| { | |
| "SRC_VOCAB_SIZE": 10000, | |
| "TGT_VOCAB_SIZE": 12000, | |
| "NUM_LAYERS": 6, | |
| "NUM_HEADS": 8, | |
| "EMBEDDING_DIM": 512, | |
| "FFN_DIM": 2048, | |
| "MAX_SEQUENCE_LENGTH": 5000, | |
| "BATCH_SIZE": 64, | |
| "EPOCHS": 20, | |
| "LEARNING_RATE": 0.0001, | |
| "WEIGHT_DECAY": 1e-05, | |
| "DROPOUT": 0.1, | |
| "LABEL_SMOOTHING": 0.0, | |
| "BETAS": [ | |
| 0.9, | |
| 0.98 | |
| ], | |
| "EPS": 1e-09, | |
| "WARMUP_STEPS": 4000, | |
| "DEVICE": "cuda", | |
| "SEED": 42, | |
| "SAVE_PATH": "checkpoints/transformer.pt", | |
| "PRINT_EVERY": 100, | |
| "SAVE_EVERY": 1 | |
| } |