- A pre-trained classifier model saved as a pickle file.
- This model is used to classify intents and provide additional metadata for SQL-related queries.
- A standalone script to test the classifier.
- It loads the classifier model and classifies a given prompt.
- A Jupyter Notebook that contains the pipeline for training and evaluating the intent classifier.
- It includes data preprocessing, model training, and evaluation steps.
To test the classifier using the classifier_test.py
script:
- This script will load the classifier model from
combined_sql_classifier.pkl
and classify a given prompt.
To run and train the classifier using the Jupyter Notebook:
- Open
Final_Intent_Classifier_Pipeline.ipynb
in Jupyter Notebook or Google Colab. - Follow the steps in the notebook to preprocess the data, train the model, and evaluate its performance.