Yahoo finance support is broken, or do we need a particular version of yfinance lib?

[trivalent]

Running a simple code with the following code results in a failure:

at = AutoTrader()
at.configure(verbosity=1, show_plot=True, feed="yahoo")
at.add_strategy("ema_crossover")
at.backtest(start_dt=datetime.now() - timedelta(days=190), end_dt=datetime.now())
at.virtual_account_config(initial_balance=100000, leverage=1)
at.run()

File "E:\workspace\python\AutoTrader\autotrader\autotrader.py", line 1468, in _main
bot = AutoTraderBot(
File "E:\workspace\python\AutoTrader\autotrader\autobot.py", line 203, in init
my_strat.create_plotting_indicators(data)
File "E:\workspace\python\AutoTrader\strategies\ema_crossover.py", line 31, in create_plotting_indicators
"data": TA.EMA(data, self.parameters["fast_ema"]),
File "C:\Users\rahul\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\finta\finta.py", line 15, in wrap
args[i] = args[i].rename(columns={c: c.lower() for c in args[i].columns})
File "C:\Users\rahul\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\finta\finta.py", line 15, in
args[i] = args[i].rename(columns={c: c.lower() for c in args[i].columns})
AttributeError: 'tuple' object has no attribute 'lower'

yfinance version: yfinance-0.2.50
AutoTrader version: latest/main branch.

[trivalent]

It appears that yFinance is appending the ticker symbol into the column names as well. Adding following lines in brokers/yahoo.py seems to fix this issue:

        # Remove excess data
        if count is not None and start_time is None and end_time is None:
            data = data.tail(count)
            
        **# fix for appending Ticker Symbol - Tested for Single Ticker download only
        if isinstance(data.columns, pd.MultiIndex):
            data.columns = data.columns.droplevel(1)
            data.drop(columns=['Adj Close'], inplace=True)

        data.columns = [col.split(" ")[-1] for col in data.columns]**

        if data.index.tzinfo is None:

However, this seems to introduce another issue:
Error when updating strategy: Invalid comparison between dtype=datetime64[ns, UTC] and datetime

which looks like the data index here is not in proper format.

[trivalent]

Adding the below looks to fix this as well:

        data.columns = [col.split(" ")[-1] for col in data.columns]

        if data.index.tzinfo is None:
            # Data is naive, add UTC timezone
            data.index = data.index.tz_localize(timezone.utc)
        else:
            # Convert to UTC
            data.index = data.index.tz_convert(timezone.utc)
            
        # fix the index as well
        **data.index = pd.to_datetime(data.index).tz_localize(None)**