search

OdysseasKr / neural-disaggregator

Code for NILM experiments using Neural Networks. Uses Keras/Tensorflow and the NILMTK.
MIT License
138 stars 58 forks source link

train_across_buildings and train_across_buildings_chunk #25

Closed HouKun-github closed 4 years ago

HouKun-github commented 4 years ago

Hi, first of all, thank you for your work. I have a question that I do n’t quite understand. I hope to get your answers. What is the use of 'train_across_buildings_chunk'? I do n’t know what this training is different from the previous 'train on chunk' It seems that 'train_across_buildings_chunk' is not used in the program. What is the purpose of this code?

OdysseasKr commented 4 years ago

Hello, Every disaggregator has two methods train and train_across_buildings. The first one trains using one meter while the second receives an array of meters. Each of these methods use the train_on_chunk and the train_across_buildings_chunk. They separate a chunk of the data and train the model.

Basically the only difference is that train_across_buildings_chunk supports chunks from multiple buildings.

HouKun-github commented 4 years ago

Hello, Every disaggregator has two methods train and train_across_buildings. The first one trains using one meter while the second receives an array of meters. Each of these methods use the train_on_chunk and the train_across_buildings_chunk. They separate a chunk of the data and train the model.

Basically the only difference is that train_across_buildings_chunk supports chunks from multiple buildings.

Thank you for your answer, I will try to train with the data of the series table to evaluate the performance of both ways.

HouKun-github commented 4 years ago

Hello, Every disaggregator has two methods train and train_across_buildings. The first one trains using one meter while the second receives an array of meters. Each of these methods use the train_on_chunk and the train_across_buildings_chunk. They separate a chunk of the data and train the model. Basically the only difference is that train_across_buildings_chunk supports chunks from multiple buildings.

Thank you for your answer, I will try to train with the data of the series table to evaluate the performance of both ways. Hello, sorry to bother you, if I want to use a series of tables for training, how can I modify the code on the original code, does the "nilmtk" toolkit provide training data using some table data, I don't seem to find it. Hope to get your help, thank you.

OdysseasKr commented 4 years ago

Hello, Every disaggregator has two methods train and train_across_buildings. The first one trains using one meter while the second receives an array of meters. Each of these methods use the train_on_chunk and the train_across_buildings_chunk. They separate a chunk of the data and train the model. Basically the only difference is that train_across_buildings_chunk supports chunks from multiple buildings.

Thank you for your answer, I will try to train with the data of the series table to evaluate the performance of both ways.

Please note that you are not supposed to call the train_on_chunk and train_across_buildings_chunk directly. Instead use the methods train and train_across_buildings.