Line Plot¶

line styles¶

import numpy as np
import plotly.graph_objects as go

fig = go.Figure()

x = np.array([1, 2, 3, 4, 5])
y = np.array([2, 10, 4, 15, 9])
line_styles = ['0', '8,3', '8,3,2,3', '3,2', '14,2', '8,3,2,3,2,3', '4,3', '6,3,2,3,2,3,2,3', '2,5']
cnt = len(line_styles)

for i, line_stsyle in enumerate(line_styles):
    fig.add_trace(go.Scatter(
        x=x,
        y=y + 5*i,
        mode='lines',
        line=dict(color='blue', dash=line_styles[i], width=2 * (1 - i * 0.4 / cnt)),
    ))

fig.show()

custom line styles¶

def get_line_widths(cnt, width=2):
    line_widths = [width * (1 - i * 0.4 / cnt) for i in range(cnt)]
    return line_widths

def get_line_styles(cnt):
    linestyles = [
        '0', '8,3', '8,3,2,3', '3,2', '14,2',
        '8,3,2,3,2,3', '4,3', '6,3,2,3,2,3,2,3', '2,5'
    ]
    line_styles = (
        linestyles * (cnt // len(linestyles)) + linestyles[:cnt % len(linestyles)]
    )
    return line_styles

example¶

Plot two columns from two dfs as lines and set the line color, legend and figure size.

if name is not set will be default to the df val col name and line will be using default color rotation

fig = go.Figure()
fig.add_trace(go.Scatter(x=df1['ts'], y=df1['val'], mode='lines', name='line-1', line=dict(color='blue')))
fig.add_trace(go.Scatter(x=df2['ts'], y=df2['val'], mode='lines', name='line-2', line=dict(color='red')))
fig.update_layout(
    xaxis_title='ts',
    yaxis_title='val',
    width=28 * 37.795,  # Convert cm to pixels (1 cm = 37.795 pixels)
    height=14 * 37.795,  # Convert cm to pixels (1 cm = 37.795 pixels)
)  
fig.show()