(venv) PS D:\Zidan\1. Machine Learning\sum_liputan6> python train_evaluate.py --model bert --train_limit 2000 --eval_limit 200
Warning: rouge-score tidak tersedia. Install dengan: pip install rouge-score
=== Training BERT Model ===
Loading training data...
Training data: 2000 articles
2025-10-02 13:20:14.631655: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2025-10-02 13:20:18.591268: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
Extracting BERT features...
Training classifier...
Training completed in 1375.99 seconds

Evaluating on dev set...

=== Evaluating on DEV set ===
Evaluating on 200 articles...
Extractive Performance:
  accuracy: 0.7964
  precision: 0.3675
  recall: 0.3475
  f1: 0.3572
ROUGE Scores:
  rouge-1: 0.0000
  rouge-2: 0.0000
  rouge-l: 0.0000

Model saved to saved_models/bert_model.pkl