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
| { | |
| "model_name": "microsoft/graphcodebert-base", | |
| "output_dir": "output_checkpoints/graphcodebert-best/", | |
| "num_epochs": 2, | |
| "max_steps": -1, | |
| "batch_size": 256, | |
| "learning_rate": 1e-06, | |
| "max_length": 512, | |
| "num_labels": 2, | |
| "use_wandb": true, | |
| "freeze_base": true, | |
| "loss_type": "r-drop", | |
| "focal_alpha": 1.0, | |
| "focal_gamma": 2.0, | |
| "r_drop_alpha": 6.0, | |
| "infonce_temperature": 0.07, | |
| "infonce_weight": 0.5, | |
| "seed": 42, | |
| "wandb_run_name": "graphcodebert-rdrop-best", | |
| "resume_from_checkpoint": "checkpoints/graphcodebert-best/checkpoint-1400", | |
| "save_steps": 100, | |
| "eval_steps": 20, | |
| "logging_steps": 2, | |
| "label_smoothing": 0.5, | |
| "adversarial_epsilon": 0.5, | |
| "use_swa": false, | |
| "swa_start_epoch": 0, | |
| "swa_lr": 1e-05, | |
| "data_augmentation": true, | |
| "aug_rename_prob": 0.8, | |
| "aug_format_prob": 0.8, | |
| "mixup_alpha": 1.0, | |
| "low_pass_keep_ratio": 0.5, | |
| "freq_consistency_weight": 0.5, | |
| "hidden_dropout_prob": 0.3, | |
| "attention_probs_dropout_prob": 0.3, | |
| "classifier_dropout": 0.3, | |
| "device": "cuda" | |
| } |