Btukfyl

I am using Keras in multi-gpu, with Tensorflow backend on 2 gpus. I am using a generator (keras.utils.Sequence) to load my data in batch mode (BS = 64). Therefore I am using the fit_generator class, providing it with my train and validation data and steps.
I noticed a strange behaviour starting from the 2nd epoch on. Basically, the first 3 steps of each epoch are completed in just 8/9 seconds each, then the network starts taking longer and longer (as it should do). Logs are the following:

Epoch 00001: val_acc improved from -inf to 0.46875, saving model to data/subs_best_model.h5

Epoch 2/32

 1/29 [>.............................] - ETA: 8s - loss: 1.0664 - acc: 0.5000

 2/29 [=>............................] - ETA: 8s - loss: 1.1384 - acc: 0.4531

 3/29 [==>...........................] - ETA: 9s - loss: 1.0915 - acc: 0.5052

 4/29 [===>..........................] - ETA: 42:03 - loss: 1.1064 - acc: 0.5117

 5/29 [====>.........................] - ETA: 56:02 - loss: 1.1173 - acc: 0.4969

 6/29 [=====>........................] - ETA: 1:03:13 - loss: 1.0964 - acc: 0.4974

 7/29 [======>.......................] - ETA: 1:06:45 - loss: 1.0740 - acc: 0.5067

 8/29 [=======>......................] - ETA: 1:08:35 - loss: 1.0592 - acc: 0.5195

 9/29 [========>.....................] - ETA: 1:08:53 - loss: 1.0580 - acc: 0.5191

Do you know what could cause this anomaly/strange behaviour?

EDIT:

My DataGenerator is inspired by this implementation

The code I use for the fit_generator is as follows:

params = {'batch_size':  TrainConfig.BATCH_SIZE,

              'dim' : ( TrainConfig.BATCH_SIZE, 1, TrainConfig.SAMPLES),

              'labels_dim': ( TrainConfig.BATCH_SIZE,),

              'n_classes' : TrainConfig.OUTPUT_DIM}



training_generator = DataGenerator(train_set, **params)

validation_generator = DataGenerator(val_set, **params)



training_steps_per_epoch = int(1.*len(train_set) / batch_size)

validation_steps_per_epoch = int(1.*len(val_set) / batch_size)



history = model.fit_generator(generator=training_generator,

                              verbose=1,

                              use_multiprocessing=False,

                              workers=1,

                              steps_per_epoch=training_steps_per_epoch,

                              epochs=epochs,

                              validation_data=validation_generator,

                              validation_steps =validation_steps_per_epoch,

                              callbacks=callbacks)