Q$2.$ Abstractive Summarization using T$5$: Code $[5]$

In this question, you will implement the summarization task using the pre$-$trained model T$5.$

$1.$ Load the dataset cnn$-$dailymail with $3\mathrm{.}0\mathrm{.}0$ version. Use test set for testing purposes. Since the model is pre$-$trained. vou do not need the train set.

$2.$ Using transformers library from Huggingface $\backslash (\{\}^\{3\}\backslash ),$ load the t$5-$small $\backslash (\{\}^\{4\}\backslash )$ model. Note that this will take around $240$ MB of space.

$3.$ Pre$-$process the test set articles column as per step $2$ of Q$1.$ You do not need to add start and end tokens for this question. Generate the summary of each row using the T$5$ model. You may use T$5$ForConditionalGeneration $\backslash (\{\}^\{5\}\backslash )$ for this task.

$4.$ Evaluate the model using average ROUGE$-1,$ ROUGE$-2$ and ROUGE$-$L scores on the test set.