Sentiment Analysis for tweets
After creating virtual environment, install python version>=3.8
conda create -n sentiment
Install all the required dependencies from requirements.txt file
pip install -r requirements.txt
- Create folder named data
- Add training data in csv format to the folder
- run training.py script
python training.py
uvicorn app:app --reload
Hit the url with the text whose sentiment needs to be identified, the api returns a dictonary with the text and its predicted sentiment
- Create support for CLI
- Make the pipeline adaptive to dynamic data inputs