Predicting emotions for unseen tweets

[khairiyah98]

Hi, can I know if we can use these to predict emotions for any tweets? Or do we need to train them downstream?

[nikicc]

Hi, can I know if we can use these to predict emotions for any tweets?

Yes, the models are already trained and can be used directly.

Or do we need to train them downstream?

No need. They are already trained.

[khairiyah98]

Alright, thank you!

May I know what are the preprocessing steps required for the tweet input into the model? For e.g. Do I need to remove emoticons/emojis?

[nikicc]

@khairiyah98 I don't think you need to do any preprocessing. Feel free to leave the emoticons in.

[Akshar-code]

Hii @nikicc in your demo.ipynb file i can see you fed the tweets are a tuple

Im wondering how i can do this if i have a csv file with rows of tweets? please help me

Also @khairiyah98 have you made any progress with this aspect?

[nikicc]

Hi @Akshar-code!

Hii @nikicc in your demo.ipynb file i can see you fed the tweets are a tuple

It's actually a list, not a tuple. But tuple might also work, not sure 🤷‍♂️.

Im wondering how i can do this if i have a csv file with rows of tweets? please help me

So if you have a file with the structure as shown in the above image, you just need to read it line by line into a list of strings. Check this StackOverflow answers for code snippets that do exactly this.