Instructions to use dzungpham/graphcodebert-code-classification with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use dzungpham/graphcodebert-code-classification with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("dzungpham/graphcodebert-code-classification", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| 2026-04-15 07:19:05,558 - INFO - __main__ - Logging to file: ./taskA-codebert-base/training.log | |
| 2026-04-15 07:19:05,563 - INFO - __main__ - Loading tokenizer from: microsoft/codebert-base | |
| 2026-04-15 07:19:05,823 - INFO - __main__ - Loading model from: microsoft/codebert-base | |
| 2026-04-15 07:19:06,547 - INFO - __main__ - Model loaded successfully | |
| 2026-04-15 07:19:06,551 - INFO - __main__ - Model config - num_labels: 2, hidden_size: 768 | |
| 2026-04-15 07:19:06,698 - INFO - __main__ - Model on device: cuda | |
| 2026-04-15 07:19:06,702 - INFO - __main__ - Freezing base model weights, training classifier head only... | |
| 2026-04-15 07:19:06,707 - INFO - __main__ - Base model frozen - only classifier head is trainable | |
| 2026-04-15 07:19:06,710 - INFO - __main__ - Loading datasets from Hugging Face Hub... | |
| 2026-04-15 07:19:07,548 - INFO - __main__ - Train samples: 500000, Val samples: 100000 | |
| 2026-04-15 07:19:07,552 - INFO - __main__ - Tokenizing datasets... | |
| 2026-04-15 07:22:59,746 - INFO - __main__ - W&B set to offline mode for notebook stability | |
| 2026-04-15 07:22:59,804 - INFO - __main__ - Starting training... | |