Unexpected behaviour Python recipe

SanderdeQueleri

Weird Python behavior:

When running a large script and printing the values just before assigning the dataframe to one of the output files i ran into this weird behavior that's returning a completely unexpected output. Here is the code i'm running, the printed output of transaction_type column and then the actual dataset that is missing the value.

1. Code for output_df that is used as input for the leading_forecast dataiku output dataset

 

 2. Print statement of the output_df, as you can see it contains V and C

3. Output dataframe in Dataiku:

 

As you can see it only contains C, and i would expect here to see atleast the V from the printstatement and a B from elsewhere in the code. How can this dataset be different from the print statement as it is literally taken from there...

Can someone please inform me why this is happening

---

Operating system used: Windows 10

Alexandru

Hi @SanderdeQueleri

As mentioned on support as well this can be due to the pandas version used. Since DSS 9.0.3 only supports up to pandas 1.0.

Support for support for Pandas 1.1, Pandas 1.2 and Pandas 1.3 was added in https://doc.dataiku.com/dss/latest/release_notes/10.0.html#version-10-0-4-march-7th-2022

Thanks