Text Classification REST API

dataset.py

Data model

{
  "t": <string>,
  "c": <number>
}

Running python dataset.py will create 3 separate files:

1. data.csv contains all data
2. train_data.csv contains 80% of the dataset (meant for training)
3. test_data.csv contains 20% of the dataset (meant for testing)

svm.py

Running python svm.py will train the SVM and print classification report using train_data.csv and test_data.csv.

textclf.py

Initialize job queue:

celery -A textclf worker

License

ISC