Hello. I$$m getting a couple of errors below when I try to create base line RNN model and second RNN model. Any sugguestions on how to fix?

You are a data scientist in an AI company. You are given a dataset of restaurant reviews. This is the sentiment$140$ dataset. It contains $1,600,000$ tweets with six columns:

$1.$ target: the polarity of the tweet $(0=$ negative, $2=$ neutral, $4=$ positive$)$

$2.$ ids: The id of the tweet $($ for example, $2087)$

$3.$ date: the date of the tweet $($for example, Sat May $1623$:$58$:$44$ UTC $2009)$

$4.$ flag: The query $($lyx$).$ If there is no query, then this value is NO$\_$QUERY.

$5.$ user: the user that tweeted $($for example, robotickilldozr$)$

$6.$ text: the text of the tweet $($for example, $"$Lyx is cool"$)$

Data Source: Sentiment$140$ DatasetLinks to an external site.

The target is the polarity of the tweet. The features are the text. You are asked to perform sentiment analysis using deep learning by using the attached Jupyter Notebook and writing a Python script, and running all the cells. You only need to submit a JupyterNotebook.

$1.$ Download the dataset that is about $81$ MB from Kaggle into the local disk and unzip it$.$

$2.$ Clean and preprocess the text data and split into training and test dataset.

$3.$ Build a baseline RNN model using embedding layer and GRU on the training dataset and evaluate it on the test dataset.

$4.$ Build a second RNN model using embedding layer and LSTM and evaluate it on the test dataset.

$5.$ Build a third RNN model using embedding layer and GRU and LSTM and evaluate it on the test dataset.

$6.$ Which model do you recommend for the model in Q$3,$ Q$4,$ and Q$5?$ Justify your answer.

import zipfile

# Unzip the downloaded file

with zipfile.ZipFile$(\text{'}$path$\_$to$\_$downloaded$\_$file.zip', $\text{'}$r$\text{'})$ as zip$\_$ref:

zip$\_$ref.extractall$(\text{'}$path$\_$to$\_$extract$\_$dataset'$)$

import pandas as pd

from sklearn.model$\_$selection import train$\_$test$\_$split

import re

from tensorflow.keras.preprocessing.text import Tokenizer

from tensorflow.keras.preprocessing.sequence import pad$\_$sequences

# Load the dataset into a pandas DataFrame

data$\_$path $=$ 'path$\_$to$\_$extract$\_$dataset$/$training$\mathrm{.}1600000.$processed.noemoticon.csv$\text{'}$

columns $=[\text{'}$target$\text{'},$ 'ids', 'date', 'flag', 'user', 'text'$]$

df $=$ pd$.$read$\_$csv$($data$\_$path, encoding$=$'latin$-1\text{'},$ header$=$None, names$=$columns$)$

# Clean the text by removing unnecessary characters

df$[\text{'}$text$\text{'}]=$ df$[\text{'}$text$\text{'}].$apply$($lambda x: re$.$sub$($r$\text{'}[^\backslash$w$\backslash$s$]\text{'},\text{'}\text{'},$ str$($x$).$lower$()))$

# Split the data into training and test datasets

train$\_$df$,$ test$\_$df $=$ train$\_$test$\_$split$($df$,$ test$\_$size$=0\mathrm{.}2,$ random$\_$state$=42)$

from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Embedding, GRU, Dense

# Tokenize the text data

tokenizer $=$ Tokenizer$()$

tokenizer.fit$\_$on$\_$texts$($train$\_$df$[\text{'}$text$\text{'}])$

vocab$\_$size $=$ len$($tokenizer$.$word$\_$index$)+1$

# Convert text data to sequences

train$\_$sequences $=$ tokenizer.texts$\_$to$\_$sequences$($train$\_$df$[\text{'}$text$\text{'}])$

test$\_$sequences $=$ tokenizer.texts$\_$to$\_$sequences$($test$\_$df$[\text{'}$text$\text{'}])$

# Pad sequences to a fixed length

max$\_$len $=100$

train$\_$data $=$ pad$\_$sequences$($train$\_$sequences, maxlen$=$max$\_$len, padding$=$'post'$)$

test$\_$data $=$ pad$\_$sequences$($test$\_$sequences, maxlen$=$max$\_$len, padding$=$'post'$)$

# Build the baseline RNN model

model$1=$ Sequential$()$

model$1.$add$($Embedding$($vocab$\_$size, $100,$ input$\_$length$=$max$\_$len$))$

model$1.$add$($GRU$(64))$

model$1.$add$($Dense$(1,$ activation$=$'sigmoid'$))$

model$1.$compile$($optimizer$=$'adam', loss$=$'binary$\_$crossentropy', metrics$=[\text{'}$accuracy$\text{'}])$

# Train the model

model$1.$fit$($train$\_$data, train$\_$df$[\text{'}$target$\text{'}],$ validation$\_$data$=($test$\_$data, test$\_$df$[\text{'}$target$\text{'}]),$ epochs$=5)$

Error$*****$ above$-$ValueError: Unrecognized keyword arguments passed to Embedding: $\{\text{'}$input$\_$lenght': $100\}$

$$

$$ from tensorflow.keras.layers import LSTM

$$

$$ # Build the second RNN model

$$ model$2=$ Sequential$()$

$$ model$2.$add$($Embedding$($vocab$\_$size, $100,$ input$\_$length$=$max$\_$len$))$

$$ model$2.$add$($LSTM$(64))$

$$ model$2.$add$($Dense$(1,$ activation$=$'sigmoid'$))$

$$

$$ model$2.$compile$($optimizer$=$'adam', loss$=$'binary$\_$crossentropy', metrics$=[\text{'}$accuracy$\text{'}])$

$$

$$ # Train the model

$$ model$2.$fit$($train$\_$data, train$\_$df$[\text{'}$target$\text{'}],$ validation$\_$data$=($test$\_$data, test$\_$df$[\text{'}$target$\text{'}]),$ epochs$=5)$

ERROR ABOVE$***$ ValueError: Unrecognized keyword arguments passed to Embedding: $\{\text{'}$input$\_$length': $100\}$

$$

# Evaluate the models on the test dataset

$\_,$ acc$1=$ model$1.$evaluate$($test$\_$data, test$\_$df$[\text{'}$target$\text{'}])$

$\_,$ acc$2=$ model$2.$evaluate$($test$\_$data, test$\_$df$[\text{'}$target$\text{'}])$

$\_,$ acc$3=$ model$3.$evaluate$($test$\_$da