2026-04-28 04:22:23,122 - INFO - Loading model and tokenizer from: checkpoints/graphcodebert-base-lowLR-highBatchSize/checkpoint-1022
2026-04-28 04:22:23,386 - INFO - ===== Model Architecture =====
2026-04-28 04:22:23,387 - INFO - 
RobertaForSequenceClassification(
  (roberta): RobertaModel(
    (embeddings): RobertaEmbeddings(
      (word_embeddings): Embedding(50265, 768, padding_idx=1)
      (position_embeddings): Embedding(514, 768, padding_idx=1)
      (token_type_embeddings): Embedding(1, 768)
      (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
      (dropout): Dropout(p=0.3, inplace=False)
    )
    (encoder): RobertaEncoder(
      (layer): ModuleList(
        (0-11): 12 x RobertaLayer(
          (attention): RobertaAttention(
            (self): RobertaSdpaSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.3, inplace=False)
            )
            (output): RobertaSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.3, inplace=False)
            )
          )
          (intermediate): RobertaIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
            (intermediate_act_fn): GELUActivation()
          )
          (output): RobertaOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
            (dropout): Dropout(p=0.3, inplace=False)
          )
        )
      )
    )
  )
  (classifier): RobertaClassificationHead(
    (dense): Linear(in_features=768, out_features=768, bias=True)
    (dropout): Dropout(p=0.3, inplace=False)
    (out_proj): Linear(in_features=768, out_features=2, bias=True)
  )
)
2026-04-28 04:22:23,389 - INFO - ===== Parameter Summary =====
2026-04-28 04:22:23,390 - INFO - Total Parameters:         124,647,170
2026-04-28 04:22:23,391 - INFO - Trainable Parameters:     124,647,170
2026-04-28 04:22:23,392 - INFO - Non-trainable Parameters: 0
2026-04-28 04:22:23,393 - INFO - ===== Tokenizer Summary =====
2026-04-28 04:22:23,408 - INFO - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
2026-04-28 04:22:23,409 - INFO - ===== End of Architecture Log =====
2026-04-28 04:22:23,831 - INFO - Loading dataset from: /kaggle/input/datasets/dzung271828/semeval/Task_A/test.parquet
2026-04-28 04:22:23,832 - INFO - Detected .parquet file – loading directly with datasets (memory-mapped)
2026-04-28 04:22:30,067 - INFO - Loaded Parquet file with 500000 examples (memory-mapped)
2026-04-28 04:22:30,068 - INFO - Columns found: ['ID', 'code', '__index_level_0__']
2026-04-28 04:22:30,072 - INFO - Tokenizing dataset...
2026-04-28 04:27:31,809 - INFO - Running inference on 500000 examples...
2026-04-28 08:29:06,190 - WARNING - No 'label' column found. Skipping metric calculation.
2026-04-28 08:29:11,935 - INFO - ✅ Predictions saved to test/inference/graphcodebert-base-lowLR-highBatchSize/checkpoint-1022/checkpoint-1022-submission.csv