Skip to content

hzue/ACM-MM2017-SMP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ACM MM 2017 - Social Media Prediction

This small project just run CNN for more practical experience and compare with different usage of CNN (and different features, but not the point).

Task 1

  • Extract features from VGG19 at fc1 layer (4096), and train it with random forest

    • MAE: 1.67, MSE: 4.41 VGG4096_rf
  • Fine tune VGG19 (using rmsprop first and then using SGD for stable descent)

    • We shrink the fc layer's neuron number because our machine is without GPU
    • MAE: 1.56, MSE: 3.94 VGG19
  • Use external data like wiews, favorate ... may fail in testing stage (they directly gave photo without link).

    • MAE: 0.63, MSE: 0.74 meta_predict

About

The challenge in ACM MM 2017.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages