search

berkeley-dsep-infra / datahub

JupyterHubs for use by Berkeley enrolled students
https://docs.datahub.berkeley.edu
BSD 3-Clause "New" or "Revised" License
65 stars 39 forks source link

Datahub python interpreter has problems displaying Pandas dataframes #5581

Open jdmarshl opened 8 months ago

jdmarshl commented 8 months ago

Bug description

My students I and encountered this problem starting last week. When you call a dataframe, either by name or using df.head(), there is an error message that indicates an indexing problem. Does not happen locally. I am attaching a pdf that shows the output. It seems to do it whenever we call a df object. 14_MoreClassifiers_solutions.pdf

Environment & setup

How to reproduce

create a Pandas dataframe call the dataframe, either by itself or using .head() interpreter displays error message saying that it has an indexing problem across the columns

jdmarshl commented 8 months ago

even did it for the following code block which is there as a kind of toy example

birds = pd.DataFrame({'color':['white', 'grey', 'white'], 'quacks':['yes', 'yes', 'no'], 'type':['duck', 'duck', 'goose']}) birds

jdmarshl commented 8 months ago

Further note--I restarted the kernel, and dataframes displayed as expected. The problem seems to occur when the students or I have been working for awhile in a notebook and then try to display a dataframe. I cannot explain the difference in behavior when I restart the kernel. I will tell students to do that for now.

shaneknapp commented 8 months ago

hey @jdmarshl ! i tried to replicate this earlier today, but was unable to. if this happens again, could you let us know here immediately. 

The problem seems to occur when the students or I have been working for awhile in a notebook and then try to display a dataframe. I cannot explain the difference in behavior when I restart the kernel.

yeah, this is really strange and i'm at a loss to explain it.

jdmarshl commented 8 months ago

I will be careful to see what the state of my (or someone else's) notebook is when the problem occurs. It does not happen when I start the kernel and run a notebook all the way through. I will check on what's happening when students are doing the their labs and let you know when I learn more. jdm

Jon Marshall, Ph.D. (he/him) Assistant Dean JSP/Legal Studies

On Mon, Mar 4, 2024 at 11:46 AM shane @.***> wrote:

hey @jdmarshl https://github.com/jdmarshl ! i tried to replicate this earlier today, but was unable to. if this happens again, could you let us know here immediately.

The problem seems to occur when the students or I have been working for awhile in a notebook and then try to display a dataframe. I cannot explain the difference in behavior when I restart the kernel.

yeah, this is really strange and i'm at a loss to explain it.

— Reply to this email directly, view it on GitHub https://github.com/berkeley-dsep-infra/datahub/issues/5581#issuecomment-1977337175, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHXFERYPCXH2W66OSPB7RXDYWTFSTAVCNFSM6AAAAABEEKBX4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZXGMZTOMJXGU . You are receiving this because you were mentioned.Message ID: @.***>