forresti / SqueezeNet

SqueezeNet: AlexNet-level accuracy with 50x fewer parameters
BSD 2-Clause "Simplified" License
2.17k stars 723 forks source link

Has anyone successfully trained Squeezenet with residual connections? #42

Closed stoneyang closed 6 years ago

stoneyang commented 7 years ago

I have trained the two versions of Squeezenet, with success, thanks @forresti !

When training the one with residual connections, I am stucked. Whatever learning policy I took, the one shipped in this repo, or the plainly step, I cannot train it to the results given in the paper. The accuracy is a bit lower than Squeezenet v1.0....

I know that I should post this in that repo, but I can't find issues tab there....

Anyone could shed me some light? Thanks in advance!