Tm1py: function with dataframe adding decimal place for numeric data

[yyzz1010]

Describe the bug
I've tried execute_mdx_dataframe_shaped and execute_view_dataframe_shaped to generate dataframe, but there is extra decimal place for numeric data, i.e. data in TM1: 6; data in dataframe: 6.0

To Reproduce
run execute_mdx_dataframe_shaped / execute_view_dataframe_shaped to generate dataframe and compare TM1 data

Expected behavior
data in dataframe exactly same as TM1 data

Version
TM1py Version: 1.5.0
TM1 Server Version: 11.8.0.33

Additional context
these functions are using same mdx/cube view

[rkvinoth]

I could not replicate this issue. Could you provide more details.. like the MDX, the cube structure and the measure dimension element types?

[MariusWirtz]

Hi @yyzz1010,

the behavior you describe is thenatural way python/pandas works.
Pandas determines independently whether to use integers or floats for the values in the column.
I noticed that if there are no NaN or None values in the view, pandas uses integers.

However, you can overrule pandas decision-making on data-types quite easily.
Here is a sample:

``` python
import pandas as pd

from TM1py import TM1Service

with TM1Service(address="", port=12354, ssl=True, user="admin", password="apple") as tm1:
mdx = """
SELECT
{[d1].[1], [d1].[2], [d1].[3]} ON ROWS,
{[d2].[1], [d2].[2]} ON COLUMNS
FROM [c1]"""
df = tm1.cells.execute_mdx_dataframe_shaped(mdx)
print("Original:")
print(df)

df = df.fillna(0)
for col in ['1', '2']:
    df[col] = pd.to_numeric(df[col], errors='coerce').astype(int)

print("Edited:")
print(df)

it generates this output:

Original:
d1 1 2
0 1 1.0 2.0
1 2 5.0 6.0
2 3 NaN NaN
Edited:
d1 1 2
0 1 1 2
1 2 5 6
2 3 0 0

Process finished with exit code 0

```

[yyzz1010]

oh thx for the explanation Marius!! i was just using below as workaround:

for x in range(len(df.columns)):
    df[list(df.columns)[x]] = df[list(df.columns)[x]].astype("float", errors="ignore")

df.to_csv(path_or_buf=file_path, index=False, na_rep="0", float_format="%.f")