[Machine Learning] Hi, I would like to ask how can I make my validation, test, and training accuracy >=75%? i have 4000 dataset and the train-test-validation set is 60-20-20. I used scikit-learn's MLPClassifier using the parameters attached as a photo.

MLPClassifier(alpha=0.1, batch_size=386, hidden_layer_sizes=[512, 256, 128, 64], Learning_rate_init=0.0001, max_iter=25, random_state=1, solver='lbfgs', verbose=True) train precision recall fl-score support lc lb la 2 0.63 0.61 0.52 0.66 0.48 0.70 0.52 0.57 0.55 0.66 0.52 0.61 836 2345 1531 1074 accuracy macro avg weighted avg 0.61 0.60 0.57 0.60 0.60 0.58 0.60 5786 5786 5786 test precision recall fl-score support lc lb la 2 0.63 0.61 0.51 0.70 0.46 0.70 0.51 0.60 0.53 0.65 0.51 0.64 209 587 383 268 accuracy macro avg weighted avg 0.61 0.60 0.57 0.60 0.60 0.59 0.60 1447 1447 1447 PISA Scalability of the model Performance of the model 09 Training score Cross-validation score 22.5 0.56 20.0 0.8 175 0.54 15.0 0.7 ne times 12.5 Score 0.52 10 0 0.6 050 7.5 50 0.48 05 2.5 1000 2000 5000 6000 1000 2000 5000 6000 50 7.5 15.0 17.5 20.0 3000 4000 Training examples 3000 4000 Training examples 10.0 12.5 fit_times