I have a project to detect faces using tiny yolov1 with Keras and TensorFlow and I have to train the model from scratch. When I train the model using the dataset https://pixeldrain.com/u/wnUrWG2k. The loss value does not decrease much in each epoch and when I plot the y_pred and y_validation they are both straight lines.
model = Model(inputs=inputs, outputs=yolo_outputs)
model.compile(loss=yolo_loss, optimizer='adam')
history = model.fit(X_train, y_train, validation_split=0.33, epochs=5, batch_size=10)`
this is my model loss plot loss plot
