Float numbers as labels at scatter plot

[defaultUser3214]

Is your feature request related to a problem? Please describe. A clear and concise description of what the problem is. Ex. I'm always frustrated when [...] I would like to plot a scatter plot with labels that are floats and not int values. I think many data scientists have the same problem and it would be great if you Y data could also be float. Describe the solution you'd like A clear and concise description of what you want to happen. The visdom experienced an exception while running: AssertionError('labels should be integers') If you use float values as y values for a scatter plot from the example demo.py https://github.com/facebookresearch/visdom/blob/master/example/demo.py Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered. I tried to do this workaround for i in range(0, len(data["labels"])): data["labels"] = int(data["labels"] * 100000)

Y = np.random.rand(100)

old_scatter = viz.scatter(

    Y = data["labels"],
    X = [list(a) for a in zip(data["top"], data["middle"])],

which resulted in a The visdom experienced an exception while running: TypeError("cannot convert the series to <class 'int'>")

Additional context Add any other context or screenshots about the feature request here.

[JackUrb]

Hi @defaultUser3214 - labels are a fairly complicated component of the scatter implementation, as they're used to determine indexes into the legend and colormap. Supporting other datatypes would lead to a further increase in complexity where we'd need a map between labels and indices. It could be potentially possible, but I don't necessarily see a clear and clean solution for doing this.

[SysOverdrive]

Is there any other way to display/load float numbers on the side of a scatter plot ?

For example when calculating some metrics (dice/f score) of a prediction and you want to add that to the scatter plot in a corner?