NER_training

This repository can be used to train NER models.

Setup Instructions

1. Install Dependencies: Ensure you have all the necessary dependencies installed by running the following command:

   pip install -r requirements.txt

2. Prepare Training Data: The training data must be kept in the data folder combined into a single file named train.csv. The CSV file must follow the same structure as this HF dataset.

3. Prepare Test Data (Optional): Another CSV file named test.csv can be added to the data folder for evaluation metrics. If test.csv is not present, the evaluation will be done on train.csv itself. The evaluation results will be populated in the logs folder.

4. Run Training: Once the data is prepared, run the training script using the following command:

   python3 '/trainer/train_bert.py'

Follow these instructions to set up and train your NER model using this repository.