PROBLEM $1$: Sentiment Analysis$(50$ points$)$

Dataset: We$$ll use the Sentiment Labelled Sentences Data Set from the UCI Machine Learning Repository. This dataset includes $3,000$ sentences labeled as positive or negative, collected from three sources: IMDb, Amazon, and Yelp. Each source contributes $1,000$ sentences. $[$Download Link$]$

We will use Pytorch to implement and train a Multi$-$layer Perceptron $($MLP$)$ on this dataset. Please follow the instructions of the code templete $($p$1-$template.py$)$ and implement the parts denoted by $$Your code here$.$ The output of your implementation should like follows:

First $5$ rows of the dataset:

sentence label

$0$ So there is no way for me to plug it in here i$...0$

$1$ Good case, Excellent value. $1$

$2$ Great for the jawbone. $1$

$3$ Tied to charger for conversations lasting more... $0$

$4$ The mic is great. $1$

Missing values in each column:

sentence $0$

label $0$

dtype: int$64$

Number of duplicate rows:

$17$

Label distribution:

label

$11386$

$01362$

Name: count, dtype: int$64$

Epoch $[(1)/(20)],$ Loss: $0\mathrm{.}6935$

Epoch $[(2)/(20)],$ Loss: $0\mathrm{.}6883$

Epoch $[(3)/(20)],$ Loss: $0\mathrm{.}6784$

Epoch $[(4)/(20)],$ Loss: $0\mathrm{.}6634$

Epoch $[(5)/(20)],$ Loss: $0\mathrm{.}6372$

Epoch $[(6)/(20)],$ Loss: $0\mathrm{.}6035$

Epoch $[(7)/(20)],$ Loss: $0\mathrm{.}5657$

Epoch $[(8)/(20)],$ Loss: $0\mathrm{.}5188$

Epoch $[(9)/(20)],$ Loss: $0\mathrm{.}4699$

Epoch $[(10)/(20)],$ Loss: $0\mathrm{.}4224$

Epoch $[(11)/(20)],$ Loss: $0\mathrm{.}3816$

Epoch $[(12)/(20)],$ Loss: $0\mathrm{.}3367$

Epoch $[(13)/(20)],$ Loss: $0\mathrm{.}3002$

Epoch $[(14)/(20)],$ Loss: $0\mathrm{.}2716$

Epoch $[(15)/(20)],$ Loss: $0\mathrm{.}2457$

Epoch $[(16)/(20)],$ Loss: $0\mathrm{.}2331$

Epoch $[(17)/(20)],$ Loss: $0\mathrm{.}2210$

Epoch $[(18)/(20)],$ Loss: $0\mathrm{.}2027$

Epoch $[(19)/(20)],$ Loss: $0\mathrm{.}2035$

Epoch $[(20)/(20)],$ Loss: $0\mathrm{.}1813$

Evaluation Metrics:

Accuracy : $76\mathrm{.}36\%$

However, feel free to use other models or training techniques as long as the final accuracy improves. This program should be able to run on the local cpu.

PROBLEM $1$: Sentiment Analysis$(50$ points$)$

Dataset: We'll use the Sentiment Labelled Sentences Data Set from the UCI Machine Learning

Repository. This dataset includes $3,000$ sentences labeled as positive or negative, collected from

three sources: IMDb, Amazon, and Yelp. Each source contributes $1,000$ sentences. $[$Download

Link$]$

We will use Pytorch to implement and train a Multi$-$layer Perceptron $($MLP$)$ on this dataset.

Please follow the instructions of the code templete $($

p$1-$template.py$)$ and implement the parts

denoted by "Your code here". The output of your implementation should like follows:

First $5$ rows of the dataset:

sentence label

$0$ So there is no way for me to plug it in here i$...0$

$1$ Good case, Excellent value. $1$

$2$ Great for the jawbone. $1$

$3$ Tied to charger for conversations lasting more... $0$

$4$

The mic is great. $1$

Missing values in each column:

sentence $0$

label $0$

dtype: int$64$

Number of duplicate rows:

$17$

Label distribution:

label

$11386$

$0,1362$

Name: count, dtype: int$64$

Epoch $[(1)/(20)],$ Loss: $0\mathrm{.}6935$

Epoch $[(2)/(20)],$ Loss: $0\mathrm{.}6883$

Epoch $[(3)/(20)],$ Loss: $0\mathrm{.}6784$

Epoch $[(4)/(20)],$ Loss: $0\mathrm{.}6634$

Epoch $[(5)/(20)],$ Loss: $0\mathrm{.}6372$

Epoch $[(6)/(20)],$ Loss: $0\mathrm{.}6035$

Epoch $[(7)/(20)],$ Loss: $0\mathrm{.}5657$

Epoch $[(8)/(20)],$ Loss: $0\mathrm{.}5188$

Epoch $[(9)/(20)],$ Loss: $0\mathrm{.}4699$

Epoch $[(10)/(20)],$ Loss: $0\mathrm{.}4224$

Epoch $[(11)/(20)],$ Loss: $0\mathrm{.}3816$

Epoch $[(12)/(20)],$ Loss: $0\mathrm{.}3367$

Epoch $[(13)/(20)],$ Loss: $0\mathrm{.}3002$

Epoch $[(14)/(20)],$ Loss: $0\mathrm{.}2716$

Epoch $[(15)/(20)],$ Loss: $0\mathrm{.}2457$

Epoch $[(16)/(20)],$ Loss: $0\mathrm{.}2331$

Epoch $[(17)/(20)],$ Loss: $0\mathrm{.}2210$

Epoch $[(18)/(20)],$ Loss: $0\mathrm{.}2027$

Epoch $[(19)/(20)],$ Loss: $0\mathrm{.}2035$

Epoch $[(20)/(20)],$ Loss: $0\mathrm{.}1813$

Evaluation Metrics:

Accuracy :( $76\mathrm{.}36)/(\%)$

However, feel free to use other models or training techniques as long as the final accuracy

improves. This program should be able to run on the local cpu.