Get pandas dataframes from lists of alpaca objects

[sokol11]

Is there a utility function, or some other way to get pandas dataframes from account.list_…()? For example, api.get_barset() has the .df attribute but account.list_positions() does not. Is what converts alpaca objects to dataframes an internal alpaca function that I can import and run on the output of account.list_positions(), .list_orders(), etc? Any insight is appreciated. Thanks!

[sokol11]

In case someone else searches for this, I realized that a straightforward way to do this is:

positions = api.list_positions()
positions = pd.DataFrame({
     'ticker': [x.symbol for x in positions],
     'qty': [x.qty for x in positions], etc
})

Still, I wonder if there’s a built-in function somewhere that yields the same result.

[Derek]

While not a built-in function. I saw a solution from nad that I’ve found very useful. You flatten the object by calling vars(object) on it first. Then handle the indexing as you roll up the dataframe:

@nad

    orders = api.list_orders(status='all', limit=orderTotal, after=today)
    if not orders:
        return
    dfOrders = pd.DataFrame()
    for o in orders:
        # convert dot notation to dict
        d = vars(o)
        # import dict into dataframe
        df = pd.DataFrame.from_dict(d, orient='index')
        # append to dataframe
        dfOrders = dfOrders.append(df, ignore_index=True)

[elTexacan]

@sokol11

[x.symbol for x in positions]

do you mean…

positions = api.list_positions()
positions = pd.DataFrame({'ticker': [1.symbol], 'qty': [2.qty],.....})

sry i know this is probably a silly question

[elTexacan]

lol that looks way easier but yet i’ve tried it and still can’t get it working in colab. ugh i’m a noob.

[Sean_Corcoran]

try pd.DataFrame([position._raw for position in positions])