dzungpham/graphcodebert-code-classification

12026-04-17 08:00:34,522 - INFO - train_pipeline - Logging to ./output_checkpoints/graphcodebert-robust/training.log22026-04-17 08:00:34,525 - INFO - train_pipeline - ===== Training Configuration =====32026-04-17 08:00:34,526 - INFO - train_pipeline - model_name           : microsoft/graphcodebert-base42026-04-17 08:00:34,528 - INFO - train_pipeline - output_dir           : ./output_checkpoints/graphcodebert-robust52026-04-17 08:00:34,529 - INFO - train_pipeline - num_epochs           : 562026-04-17 08:00:34,531 - INFO - train_pipeline - batch_size           : 3272026-04-17 08:00:34,533 - INFO - train_pipeline - learning_rate        : 2e-0582026-04-17 08:00:34,535 - INFO - train_pipeline - max_length           : 51292026-04-17 08:00:34,536 - INFO - train_pipeline - num_labels           : 2102026-04-17 08:00:34,538 - INFO - train_pipeline - use_wandb            : True112026-04-17 08:00:34,540 - INFO - train_pipeline - freeze_base          : True122026-04-17 08:00:34,541 - INFO - train_pipeline - loss_type            : r-drop132026-04-17 08:00:34,542 - INFO - train_pipeline - focal_alpha          : 1.0142026-04-17 08:00:34,544 - INFO - train_pipeline - focal_gamma          : 2.0152026-04-17 08:00:34,545 - INFO - train_pipeline - r_drop_alpha         : 4.0162026-04-17 08:00:34,546 - INFO - train_pipeline - infonce_temperature  : 0.07172026-04-17 08:00:34,548 - INFO - train_pipeline - infonce_weight       : 0.5182026-04-17 08:00:34,550 - INFO - train_pipeline - seed                 : 42192026-04-17 08:00:34,552 - INFO - train_pipeline - resume_from_checkpoint : None202026-04-17 08:00:34,553 - INFO - train_pipeline - label_smoothing      : 0.1212026-04-17 08:00:34,554 - INFO - train_pipeline - adversarial_epsilon  : 0.5222026-04-17 08:00:34,556 - INFO - train_pipeline - use_swa              : True232026-04-17 08:00:34,557 - INFO - train_pipeline - swa_start_epoch      : 2242026-04-17 08:00:34,558 - INFO - train_pipeline - swa_lr               : 1e-05252026-04-17 08:00:34,559 - INFO - train_pipeline - data_augmentation    : True262026-04-17 08:00:34,561 - INFO - train_pipeline - aug_rename_prob      : 0.3272026-04-17 08:00:34,562 - INFO - train_pipeline - aug_format_prob      : 0.3282026-04-17 08:00:34,564 - INFO - train_pipeline - =================================292026-04-17 08:00:35,711 - INFO - train_pipeline - Model placed on cuda302026-04-17 08:00:35,716 - INFO - train_pipeline - ===== Model Architecture =====312026-04-17 08:00:35,718 - INFO - train_pipeline - 32RobertaForSequenceClassification(33  (roberta): RobertaModel(34    (embeddings): RobertaEmbeddings(35      (word_embeddings): Embedding(50265, 768, padding_idx=1)36      (position_embeddings): Embedding(514, 768, padding_idx=1)37      (token_type_embeddings): Embedding(1, 768)38      (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)39      (dropout): Dropout(p=0.1, inplace=False)40    )41    (encoder): RobertaEncoder(42      (layer): ModuleList(43        (0-11): 12 x RobertaLayer(44          (attention): RobertaAttention(45            (self): RobertaSdpaSelfAttention(46              (query): Linear(in_features=768, out_features=768, bias=True)47              (key): Linear(in_features=768, out_features=768, bias=True)48              (value): Linear(in_features=768, out_features=768, bias=True)49              (dropout): Dropout(p=0.1, inplace=False)50            )51            (output): RobertaSelfOutput(52              (dense): Linear(in_features=768, out_features=768, bias=True)53              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)54              (dropout): Dropout(p=0.1, inplace=False)55            )56          )57          (intermediate): RobertaIntermediate(58            (dense): Linear(in_features=768, out_features=3072, bias=True)59            (intermediate_act_fn): GELUActivation()60          )61          (output): RobertaOutput(62            (dense): Linear(in_features=3072, out_features=768, bias=True)63            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)64            (dropout): Dropout(p=0.1, inplace=False)65          )66        )67      )68    )69  )70  (classifier): RobertaClassificationHead(71    (dense): Linear(in_features=768, out_features=768, bias=True)72    (dropout): Dropout(p=0.1, inplace=False)73    (out_proj): Linear(in_features=768, out_features=2, bias=True)74  )75)762026-04-17 08:00:35,722 - INFO - train_pipeline - ===== Parameter Summary =====772026-04-17 08:00:35,723 - INFO - train_pipeline - Total Parameters:         124,647,170782026-04-17 08:00:35,724 - INFO - train_pipeline - Trainable Parameters:     592,130792026-04-17 08:00:35,725 - INFO - train_pipeline - Non-trainable Parameters: 124,055,040802026-04-17 08:00:35,727 - INFO - train_pipeline - ===== Tokenizer Summary =====812026-04-17 08:00:35,747 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']822026-04-17 08:00:35,749 - INFO - train_pipeline - ===== End of Architecture Log =====832026-04-17 08:00:35,751 - INFO - train_pipeline - Data augmentation enabled (rename=0.3, format=0.3)842026-04-17 08:00:36,645 - INFO - train_pipeline - === Starting training with robust regularisation ===85