Colours and symbolics

# Use plotly express to plot petal widths and lengths of the iris data set (imported from plotly). 
# Colour and symbol by species.
import plotly.express as px
df = px.data.iris()
fig = px.scatter(df, x="petal_width", y="petal_length", color="species", symbol="species")
fig.show()

import plotly.graph_objects as go

df = px.data.iris()
x = df["sepal_width"]
y = df["sepal_length"]

# Main plot
f = go.FigureWidget([go.Scatter(x=x, y=y, mode='markers')])
f.update_layout(xaxis_title="Sepal width", yaxis_title="Sepal length")
# Set xlimits and ylimits
f.update_yaxes(range=[4, 8.1])
f.update_xaxes(range=[1.8, 4.5])

# Create a color list based on px.colors.qualitative.Vivid, which has three colors.
colors = [px.colors.qualitative.Vivid[i] for i in df["species_id"]]
sizes = [10] * len(x)

# Attributes of the scatter object
scatter = f.data[0]
scatter.marker.color = colors
scatter.marker.size = sizes
f.layout.hovermode = 'closest'

# Create our callback function
def update_point(trace, points, selector):
    species = df["species_id"][points.point_inds[0]]
    # Change all ellements in cols that have df.species != species gray
    # to '#bae2be'. Keep in mind that df.species is a series and that cols is a list.
    cols = ['#BBBBBB' if s != species else px.colors.qualitative.Vivid[s] for s in df.species_id]
    size = [14 if s == species else 10 for s in df.species_id]
    
    with f.batch_update():
        scatter.marker.color = cols
        scatter.marker.size = size

def reset_point(trace, points, selector):
    with f.batch_update():
        scatter.marker.color = colors
        scatter.marker.size = sizes

# Assign the callback function to the scatter object
scatter.on_hover(update_point)
scatter.on_unhover(reset_point)

f # Do not use .show() or the figure will not be interactive

df = px.data.iris()
fig = px.scatter(df, x="sepal_width", y="sepal_length", color="species")

# Add a red cross behind the samples with lowest and highets sepal width
df_extreme = df[(df.sepal_width == df.sepal_width.min()) | (df.sepal_width == df.sepal_width.max())]
fig.add_trace(go.Scatter(x=df_extreme.sepal_width, y=df_extreme.sepal_length, mode="markers",
                         marker=dict(color="black", size=12, symbol="cross"), showlegend=False))
fig.add_trace(go.Scatter(x=df_extreme.sepal_width, y=df_extreme.sepal_length, 
                         mode="markers", marker=dict(color="red", size=12, symbol="cross-open"), 
                         showlegend=False))

# Traces are plotted in the order they are added. To plot the crosses behind the original data, 
# we need to change the order of the traces.
fig.data = (fig.data[3], fig.data[4], fig.data[0], fig.data[1], fig.data[2])

# Add title and axis labels
fig.update_layout(title="Can you spot the outliers?", 
                  xaxis_title="Sepal width", yaxis_title="Sepal length")
fig.show()

# Plot a random sample of 1000 normal distributed points as a 2D scatter plot.
import numpy as np
np.random.seed(1)
N = 1000
random_x = np.random.randn(N)
random_y = np.random.randn(N)
dist_origin = np.sqrt(random_x ** 2 + random_y ** 2)

# Let the colour be blue and the opacity 0.7 for points closer than 2 from the origin. 
# Use opacity = 0.2 for all other points.
fig = px.scatter(x=random_x[dist_origin > 2], y=random_y[dist_origin > 2], opacity=0.2, color_discrete_sequence=['blue'])
fig.add_trace(go.Scatter(x=random_x[dist_origin <= 2], y=random_y[dist_origin <= 2], 
                         mode="markers", marker=dict(color="blue", opacity=0.7), showlegend=False))

# Add a circle with radius 2 around the origin
fig.add_shape(type="circle", xref="x", yref="y", x0=-2, y0=-2, x1=2, y1=2, line_color="LightSeaGreen")
fig.update_layout(title="Softening outliers using opacity")
fig.show()

# Use plotly express to plot petal widths and lengths of the iris data set (imported from plotly).
# Colour by species. 
# Use a colour scale friendly to colour blind people.
import plotly.express as px
df = px.data.iris()
fig = px.scatter(df, x="petal_width", y="petal_length", color="species", 
                 color_discrete_sequence=px.colors.qualitative.Vivid)
fig.show()

# Add ipywidgets slider to control the number of rectangles.
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import ipywidgets as widgets
def plot_rectangles(n):
    # Set figure size to 10 x 4
    _, ax = plt.subplots(figsize=(10, 4))
    for i in range(n):
        rectR = patches.Rectangle((i / n, 5), 1 / n, 1, facecolor=(i / (n-1), 0, 0))
        ax.add_patch(rectR)
        rectG = patches.Rectangle((i / n, 4), 1 / n, 1, facecolor=(0, i / (n-1), 0))
        ax.add_patch(rectG)
        rectB = patches.Rectangle((i / n, 3), 1 / n, 1, facecolor=(0, 0, i / (n-1)))
        ax.add_patch(rectB)
        rectY = patches.Rectangle((i / n, 2), 1 / n, 1, facecolor=(i / (n-1), i / (n-1), 0))
        ax.add_patch(rectY)
        rectC = patches.Rectangle((i / n, 1), 1 / n, 1, facecolor=(0, i / (n-1), i / (n-1)))
        ax.add_patch(rectC)
        rectM = patches.Rectangle((i / n, 0), 1 / n, 1, facecolor=(i / (n-1), 0, i / (n-1)))
        ax.add_patch(rectM)
    # Set y limits
    ax.set_ylim(0, 6)
    plt.show()
widgets.interact(plot_rectangles, n=widgets.IntSlider(min=20, max=100, step=1, value=20))

# There are many colourmaps in matplotlib.
from matplotlib import colormaps
list(colormaps)