1

Once the training is done and Model is generated,Model Size can vary according to the dataset and algorithm used. I want to know what is the range (in MBs) the ("generally") Model Size can vary.

Amazon ML sets the limit of Model Size to be between 1 MB to 1GB.

The question is mainly revolved about collecting the information about what is the average size of models generated by organizations? Most Models generated by organization are of how much size ?

Any pointers in the related field will be helpful.

Ionuț G. Stan
  • 176,118
  • 18
  • 189
  • 202
rishabh.bhardwaj
  • 378
  • 4
  • 12
  • 1
    This is not a precise question. How is model-size defined? Memory of your trained classificator/regressor? There are no limits. A SVM for example will grow as the number of support-vectors are growing. One could argue, that a high number of support-vectors is a bad sign regarding generalization, but this is not always the case. And a huge dataset will generate many support-vectors (absolute vs. relative value). Another example are Deep Neural Networks: they will get huge (because they need to store all the weights.) I expect the usual NNs in Machine Vision to be bigger than 1GB. – sascha Jul 15 '16 at 11:53

1 Answers1

-2

Model size depends on the product being used and what is included in the model. This can vary from implementation to implementation, type of problem (classification, regression), algorithm (SVM, neural net etc.), data type (image, text etc.), feature size etc.

neerajkh
  • 1,233
  • 8
  • 9