optstrategy trouble with Unix timestamps

adenosine

My dataset's datetime-stamp is formatted in Unix float notation. I've divided the datetime-stamp by 1000 so it can be passed with GenericCSVData no problem in a normal addstrategy backtest with the parameter dtformat=2. However, when trying to optimize with optstrategy, I get an error: AttributeError: Can't pickle local object 'GenericCSVData.start.<locals>.<lambda>'

I assume the dtformat=2 parameter I've set is the lambda function so I've just manually converted the Unix datetime-stamps into the %Y-%m-%d %H:%M:%S format and removed the dtformat=2 parameter. However, I then get another error that my format is still incorrect:

ValueError: time data '20/09/01 02:00:00' does not match format '%Y-%m-%d %H:%M:%S'

Is there something I'm missing here?

ab_trader

@adenosine
your date-time format is more %y/%m/%d %H:%M:%S rather than %Y-%m-%d %H:%M:%S. or %d/%m/%y %H:%M:%S since it is not clear where the year is shown.

adenosine

@ab_trader I took another stab and tried making the year more clear and get the same error.

Is there a difference between %y and %Y?

ab_trader

@adenosine said in optstrategy trouble with Unix timestamps:

Is there a difference between %y and %Y?

Sure, there is a difference. google, what is python date time string format

Also you separate the numbers by /, but in the format description you use -. This can affect it as well.

adenosine

@ab_trader You were right. It was the - that was throwing it off . Thank you.