Is Float64 not supported in Streamlit?

🎈 Using Streamlit
1

I was trying to view my dataset using streamlit. While I could view simple datasets easily, I have one containing Float-64 values…

And even when using my classifier, I got all the predict value same…!

image
image745×177 6.88 KB

while in real, it has data like this:
image

And the problem is that all it shows for my data is null values throughout the dataset.

What might be the solution for this? or what might’ve been the problem? because, for other datasets, it shows their values correctly… like this Iris dataset:

image
image758×249 14.5 KB

2

Hi @asmeedhungana, welcome to the Streamlit community!

There’s a couple of distinctions here to make. First, what is displayed as a table isn’t necessarily going to exactly represent the underlying data, as only some number of significant digits will be displayed. Because you are using such small values, they do come out to 0.0000 for the most part at the level of significant digits.

Separately, if your classifier comes out incorrect, it could be one of many things, none of which are Streamlit-specific. For example, how you’re reading the data into Python could be an issue (either using pandas or numpy), or your classifier might not have the right arguments being passed to it.

But as I mentioned, this is a general Python debugging issue, not a Streamlit one. Good luck! :slight_smile:

1 Like
3

ohh alright! Thanks for the quick response @randyzwitch !
I was worried because my task is only to make a simple gui for the ml project my friends are doing and when I found out about streamlit and tried it, it was just WOW! I don’t have much background in python and ML…
I nearly got shattered thinking I might not be able to do it in streamlit and would have to learn flask and all for the demo! Thank god it’s not the case!
:relieved:
I’m bugging my friends now! :rofl:

1 Like