Open Lukzin opened 5 years ago
The paper has been updated. The last but one 'bneck' is updated in 'Table 1'. The 'exp size' is no longer 672 but 960. Please check it out. Is it possible to provide a new pre-trained model based on the updated network structure?
thank you
The paper has been updated. The last but one 'bneck' is updated in 'Table 1'. The 'exp size' is no longer 672 but 960. Please check it out. Is it possible to provide a new pre-trained model based on the updated network structure?