Changed Metric To Categorical Cross Entropy

src/training/_build_model.py

@@ -3,7 +3,7 @@
 from keras.api.layers import Input, Conv2D, BatchNormalization, LeakyReLU, GlobalAveragePooling2D, Dense, Dropout, TimeDistributed, GRU, Add, LayerNormalization, MultiHeadAttention
 from keras.api.optimizers import Adam
 from keras.api.regularizers import l2
-from keras.api.metrics import Accuracy, BinaryCrossentropy, AUC, Precision, Recall, MeanSquaredError
+from keras.api.metrics import Accuracy, CategoricalCrossentropy, AUC, Precision, Recall, MeanSquaredError
 from src.utils.path_utils import find_project_directory
 
 from src.training._load_dataset import MAX_MOVES, BATCH_SIZE
@@ -67,7 +67,7 @@ def build_model():
         model.compile(
             optimizer=Adam(learning_rate=1e-4),
             loss='category_crossentropy',
-            metrics=[Accuracy(), BinaryCrossentropy(), AUC(), Precision(),
+            metrics=[Accuracy(), CategoricalCrossentropy(), AUC(), Precision(),
                      Recall(), MeanSquaredError()]
         )