Python Dataframe to JMP Datatable without intermediate csv

[HB]

It looks like the interface between a python dataframe and a jmp datatable goes through a csv file. Is that true?

For example:

Python Init(Path("C:\Program Files (x86)\Microsoft Visual Studio\Shared\Python37_64\python37.dll"));
Python Submit( "\[import pandas as pd
df = pd.read_parquet('c:/temp/tst.parquet')
]\" );
dt = Python Get(df); // this method is saving the df as csv and then opens it in jmp. Is there a way to do that in memory
dt << New Data View;
Python Term();

In the above code I see JMP opening a progress bar for reading a csv during the Python Get(df) call. This is taking a very long time in large files.

We would like a way to import the data-frame to data-table w/o the time to save as csv and reading it

[Ryan_Gilmore]

We are archiving this request. If this is still important please comment with additional details and we will reopen. Thank you!

[hogi]

Thanks for implementing this Wish in Jmp 18

[hogi]

@Sarah-Sylvestre  - delivered?

[Ressel]

@Sarah-Sylvestre, it looks like the Python dataframe still goes via a csv to become a JMP table. Am I misunderstanding something?

[Paul_Nelson]

JMP 14-17 Python Send ( dt )   created a copy of the data table as a pandas dataframe.  

JMP 18_+ creates a jmp.DataTable object which is a live reference to the data.

Python Get ( dt ) get's a JSL reference to the Python data table object.

Python Get ( df )  creates a data table from a pandas dataframe that is a COPY of the data. 

JMP 18 - Python Get ( df )  does indeed use CSV to get a DataFrame as JMP DataTable, that was left for compatibility.  You should be making a dt from the df in Python and using the dt with Python Get().

JMP 19 adds Portable Dataframe Protocol support so that it is easy to create a JMP DataTable from a Pandas dataframe in memory without having to walk the columns and build columns yourself.  dt = jmp.from_dataframe( df )

In JMP 19 Python Get ( df ) should create a new data table which is a COPY directly from the dataframe in memory.  If it cannot automatically convert in memory for whatever reason, it falls back to CSV instead of failing.