needing help with index error:

$9.$Visualize the first five sentences embedding for each tweet using PCA

import spacy

# Load the large model

nlp $=$ spacy.load$("$en$\_$core$\_$web$\_$lg$")$

tweets.index $=$ range$(0,$len$($tweets$))$

#data$\_$bc $=[]$

data$\_$ac $=[]$

# Generate only the first $5$ sentence$/$doc embeddings

top$\_$n $=5$

for idx, row in tweets.iterrows$()$:

if idx $$ top$\_$n:

data$\_$ac$.$append$([$row$[\text{'}$tweet$\text{'}],$ nlp$($row$[\text{'}$tweet$\text{'}]).$vector, row$[\text{'}$target$\text{'}]])$

else:

break

#Store the results in the dfs

df$\_$ac $=$ pd$.$DataFrame$($data$\_$ac$,$ columns$=[\text{'}$review$\_$ac$\text{'},$ "vector", "target"$])$

pprint$($df$\_$ac$)$

from sklearn.decomposition import PCA

df$\_$ac$.$index $=$ range$(0,$len$($df$\_$ac$))$

X$\_$ac $=$ np$.$array$($list$($df$\_$ac$["$vector$"]))$

# Extract $2$ principal componets using PCA

pca $=$ PCA$($n$\_$components$=2)$

pca$\_$ac$\_2$d $=$ pca.fit$\_$transform$($X$\_$ac$)$

review$\_$ac $=$ df$\_$ac$.$review$\_$ac

fig, ax $=$ plt$.$subplots$(2,1,$ figsize $=(17\mathrm{.}8,10*2))$

fig.suptitle$(\text{'}$Visualize Sentence Vectors Using PCA', fontsize$=24)$

# Color the sentence embeddings based on the airline sentiment

colors $=\{\text{'}4\text{'}$:'green', $\text{'}0\text{'}$:'red'$\}$

ax$[1].$scatter$($pca$\_$ac$\_2$d$[$:$,0],$pca$\_$ac$\_2$d$[$:$,1],$ s $=100,$ c $=$ df$\_$ac$["$target$"].$map$($colors$),$

cmap $=$ "coolwarm", edgecolor $=$ "None", alpha$=0\mathrm{.}5)$

# Label the points using the texts

for idx, txt in enumerate$($review$\_$ac$)$:

ax$[1].$annotate$($txt$,($pca$\_$ac$\_2$d$[$idx$,0],$pca$\_$ac$\_2$d$[$idx$,1]))$ #IndexError: index $5$ is out of bounds for axis $0$ with size $5$

ax$[1].$set$\_$title$(\text{'}$Scatter Plot for Cleaned Texts', fontsize$=20)$

plt$.$show$()$

example of df:

example of df:

review$\_$ac $\backslash$

$0$ cant sleep gahh. better try again atleast. ni$...$

$1$ @zincous It's still in the works... I promise

$2$ @realAdamBeyer how can I get drumcode products...

$3$ @Cupcake$\_$Smile Thanks $.$ i appreciate that $.$

$4$ @$\_$MeesLovesYou$\_$ Grounded from the computer... $...$

vector target

$0[1\mathrm{.}5666437,2\mathrm{.}5385,-4\mathrm{.}0792966,-1\mathrm{.}8616177,-0...0$

$1[0\mathrm{.}35162392,2\mathrm{.}2831438,-1\mathrm{.}9513298,-1\mathrm{.}5335256...4$

$2[-0\mathrm{.}29363254,2\mathrm{.}0723453,-4\mathrm{.}4491,-1\mathrm{.}4692746,...4$

$3[0\mathrm{.}84563166,-1\mathrm{.}7277744,-2\mathrm{.}7835429,-2\mathrm{.}918788...4$

$4[-0\mathrm{.}7086367,-0\mathrm{.}63313067,-2\mathrm{.}168379,-0\mathrm{.}609265...0$

dataset head:

target tweet len$\_$target count$\_$word

$0$ cant sleep gahh. better try again atleast. ni$...508$

$4$ @zincous It's still in the works... I promise $468$

$4$ @realAdamBeyer how can I get drumcode products... $629$

$4$ @Cupcake$\_$Smile Thanks $.$ i appreciate that $.447$

$0$ @$\_$MeesLovesYou$\_$ Grounded from the computer... $...618$