How to plot line graph from each row - no connection with other rows

đź“Š Plotly Python
1

Hi,

I am looking to draw a line graph from a panda dataframe. Each line (3 point line) should be from each row and won’t have any connection with the next row’s data. I looked and couldn’t find how can I do it with plotly.

Just to provide detail:

Each row has say 4 columns for each event, where event starts but there is wait after that i.e. start_datetime, end_datetime, wait_duration, run_duration.

Datetime is in date time format and duration is in seconds.

I am looking to draw datetime on x-axis and duration on y -axis.

Need to draw 2 connecting lines of different color for each row i.e. first a vertical line - start_datetime to wait_duration. Then second line connecting from end of first line i.e. wait_duration to run_duration which should be finishing at end_datetime.

Same for each row.

Just wondering if something like this will be possible in plotly, I will highly appreciate if so some code snippet can be provided to solve this. Thanks

2

Hi,

Welcome to the community! :slight_smile:

Here’s how you can create this chart row-by-row:

fig = go.Figure()
for idx, row in df.iterrows():
    fig.add_trace(
        go.Scatter(
            x=[row["start_datetime"], row["start_datetime"]],
            y=[0, row["wait_duration"]],
            mode="lines",
            line={"color": "red"},
            showlegend=False
        )
    )
    fig.add_trace(
        go.Scatter(
            x=[row["start_datetime"], row["end_datetime"]],
            y=[row["wait_duration"], row["run_duration"]],
            mode="lines",
            line={"color": "blue"},
            showlegend=False
        )
    )

As you are adding traces manually, this can be very inefficient for large datasets. Besides, I am explicitly removing the traces from the legend (or there will be one for each legend).

Another approach would be to create all the red/blue lines together by “patching together” the coordinates for line in each row with None as separator. This is for the first (red) group of lines:

red_x = []
red_y = []
for idx, row in df.iterrows():
    _x = [row["start_datetime"], row["start_datetime"]]
    _y = [0, row["wait_duration"]]
    red_x.append(_x[0])
    red_x.append(_x[1])
    red_y.append(_y[0])
    red_y.append(_y[1])
    red_x.append(None) # to separate lines, or they will be connected
    red_y.append(None) # for each adjacent row
    
    
fig.add_trace(
    go.Scatter(
        x=red_x,
        y=red_y,
        mode="lines",
        line={"color": "red"}
    )
)
3

That’s great, thanks jlfsjunior. Much Appreciated.