Python (Pandas): count of records between values at a given time (Packets in Flight) -

Python (Pandas): count of records between values at a given time (Packets in Flight) -

i have dataframe based on packet logs wireless channel has indexes "transmitted" , "received" beingness respective floats of time since start.

i want distribution of how many packets in air @ time. @ moment

t_packets = pd.series([df[(df.transmitted < t) & (t < df.received)].count().max() t in range(tmax)])

this computationally nasty (~40s each iteration) haven't been able work out improve way limited pandas experience.

anyone got ideas 'better' way accomplish this?

basically, think want flatten info , create timeseries out of it.

tx = pd.dataframe(index=df.transmitted) rx = pd.dataframe(index=df.received) tx['p'] = 1 #adding packet rx['p'] = -1 #receiving packet #create time series here t = pd.concat([tx, rx]) t.sort().cumsum()

it's easy here figuring out actual times. can resample create clearer, should it.

edit: adding how resample seconds:

#assuming indexed times in seconds t.sort(inplace=true) t.index = pd.to_datetime(t.index * 10e9) #to convert nanoseconds seconds t.resample('s', how='sum').cumsum()

python pandas dataframes