I have a series of start and finish times, from which I want to plot horizontal duration lines.
( (start1, finish1), (start2, finish2) … ) where start and finish are timestamps
I cannot find any guidance on how to achieve this either in the documentation or by other searching.
One possible workaround might be to make a series of narrow rectangles, but that feels like a kludge, when others enthuse about the wonders of Plotly !!
Hi @mike0whit,
Welcome to forum! Plotly provides the Gant chart, that with right parameters draws horizontal line from start to end time of a job. There exist two versions: ff.create_gantt(), and px.timeline() to define such a chart:
import plotly.figure_factory as ff
from datetime import datetime
df = [dict(Task="Job A", Start='2019-01-01', Finish='2019-02-01', Resource='Apple'),
dict(Task="Job B", Start='2019-03-05', Finish='2019-04-15', Resource='Grape'),
dict(Task="Job C", Start='2019-04-20', Finish='2019-09-30', Resource='Banana')]
colors = ['#7a0504', (0.2, 0.7, 0.3), 'rgb(210, 60, 180)']
fig = ff.create_gantt(df, colors=colors, index_col='Resource',
showgrid_x=True, showgrid_y=False,
show_colorbar=True, bar_width=0.025)
fig.update_layout(height=400, width=800)
fig.show()
for more details print:
help(ff.create_gantt)
or
import plotly.express as px
import pandas as pd
df = pd.DataFrame([
dict(Task="Job A", Start='2009-01-01', Finish='2009-02-28', Resource="Alex"),
dict(Task="Job B", Start='2009-03-05', Finish='2009-04-15', Resource="Alex"),
dict(Task="Job C", Start='2009-02-20', Finish='2009-05-30', Resource="Max")
])
fig = px.timeline(df, x_start="Start", x_end="Finish", y="Task", color="Resource")
fig.update_yaxes(autorange="reversed")
fig.update_traces(width=0.025) #This update generates a thinner bar (like a line) instead of the default horizontal bar
fig.show()
To get help on px.timeline()
print help(px.timeline)
Thanks very much ! I am using that now.
However, my start and end times are in UTC, ISO format, but the graph has a weird X axis (1000AD - 6000AD) rather than the last two days !!
With a flurry around August 2020 !
data sample:
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-15T17:28:19+00:00’, ‘end’: ‘2020-08-15T17:33:01+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T15:43:56+00:00’, ‘end’: ‘2020-08-14T15:43:58+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T16:40:26+00:00’, ‘end’: ‘2020-08-14T17:34:42+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T15:29:49+00:00’, ‘end’: ‘2020-08-14T15:29:50+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T16:38:57+00:00’, ‘end’: ‘2020-08-14T16:40:24+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T15:30:00+00:00’, ‘end’: ‘2020-08-14T15:30:01+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T15:29:56+00:00’, ‘end’: ‘2020-08-14T15:29:57+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T17:34:46+00:00’, ‘end’: ‘2020-08-14T18:42:31+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T15:29:58+00:00’, ‘end’: ‘2020-08-14T15:29:59+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-17T21:10:55+00:00’, ‘end’: ‘2020-08-17T21:13:28+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-14T14:55:53+00:00’, ‘end’: ‘2020-08-14T15:50:45+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-17T19:52:10+00:00’, ‘end’: ‘2020-08-17T19:54:42+00:00’}
{‘instance’: ‘i-123’, ‘start’: ‘2020-08-17T19:15:34+00:00’, ‘end’: ‘2020-08-17T19:18:06+00:00’}
I converted df[‘start’] and df[‘end’] , via pd.to_datetime() and here is what I got:
import plotly.express as px
import pandas as pd
d = [{'instance': 'i-123', 'start': '2020-08-15T17:28:19+00:00', 'end': '2020-08-15T17:33:01+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T15:43:56+00:00', 'end': '2020-08-14T15:43:58+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T16:40:26+00:00', 'end': '2020-08-14T17:34:42+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T15:29:49+00:00', 'end': '2020-08-14T15:29:50+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T16:38:57+00:00', 'end': '2020-08-14T16:40:24+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T15:30:00+00:00', 'end': '2020-08-14T15:30:01+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T15:29:56+00:00', 'end': '2020-08-14T15:29:57+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T17:34:46+00:00', 'end': '2020-08-14T18:42:31+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T15:29:58+00:00', 'end': '2020-08-14T15:29:59+00:00'},
{'instance': 'i-123', 'start': '2020-08-17T21:10:55+00:00', 'end': '2020-08-17T21:13:28+00:00'},
{'instance': 'i-123', 'start': '2020-08-14T14:55:53+00:00', 'end': '2020-08-14T15:50:45+00:00'},
{'instance': 'i-123', 'start': '2020-08-17T19:52:10+00:00', 'end': '2020-08-17T19:54:42+00:00'},
{'instance': 'i-123', 'start': '2020-08-17T19:15:34+00:00', 'end': '2020-08-17T19:18:06+00:00'}]
df = pd.DataFrame(d)
for col in ['start', 'end']:
df[col] = pd.to_datetime(df[col])
df['id'] = [idx for idx in df.index]
print(df['start'].min(), df['end'].max())
fig = px.timeline(df, x_start="start", x_end="end", y="id")
fig.update_xaxes(range=[df['start'].min(), df['end'].max()]) #THIS IS A NECESSARY UPDATE!!!!
fig.update_traces(width=0.15)
fig.show()
Data from the last two rows of the dataframe are not displayed. This is the time difference:
df['end'] - df['start']
0 00:04:42
1 00:00:02
2 00:54:16
3 00:00:01
4 00:01:27
5 00:00:01
6 00:00:01
7 01:07:45
8 00:00:01
9 00:02:33
10 00:54:52
11 00:02:32
12 00:02:32
dtype: timedelta64[ns]
@empet ,
For same dataset if I want to plot Dates on Y axis and time i.e. 00 to 24 hrs on x axis, then how can I plot horizontals lines.
Or for a particular date 2020-08-14, if I can plot start and end time on x axis. And then I can put a loop to plot other dates where start & end time can overlap, instead of incremental over y axis.
Objective is to understand during what time their is no instant and I have 3 months data.
Thanks