Mindspace AI Model - Prediction & Data Preparation

This guide explains how to generate training and test data, and how to use the prediction script.

1. Generating Train and Test Data

• Place your raw dataset (e.g., sample.xlsx) in the data/ directory.
• Run the following script to process the data and generate train.csv and test.csv:

python src/dataset.py

This will:

• Read the Excel file.
• Prepare text features and labels.
• Split the data into training and test sets.
• Save train.csv and test.csv in the data/ folder.

2. Running Predictions

• Make sure your trained model is available in the models/distilbert directory.
• Run the prediction script:

python src/predict.py

This will:

• Load the trained model.
• Predict the label for a sample input.
• Print the prediction result and available labels.

---

For more details, check the respective Python scripts in the src/ folder.