Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Self serve data - not able to filter by date

Hi Everyone
I followed the posts on loading and checking the load.
So far I have reached the point where I can access the index of the dataframe loaded.
When convenient can you please let me know if there is sample code to access only those rows for a given backtest date?
I looked at this detailed post from Dan https://www.quantopian.com/posts/access-data-in-pipeline-output-pandas-dataframe - not sure though how to only get rows for a specific date and fields within those rows.
Do I need to add a date filter in this section
def before_trading_start(context, data)
results = pipeline_output('my_pipeline')
context.pipeline_results = results

When I try for stock in context.pipeline_results.index I get a symbol that is not in the data for the backtest date I am on.
I tried context.pipeline_results.loc[:'asof_date'] == ESTdate - the result does not exclude false values.
My apologies in advance if I missed this in the documentation or in any of the posts.
Much appreciate and a pleasant day
Savio
The data has the following field labels if it helps to clarify

        symbol  limit   allocation          day    month    year    sid         asof_date           timestamp  
0   F           5.26    0.03028572  2.0    1.0        2008.0    2673    2008-01-02  2008-01-03  
1     JNJ   48.53   0.110258293 2.0    1.0        2008.0    4151    2008-01-02  2008-01-03  
1 response

There is a notebook attached to this post that may be helpful.
https://www.quantopian.com/posts/live-custom-data-what-to-expect

It seems as though you could use the .latest method in your pipeline, then sort by the asof_date column and strip out the most recent dated rows.
Still working on this myself, so I don't have a working script to share.