whats wrong with this code?

import numpy as np

def main$($filename$,$filter$\_$value,type$\_$of$\_$card$)$:

$"""$read dataset and stores transaction records"""

data$=$read$\_$csv$($filename$)$

#$[$ALANA MAKE SURE U READ THIS LATER$]$

filtered$\_$data$=$filter$\_$data$($data$,$filter$\_$value,type$\_$of$\_$card$)$

display$\_$results$($filtered$\_$data$)$

def read$\_$csv$($filename$)$:

data$=[]$

with open$($filename$,\text{'}$r$\text{'})$ as file:

reader$=$csv$.$reader$($file$)$

next$($reader$)$

for row in reader:

data.append$($row$)$

return data

def filter$\_$data$($data$,$filter$\_$value,type$\_$of$\_$card$)$:

filtered$\_$data$=[]$

for row in data:

if filter$\_$value in row and type$\_$of$\_$card in row:

filtered$\_$data.append$($row$)$

return filtered$\_$data

def display$\_$results$($filtered$\_$data$)$:

for row in filtered$\_$data:

print$($row$)$

def task$1($data$,$filter$\_$value,type$\_$of$\_$card$)$:

$"""$data contains records

filter$\_$value is an area name

type$\_$of$\_$card is name of card provider

return list containing values to $2$ decimal"""

answer$=[]$

#finding cosine difference between normal & malicious transactions

#based on IP$\_$validity$\_$score

normaltransaction$=$data$[$data$["$Transaction$\_$type"$]==$'normal'$]["$IP$\_$validity$\_$score"$]$

maltransaction$=$data$[$data$["$Transaction$\_$type"$]==$'malicious'$]["$IP$\_$validity$\_$score"$]$

#treat transactions as vectors

normalvector$=$np$.$mean$($normaltransaction$)$

malvector$=$np$.$mean$($maltransaction$)$

dotproduct$=$np$.$dot$($normalvector$,$malvector$)$

#find norm of 'normal' transaction and 'malicious' transaction vectors

norm$\_$normal$=$np$.$linalg.norm$($normalvector$)$

norm$\_$mal$=$np$.$linalg.norm$($malvector$)$

#plug into formula for cosine similarity

cosine$\_$simil$=$dotproduct$/($norm$\_$normal$*$norm$\_$mal$)$

#find cosine distance

cos$\_$dist$=1-$cosine$\_$simil

answer.append$($round$($cos$\_$dist,$2))$

#find variance for specific area

area$\_$filtered$\_$data$=$data$[$data$[\text{'}$Actual$\_$area'$]==$filter$\_$value$]$

transaction$\_$amounts$=$area$\_$filtered$\_$data$[\text{'}$Transaction$\_$amount'$]$

variance$=$np$.$var$($transaction$\_$amounts,ddof$=1)$

answer.append$($round$($variance$,2))$

#find median of authentication score for lower $25$th and upper $75$th percentile

#based on type of card

filteredcarddata$=$data$[$data$[\text{'}$Type$\_$of$\_$card'$]==$type$\_$of$\_$card$]$

authenticationscores$=$filteredcarddata$[\text{'}$Authentication$\_$score'$]$

low$25$percentile$=$np$.$percentile$($authenticationscores$,25,$interpolation$=$'lower'$)$

high$75$percentile$=$np$.$percentile$($authenticationscores$,75,$interpolation$=$'higher'$)$

lowmedian$=$np$.$median$($authenticationscores$[$authenticationscores$<=$low$25$percentile$])$

highmedian$=$np$.$median$($authenticationscores$[$authenticationscores$>=$high$75$percentile$])$

answer.append$([$round$($lowmedian$,2),$round$($highmedian$,2)])$

#filter mal transactions where 'actual' and 'origin' are different

#find elementwie product between authentication score and ip validation score

#perform correlation between resultant vector and amount column

malsusspots$=$data$[($data$[\text{'}$Transaction$\_$Type'$]==$'Malicious'$)$&$($data$[\text{'}$Actual$\text{'}]!=$data$[\text{'}$Origin$\text{'}])]$

authscores$=$malsusspots$[\text{'}$Authentication$\_$score'$]$

ipscores$=$malsusspots$[\text{'}$IP$\_$validity$\_$score'$]$

dot$\_$product$=$np$.$dot$($authscores$,$ipscores$)$

correlation$=$np$.$corrcoef$($dot$\_$product, malsusspots$[\text{'}$Transaction$\_$amount'$])[0,1]$

answer.append$($round$($correlation$,2))$

#create Nx$5$ matrix where N is number of rows in data set

#find principal component analysis $($PCA$)$ to reduce dimensionality to Nx$1$

transactiontypemap$=\{\text{'}$ATM$\text{'}$:$1,$'EFTPOS':$2,$'Internet':$3\}$

entrymodemap$=\{\text{'}$Magnetic Stripe':$1,$'Manual':$2,$'Chip Card Read':$3,\text{'}$NFC$\text{'}$:$4\}$

datacopy$[\text{'}$Transaction$\_$Type'$]=$datacopy$[\text{'}$Transaction$\_$Type'$].$map$($transactiontypemap$)$

datacopy$[\text{'}$Entry$\_$mode'$]=$datacopy$[\text{'}$Entry$\_$mode'$].$map$($entrymodemap$)$

featurematrix$=$data$\_$copy$[[\text{'}$Transaction$\_$Type','Entry$\_$mode','Transaction$\_$amount','Authentication$\_$score',$\text{'}$IP$\_$validity$\_$score'$]].$values

pca$=$PCA$($n$\_$components$=1)$

pcaresult$=$pca.fit$\_$transform$($featurematrix$)$

answer.append$($list$($np$.$round$($pcaresult$.$flatten$(),2)))$

print$($f$"$Cosine Distance: $\{$cos$\_$dist$\}")$

print$($f$"$Variance: $\{$var$\}")$

print$($f$"$Median: $\{$median$\}")$

print$($f$"$Correlation: $\{$corr$\}")$

print$($f$"$PCA: $\{$pca$\}")$