50MB dataset limitation when using Plotly.py

Hi there,

I’m using Plotly.py to show figures in my Streamlit app with big datasets. And now I get a notification of dataset oversizing, bigger than 50.0MB, is there anyway to solve this?

the exact error message is “RuntimeError: Data of size 83.4MB exceeds write limit of 50.0MB”

Check this:

Thanks for the reply! I found this github link before I posted this issue, but it didn’t solve my problem.

But later my friends and I found this error message in ~/site-packages/streamlit/server/server_util.py, and changed the parameter MESSAGE_LIMIT_SIZE to 200*1e6, making the write limit now 200 MB

1 Like

@imClumsyPanda I did the same, but it still doesn’t work for me.
Please any help

Hi,

Have you tried to re-do streamlit run again after the file changed and saved?

1 Like

Yes, i rerun the app again but still doesn’t work.
Please, i have another question is their in posibility to run a file with GB size.

If you mean upload file with GB size, you can try “streamlit run xx.py —server.maxUploadSize=1028”

Do you have multiple python environments? That might cause the modification doesn’t work well

1 Like

No i have only one envirement

You can use “where streamlit” command to check if the streamlit.exe is in the same path with the server_util

2 Likes

Thank you for your response. Actually when i restart my machine it works.
And i can display data up to 200 Mo

We recently introduced some changes with 1.4.0 to fix this problem. The default message size limit is now 200 MB (instead of 50 MB) and you can increase it even further with the config option server.maxMessageSize. More about config options here. Let me know if this works or not!