we need to prepare a DataFrame that counts the number of non$-$zero precipitation hours each day. This is done for you below when creating $`$daily$\_$counts$`$

daily$\_$counts $=($

df$.$groupby$($pd$.$Grouper$($key$="$dt$",$ freq$="$D$"))$

$.$agg$(\{$

"precip": $[$lambda x: $($x $>0).$sum$(),$ "mean", "sum"$],$

"temp": $[\text{'}$min$\text{'},$ 'max'$],$

$\})$

$.$assign$($month$=$lambda x: x$.$index.month$)$

$).$dropna$()$

daily$\_$counts.columns $=["$n$\_$precip$\_$hours", "ave$\_$hourly$\_$precip", "daily$\_$precip", "temp$\_$min", "temp$\_$max", "month"$]$

daily$\_$counts.head$()$

Using this DataFrame and the statsmodels library, fit the appropriate GLM model to predict the number of hours with non$-$zero precipitation in a day

You shoud use as features, temp$\_$min, temp$\_$max, and a collection of indidcator variables for the month. In total you will have $14$ features.

# count$\_$model $=$ smf$.$glm$($

# $...,$ # CHANGE FORMULA

# daily$\_$counts,

# family$=...$ # CHANGE FAMILY

# $)$

# your code here

count$\_$model$\_$fit $=$ count$\_$model.fit$()$

count$\_$model$\_$fit.summary$()$