 1, col + 1)]
sns.heatmap(V_result, annot=True, linewidths=2, vmin=0, vmax=20, cmap=sns.color_palette("Reds", 24))
plt.title("Optimal state values")
plt.show()