Download these three files from Google Drive, into the project directory:
They can be quite large as they are pretrained transformers (currently working on reducing the size)
This uses a few ML models along with some naive human-made NLP features to attempt to scrape a day's tweets for newly announced startups. It's built with snscrape and generates a .txt file with ≤100 URLs and tweets from 24 to 48 hours ago.
Run the get_tweets.py. It takes around ~30 minutes to pull and run the model, with most of the time being taken up by the ML classifiers and transformers.
Download these three files from Google Drive, into the project directory:
They can be quite large as they are pretrained transformers (currently working on reducing the size)
This uses a few ML models along with some naive human-made NLP features to attempt to scrape a day's tweets for newly announced startups. It's built with snscrape and generates a .txt file with ≤100 URLs and tweets from 24 to 48 hours ago.
Run the get_tweets.py. It takes around ~30 minutes to pull and run the model, with most of the time being taken up by the ML classifiers and transformers.
