Update README.md

README.md

@@ -53,12 +53,12 @@ To avoid data leakage and able to predict DDI for new drugs, the drug1 or drug2
 Using AWS EC2 g5.4xlarge instance. 24GB GPU.
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 0.001
-- train_batch_size: 256
-- eval_batch_size: 256
-- seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
+- learning_rate: 0.01
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 7
+- optimizer: Adadelta with weight_decay=1e-04
+- lr_scheduler_type: CosineAnnealingLR
 - num_epochs: 4
 ### Training results
 | Training Loss | Epoch | Validation Loss | F2     | Recall | Precision | Mcc    |