pandas df.loc[z,x]=y how to improve speed?

Question

I have identified one pandas command

timeseries.loc[z, x] = y

to be responsible for most of the time spent in an iteration. And now I am l

Answer 1

if you are adding rows inside a loop consider thses performance issues; for around first 1000 to 2000 records "my_df.loc" performance is better and gradually it is become slower by increasing the number of records in loop.

If you plan to do thins inside a big loop(say 10M‌ records or so) you are better to use a mixture of "iloc" and "append"; fill a temp datframe with iloc untill the size gets around 1000, then append it to the original dataframe, and empy the temp dataframe. this would boost your performance around 10 times