Closed dotorl127 closed 3 years ago
I'm not sure about your question, as of my understanding, the training happens inside point_pillars_training_run.py and prediction is done using point_pillars_prediction.py. Is this what you were asking?
No, What I want to say does not load dataset in every frame but seems to be loaded at once when prediction in point_pillars_prediction.py.
No, What I want to say does not load dataset in every frame but seems to be loaded at once when prediction in point_pillars_prediction.py.
As I understand the data is handled with a generator (for training, and also for prediction, definition SimpleDataGenerator() imported from processors.py) that loads the data by batch. So no the dataset is not "loaded at once".
is it importing dataset to train and prediction in once?