Loading Kears Model after crash: Do keras callbacks such as EarlyStopping consider the pre-crash training history?

Question

I am training some models on a remote machine and I lost my ssh connection. I have been saving the models at the end of each epoch with the following call back.

cp = k.callbacks.ModelCheckpoint(
 filepath=checkPointOutPutFolderPath,
 verbose=0,
 monitor="val_loss",
 save_best_only=True, 
 save_weights_only=False, 
 mode="auto", 
 save_freq="epoch", )

I also had an early stopping call back and reduce lr call back

 es = k.callbacks.EarlyStopping(monitor='val_loss', restore_best_weights=True, patience=7)
 rlronp = k.callbacks.ReduceLROnPlateau(monitor='loss', factor=.5, patience=3)

My question is what happens when I reload the model with respect to the early stopping and reduce learning rate call backs? If the validation loss did not improve for 6 iterations before the crash does the model checkpoint save that information? Will the callback fire if the first iteration also does not improve validation loss? Or will I risk overtraining if I restart it? I think It was getting close... If the callback won't fire correctly does anyone know how I could fix it?

model = k.models.load_model(checkPointOutPutFolderPath)
cbs = [es, rlronp, cp]

history = model.fit(trainGen,
                    epochs=max_epoch,
                    batch_size=batchSize,
                    validation_data=valGen,
                    shuffle=True,
                    callbacks=cbs,
                    workers=16,
                    max_queue_size=64,
                    validation_freq=1,
                               )